The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Investigating the implications of static and dynamic loading in high-performance fabrics for outdoor clothing
Static and dynamic loading
Daiva Juodsnukyte˙, Virginija Daukantiene˙ and Matas Gutauskas
Received 15 July 2007 Revised 25 September 2007 Accepted 25 September 2007
Faculty of Design and Technologies, Kaunas University of Technology, Kaunas, Lithuania
7
Abstract Purpose – This paper aims to develop the methodology for the imitation of exploitation conditions of textile products as well as to determine the exploitation peculiarities of high-performance fabrics for outdoor clothing producible in Lithuania. Design/methodology/approach – Static- and dynamic-cyclic loading was applied for the imitation of exploitation conditions as well as for the investigation of the changes in specimen geometrical parameters. Findings – The differences in the parameters of textile material stability determined under dry and wet cyclic specimen deformation were determined. The investigation results presented show that the parameters of air permeability can be used for the determination of changes in textile product shapes due to their cyclic washing as well as to the other kinds of wet technological treatment, especially in these cases when the small areas of product material are deformed. Practical implications – The problems concerned with the methodology for the evaluation of exploitation stability of high-performance fabrics (woven and knitted) for outdoor clothing are analyzed in this research. Originality/value – In most cases, the exploitation behaviour of textile materials is investigated under uniaxial or static biaxial deformation. For better imitation of real exploitation conditions of textiles the new testing methodology based on two testing methods was established (original device for punch deformation working in creep mode as well as using wet and dry specimens; device ARRV for cyclic fatigue). Keywords Textiles, Creep, Deformation, Dynamic loading, Clothing Paper type Research paper
Introduction During exploitation the high-performance fabrics for specialized outdoor clothing are exposed to straight exploitation conditions which are much heavier compare to those of ordinary civil clothing. When considering the specific wear conditions of outdoor clothing the testing methods based on the imitation of real exploitation conditions must be developed. Some published works present the researches concerning the imitation of exploitation conditions of separate (or the most critical) garment zones under biaxial (membrane or punch) deformation (Gronewald and Zoll, 1973; Celanese, 1979; Yeung et al., 2002; Yokura et al., 1986; Gutauskas and Masteikaite˙ 1997; Strazdiene˙ et al., 1997; Kisilak, 1999; Zhang et al., 1999, 2000; Strazdiene˙ and Gutauskas, 2001, 2003; Amirbayat and Namiranian, 2006; Amirbayat, 2006; Juodsnukyte˙ et al., 2006). It should
International Journal of Clothing Science and Technology Vol. 20 No. 1, 2008 pp. 7-14 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810843494
IJCST 20,1
8
be noted that these testing conditions are not the best ones for the experimental imitation of exploitation process of high-performance fabrics for outdoor clothing. In this case, high-performance textiles in outdoor clothing are not only biaxially deformed, but also cyclically abraded by surrounding objects as well as influenced by changing climate conditions. The behaviour of textiles is highly dependent on environment conditions. From the research performing, the uniaxial deformation of textiles it is known that a wet specimen is more easily and highly deformed resulting higher residual strains than a dry specimen. Our earlier investigations concerning the exploitation behaviour of textiles for outdoor clothing under the static punch deformation had shown that the remaining strains of the wet specimens are higher from 1.3 to 1.7 times compare to those of dry specimens (Juodsnukyte˙ et al., 2006). Punch deformation is very good for the imitation of exploitation process of garment. This method imitates biaxial deformation as well as shows the changes in the specimen shapes similar to those occurring in the knee and elbow zones of garment. Keeping the same level of stresses in textile material the linear strains measured in the meridian directions of textile punch-formed shell always have been determined being higher compared to the ones of uniaxial deformation. The advantage of biaxial punch deformation is such as it can to guarantee the same level of loading for the both yarn systems of textile materials as well as it can to imitate the ability of flat textiles to obtain the spatial shapes representing the instability of textiles during their exploitation process. During complex exploitation process of clothing materials, the cyclic loading takes the main position, so its experimental imitation can give new information for the evaluation of stability of textiles during their exploitation process. There are known some fatigue devices capable to imitate the conditions of clothing material exploitation more or less similar to real ones (Bukauskaite˙ and Kucˇingis, 1972; Panasiuk and Gutauskas, 1984; Panasiuk et al., 1984a, b, 1986; Amirbayat and Namiranian, 2006; Amirbayat, 2006; LST EN ISO 9237, 1997). The working principles of these devices are different as they allow changing the frequency of loads (1-2 Hz), the level of maximal deformation (10-12 per cent) or the level of loading in respect to critical force values (10-30 per cent). The work aims to develop the methodology for the imitation of exploitation conditions of textile products as well as to determine the exploitation peculiarities of high-performance fabrics for outdoor clothing producible in Lithuania. Methodology The objects of investigations were five woven fabrics and four knitted materials (Table I). The investigations concerning the exploitation behaviour of high-performance fabrics for outdoor clothing under static- and dynamic-cyclic loading were performed using special laboratory equipment. Static fatigue was carried out using punch deformation method (Figure 1) where dry and wet specimens are loaded under sustained creep process, duration t ¼ 5 h of which was the same in all the cases. Deformation height H of specimen and its changes occurring during specimen relax time were measured. Loaded wet specimen was dried using air flow under the ambient conditions. After creep process, whereas load time was 5 h, the humidity of wet specimen became equal to that of dry specimen.
Static and dynamic loading
Fiber composition, per cent Symbol A-01a A-04 A-05a A-06 A-07 T-01 T-02 T-03 T-04
Cotton
PES
60 70 30 70 70 100 100 96 100
40 30 70 30 30 – – 4 –
Weave/knitting structure Rip stop Rip stop Twill 2/1 Twill 2/1 Warp satin Plain jersey Interlock Rib 1 £ 1 Plain jersey
Density (dm2 1 weft/warp course/wale)
Area density (g, g/m2)
Thickness (d, mm)
176/276 184/280 202/260 190/216 154/154 120/130 130/90 110/100 100/130
250 265 250 250 270 265 129 261 287
0.62 0.69 0.57 0.82 1.01 1.52 0.64 1.04 1.45
9
Table I. The characteristics of investigated materials
a
Note: Material with water repellent finishing
P P R R r r
0 a
H a
H a (a)
b
b
c
b
c a
0 a
Sk.n.
rc b Sr.n.
h
a (b)
Then the specimen was unloaded. After this, remaining strain HR of the specimen as well as parameter of air permeability q in center of the specimen (using the device L14DR according to the standard LST EN ISO 9237) was determined. Parameter of permeability of woven fabrics to air was measured under 2.0 mbar air pressures for 20 cm2 area of specimen. In the case of knitted materials, measuring conditions were 0.5 mbar pressure and 5 cm2 specimen area. Dynamic creep was performed using the device ARRV with 2 Hz revolution of rotor (Panasiuk et al., 1984a, b, 1986; Panasiuk and Gutauskas, 1984; Bukauskaite˙ and Kucˇingis, 1972), when 1 ¼ const ¼ 9.3 per cent and fatigue time t ¼ const ¼ 0.5 h (Figure 2). During fatigue process a strip-shaped specimen was triply impacted, i.e. stretched-bended-abraded, (Figure 3) by steel roller (r ¼ 20.0 mm) fixed stationary on rotor (R ¼ 52.9 mm). The most sensitive fatigue process has been realized in this work (Strazdiene˙ and Gutauskas, 2001).
Figure 1. Schemes of punch deformation, when (a) H , r; (b) H . r
IJCST 20,1
Rotor
10 Specimen
Specimen
Figure 2. Working component of the device ARRV
Rollers
S1
Roller
Specimen
Figure 3. Schematic view of the positions of the specimen-roller
S2
During spinning of rotor (Figure 2) different strains occur in the specimen arms S1 and S2 (Figure 3) which decreases increasing the number of loading cycles as well as the remaining strain. In some cases, the parameter S2 have been decreased up to zero and the specimen worked under a loose beat. The changes in linear deformation of specimen due dynamic fatigue as well as in parameters of air permeability were determined. The number of specimens tested in each fabric group varied from six to ten. Measurement results were processed statistically. The error of measurements does not exceed 5-8 per cent. Results and discussion During punch deformation (Figure 1) woven fabrics have been slightly deformed H , r (Figure 1(a)) as opposite to knitted materials which were comparatively highly deformed H . r (Figure 1(b)) as well as specimen cut from wet woven fabric H $ r. Maximal deformation H of the loaded specimen can be determined measuring and calculating meridian elongation of the specimen l ¼ 2(abc) as well as computing increase of the specimen area, supposing that area of loaded specimen is a sum of surface areas of sphere section and truncated cone: S r:n þ S n:k ¼ S. The dependency
between H and S/S0 is shown in Figure 4, where S 0 ¼ pR 2 . After unloading and relax time of specimen the measurement of specimen remaining deformation HR as well as the determination of changes in specimen meridian elongation and area became complicated. It was noticed that the shape of specimen was changed after its relaxation. The relaxation process influences the evident changes in shape of the thin-walled shell occurring because of the irregular distribution of strains along material warp and weft (wale and course) directions. This reason influenced the use of the other methods for the measurement of specimen remaining strains characterizing mechanical instability of textile materials. The parameters of air permeability can be used for the characterization of changes of textile materials properties due to their technological, particularly wet, treatment. It was established that due static punch deformation the parameters of air permeability changed from 2 to 216 per cent. In all the investigated cases, the air permeability parameters of deformed wet specimens were higher from 36 to 57 per cent compared to those of dry specimens. Increase of the parameter of air permeability could be influenced by the decrease of density of woven fabric in a thin-walled shell formed due to biaxial deformation. The investigations showed that the measurement of linear deformation of soft shell, especially if it formed from knitted material, is very complicate, so the parameters of air permeability could be used to evaluate the changes of specimen geometry (Figure 5). The results of permeability of knitted materials to air before and after specimen fatigue using the device ARRV showed that the increase of air permeability was equal from 1 to 11 per cent, and only wet specimens of knitted fabric T-01 feature the air permeability decreased up to 9 per cent. Analysis of the results has been shown that after uniaxial cyclic fatigue the remaining deformation of specimen vary from 1 to 9 per cent and the parameters of air permeability remain almost unchanged. As comparison analysis of the relative parameters (linear extension l/l0, increase of specimen area S/S0 and change in air permeability q/q0) (Figure 6) showed, the tendency of these parameters changes for five woven fabrics as well as for four knitted
Static and dynamic loading
11
S/S0, l/l0 S/S0 = a + bH2; a = 1.0230; b = 0.0002;
1.4
r 2 = 0.9946;
1.3
1.2 1.1
l/l0 = a + bH2; r 2 = 0.9892; a = 1.0321; b = 0.0001;
1 0
10
20
30
40
H, mm
Figure 4. Theory dependencies between specimen deformation height H and increase of specimen area S/So as well as linear deformation l/lo, when r/R ¼ 0.53
Figure 6. The changes of relative coefficients of air permeability q/q0, linear deformation l/l0 as well as specimen area S/S0 of woven and knitted materials after punch deformation and relax time: (a) for dry specimen (b) wet specimen
Air permeability parameter q/q0
2.5 2 1.5 1 0.5 0 A-01*
A-04
A-05*
A-06
3.5 3 2.5 2 1.5 1 0.5 0
A-07
T-01
(a)
T-02
T-03
T-04
(b)
Key
– control,
– dry,
– wet
3.5
3.5
3
3
2.5
2.5 q/q0, l/l0, S/S0
Figure 5. The results of air permeability q/q0 of textile materials after punch deformation and relax time: (a) for woven fabrics (b) knitted materials
3
Air permeability parameter q/q0
12
materials due punch deformation is similar, especially for woven fabrics with water repellent finishing A-01 and A-05. It was obtained that after specimen fatigue the air permeability of woven fabric A-04 as well as knitted material T-03 slightly increased. After wet punch deformation the parameters l/l0 and S/S0 remained almost unchanged and parameter q/q0 for seven tested materials increased. As the results presented in Figure 6 show that the least changes in air permeability were determined for woven fabrics with water repellent finishing A-01 and A-05. The highest increase of air permeability was determined for woven fabric A-04 and knitted material T-03, and this parameter compared to that of dry punch deformation was 186 and 157 per cent greater, respectively. It can be explained that after wet punch deformation the pores of textile materials were irreversibly opened.
q/q0, l/l0, S/S0
IJCST 20,1
2 1.5
2 1.5
1
1
0.5
0.5
0
Key – q/q0,
0 A-01* A-04 A-05* A-06 A-07 T-01 T-02 T-03 T-04
A-01* A-04 A-05* A-06 A-07 T-01 T-02 T-03 T-04
(a)
(b)
– l/l0,
– S/S0
Conclusions . The results of the investigation showed that punch deformation carried out in creep mode as well as cyclic fatigue performed with the device ARRV suitably imitated exploitation process of high-performance fabrics for outdoor clothing. The use of these methods as well as change of environment conditions allows determining the influence of technological treatment of textile materials on stability of their geometric parameters and shape. . It was established that the local change of product shape could be determined from change in permeability of material to air as well as from other parameters used in material science.
References Amirbayat, J. (2006), “An improved analysis of bagging of textile fabrics. Part I: theoretical”, International Journal of Clothing Science & Technology, Vol. 18 No. 5, pp. 303-7. Amirbayat, J. and Namiranian, B. (2006), “An improved analysis of bagging of textile fabrics. Part II: experimental work”, International Journal of Clothing Science & Technology, Vol. 18 No. 5, pp. 308-13. Bukauskaite˙, D.A. and Kucˇingis, A.A. (1972), “The determination of elastic properties of materials knitted from elastomeric yarns”, The Works of Science Research of Lituania, No. 11, pp. 257-62 (in Russian). Celanese, T.W. (1979), “Bagging test for knitted fabrics”, J. Am. Assoc. Textile Chem. College, Vol. 8, pp. 231-3. Gronewald, K.N. and Zoll, W. (1973), “Practical methods for determining the bagging tendency in textiles”, Int. Textile Bull. Weaving, Vol. 3, pp. 273-5. Gutauskas, M. and Masteikaite˙, V. (1997), “Mechanical stability of fused textile systems”, International Journal of Clothing Science & Technology, Vol. 9 No. 5, pp. 360-6. Juodsnukyte˙, D., Gutauskas, M. and Cˇepononiene˙, E. (2006), “Mechanical stability of fabrics for military clothing”, Materials Science (Medzˇiagotyra), Vol. 12 No. 3, pp. 243-6. Kisilak, D.A. (1999), “New method of evaluating spherical fabric deformation”, Textile Research Journal, Vol. 69 No. 12, pp. 908-13. LST EN ISO 9237 (1997), “Textiles. Determination of permeability of fabrics to air”, LST EN ISO 9237. Panasiuk, L.G. and Gutauskas, M.V. (1984), “The evaluation of strength of adhesion of fused systems under mechanical fatigue”, Technology of Light Industry, No. 4, pp. 18-21 (in Russian). Panasiuk, L.G., Baltrusˇaitis, A. and Gutauskas, M.V. (1984a), “The rotor device for fatigue of textiles, films ang soft leather”, paper presented at Material Science and Technology of Products from Leather and Textiles, Vilnius, pp. 36-8 (in Russian). Panasiuk, L.G., Gutauskas, M.V. and Baltrusˇaitis, A. (1984b), “The device for cyclic pneumo-loading of specimen”, A. S. 1104386 (former SU), OIPOTZIIOT3, No. 27. (in Russian). Panasiuk, L.G., Kondratas, A.V., Titas, R. and Gutauskas, M. (1986), “The application of device ARRV for the evaluation of exploitation properties of garment materials”, Technology of Light Industry, No. 4, pp. 12-15 (in Russian).
Static and dynamic loading
13
IJCST 20,1
14
Strazdiene˙, E. and Gutauskas, M. (2001), “The peculiarities of textile behaviour in biaxial punch deformation”, International Journal of Clothing Science & Technology, Vol. 13 Nos 3/4, pp. 176-85. Strazdiene˙, E. and Gutauskas, M. (2003), “The evaluation of fused knitted systems stability”, International Journal of Clothing Science & Technology, Vol. 15 No. 3, pp. 204-10. Strazdiene˙, E., Gutauskas, M., Papreckiene˙, L. and Williams, J.T. (1997), “The behaviour of textile membranes in punch deformation process”, Materials Science (Medzˇiagotyra), Vol. 5 No. 2, pp. 50-4. Yeung, K.W., Li, Y., Zhang, X. and Yao, M. (2002), “Evaluating and predicting fabric bagging with image processing”, Textile Research Journal, Vol. 72 No. 8, pp. 693-700. Yokura, H., Nagae, S. and Niwa, M. (1986), “Prediction of fabric bagging from mechanical properties”, Textile Research Journal, Vol. 56 No. 12, pp. 748-54. Zhang, X., Li, Y., Yeung, K.W. and Yao, M. (1999), “Fabric bagging. Part I: subjective perception and psychophysical mechanism”, Textile Research Journal, Vol. 69 No. 7, pp. 511-8. Zhang, X., Li, Y., Yeung, K.W. and Yao, M. (2000), “Mathematical simulation of fabric bagging”, Textile Research Journal, Vol. 70 No. 1, pp. 18-28. Further reading Henno, J. and Jouhet, R. (n.d.), “Appareil destine a la´ mesure du pochage des tissues elastiques”, Bull. ITF, S.a., Vol. 22 No. 135, pp. 281-4 (in France). Corresponding author Virginija Daukantiene˙ can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Made-to-measure pattern development based on 3D whole body scans Hein Daanen TNO Defence, Security and Safety, Business Unit Human Factors, Department of Human Performance, Soesterberg, The Netherlands and Faculty of Human Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands, and
3D whole body scans
15 Received 12 June 2006 Accepted 2 February 2007
Sung-Ae Hong Apparel Fashion Business, Hansung University, Seoul, South Korea Abstract Purpose – New techniques are required to link 3D whole body scans to manufacturing techniques to allow for the mass-customization of clothes. This study aims to compare two methods of producing skirts based on 3D whole body scans. Design/methodology/approach – Three females participated in the study. They were scanned with an accurate 3D whole body scanner. A set of relevant 1D measures was automatically derived from the 3D scan. The measures were incorporated in a skirt pattern and the skirt was made from jeans material. The second method was based on triangulation of the scanned waist-to-hip part. The points in the 3D scan were first converted to triangles and these triangles were thereafter merged with neighboring triangles of similar orientation until about 40 triangles remained. These triangles were sewn together to form a “patchwork”-skirt. All females performed fit tests afterwards. Findings – The fit of the 3D-generated patchwork skirt was much better than the fit of the skirt generated by the 1D scan-derived measures. In the latter case, two of the three skirts were too wide because the scan-derived hip circumference exceeded the manually derived values. For the 3D generated skirt, it was necessary to enlarge the triangles with a factor of 1.025 to achieve optimal fit. Originality/value – As far as is known, this is the first study that reports a direct conversion of a 3D scan to clothing without interference of clothing patterns. The study shows that it is possible to generate a fitting patchwork skirt based on 3D scans; the intermediate step of using 1D measures derived from 3D scans is shown to be error-prone. Keywords Clothing, Dimensional measurement, Image scanners Paper type Research paper
Introduction Generally, clothing is selected in retail shops, where the customer is challenged to find garments that match his or her personal preference and at the same time fit to the body. The latter is not an easy job when the large variation in body shape between subjects is taken into account. Therefore, an increasing effort is put into manufacturing This research was financially supported by Hansung University in the year 2007. The authors acknowledge: Koen Tan for processing the whole body scans; Marie¨lle Weghorst for manufacturing the skirts; Aard Daanen for the mathematics behind triangulation; and David Bruner for the information supplied about the TC2 software.
International Journal of Clothing Science and Technology Vol. 20 No. 1, 2008 pp. 15-25 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810843502
IJCST 20,1
16
made-to-measure (MTM) clothing, where the individual is no longer at the end of the clothing chain, but instead forms the start of it. Body dimensions and preferred textiles and colors are matched in individual patterns, which are consecutively cut by single layer cutters and sewn together. This technique was recently used successfully for a complicated stage costume and men’s suit (Choi et al., 2005; Choi and Hong, 2005; Hong and Daanen, 2004). In the last decade, new tools became available to determine the body dimensions fast and reliably: 3D whole body scanners (Daanen and van de Water, 1998). However, in most cases 1D measures like chest and waist circumference are derived from the 3D scans and the 3D information is not used. In this pilot study, we describe a method to make a more direct link between 3D body shape and clothing form for a female skirt. We compare the results to the method using 1D dimensions derived from 3D scans. The purpose of this research was to provide the best fit of a MTM skirt for each different shape of individual consumer from 3D scanned data. The hip part was selected of a 3D scan, since it is the most important part for designing pants and skirts. For the evaluation, we use virtual fitting (mapping the skirt over the 3D scan) and real fitting (the subjects tries the skirt) and combinations between the two. Methods Subjects Three female subjects participated in the study. They were 40, 26 and 27 years old and weighed 61, 72 and 63 kg, respectively. Two skilled anthropometrists performed the anthropometric measurements. 3D scanner and processing software The Vitronic Viro 3D Pro 3D whole body scanner was used to scan the subjects. The system has 16 depth cameras and four color cameras. The images of the 16 depth cameras were aligned and merged using Polyworks (www.innovmetric.com). The resulting scan was stored in the binary pol format. Protocol The subjects came to the anthropometric research lab twice. In the first session, the body dimensions were measured and they were scanned seminude. They were dressed in their underwear with a top and bicycle short over it. These garments were made in such a way that they followed the shape of the body as natural as possible without compressing the skin due to elastic bands. The subjects indicated their preferred skirt length and were informed about the procedures. They were not paid and did not have to pay for the skirts. After the first session two skirts were made for each subject. The first skirt was made based on the 3D scan (3D skirt), the second skirt was based on 1D scan derived body dimensions (1D skirt). In the second session, the subjects were scanned three times to evaluate the fit: seminude, seminude with 3D skirt and seminude with 1D skirt. Alignment and visualization of the scans was performed using Integrate ( Burnsides et al., 1996).
3D skirt The 3D skirt was based on the human body shape at hip level. The parts of the 3D scan above the waist and below the hip were removed. The points in this “band” over the body were triangulated. Using the imcompress module of Polyworks (www.innovmetric.com), the number of triangles was reduced stepwise. It appeared that the reduction factor of 5 percent resulted in the minimum amount of triangles in which the shape of the hip was still clearly visible (Figure 1). This reduction corresponded to about 40 triangles. The method essentially combines neighboring triangles when their surface normal vectors are pointing in the same direction (as set by a certain tolerance). This results in a few large triangles for relatively flat surfaces and many small triangles for curvy surfaces. The triangles were projected on a flat plane perpendicular to the surface normal. One corner was moved to the (0,0) coordinate and the triangle was rotated around (0,0) so that the second coordinate was on the x-axis (x1,0). The third coordinate was (x2, y2). All coordinates were multiplied by a factor 1.025 in order to be sure that the skirt would fit over the body. The triangles were printed on paper, transferred to the textile material and then sewn together without overlap. The band was cut at the back and a zipper was inserted to enable donning of the skirt. The zipper was inserted in such a way that the textile on the left and the right side of the zipper would have touched if the zipper was absent.
3D whole body scans
17
1D skirt The 3D scan was processed using the TC2 body measurement system (www.tc2.com). The 1D dimensions hip circumference and waist circumference were derived from the 3D scan and these were used as input for the pattern design. The length of the skirt was determined by the preference of the subjects.
Notes: The left figures have a reduction factor of 1% and the right figures of 5%. The figures correspond to the front, back and side of the subject from top to bottom
Figure 1. Gouraud shaded triangles resulting from the reduction process of the 3D scanned data
IJCST 20,1
18
Textile and patterns The skirt was made of jeans textile. The physical properties of the textile are provided in Table I. The patterns for the skirts were custom made by using Investronica CAD system and converted to DXF format. The skirt consisted of a front and back part with four darts for each part at the upper side of the skirt. NARCIS (D&M Technology Co., Ltd) was used for 3D virtual sewing and try-on to test the fit of variable skirts by investigating strain, ease amount, and relative pressure.
Results Figure 2 shows a frontal view of 3D scans seminude, with the 1D skirt and 3D skirt. White paint was attached over the skirts to make them visible for the lasers of the 3D scanner. 3D skirt The 3D skirts generated by the triangulation process all had a similar shape for the subjects. A total of 37 triangles were generated for Subject 1, 41 for Subject 2 and 38 for Subject 3. The protruding point of the buttocks was the connection point of several triangles, thus enabling the accommodation of the buttocks (Figure 3). Table I. Physical properties of textiles
Fiber content
Weave
Thickness (mm)
Weight (g/m2)
Fabric count (wp £ wf/in.) (continued)
Fiber content
Weave
Thickness (mm)
Weight (g/m2)
Fabric count (wp £ wf/in.)
Cotton 100 percent
2 £ 1 twill
1.01
402.0
66.4 £ 49.2
Figure 2. Frontal view of the scans of subject three seminude (left) and with the 1D skirt (middle) and 3D skirt (right)
3D whole body scans
19 –Z
Notes: Please note the buttock point where the triangles come together. The skirt is scaled with a factor 1.025
Figure 3. Back view of the 3D skirt over the 3D scan of a typical subject (Subject 3)
The subjects indicated that the skirts fitted tight to their bodies. The enlargement factor of 1.025 was good with the zipper included (Figure 4). Figure 5 shows the enlargement. The 3D scans were superimposed to investigate the fit of the skirts. Figure 6 shows the results. The 1D skirt has a much looser fit than the 3D skirt “band.” 1D skirt Skilled anthropometrists measured the subjects and their relevant body dimensions were derived using the TC2 body measurement system. Table II shows the relevant measures. The waist circumference was considerably larger (75 mm) using the TC2 system as compared to manual measures, in particular for Subjects 2 and 3. This is probably related to the observation that the location of the waist circumference measurement
Figure 4. Back view of the 3D skirt worn by Subject 3
IJCST 20,1
20 Figure 5. Top view of the computer generated skirt of Subject 3 with (brown) and without (green) enlargement
(a)
(b)
Figure 6. Transsections of the aligned scans of subject three seminude (brown), 1D skirt (blue) and 3D skirt (green); (a) at the upper leg level; (b) just below the crotch; (c) at the hip level
(c)
was considerably lower (75 mm) for the TC2 system. Hip circumference as derived from the scan was used in the skirt design. Subjects 2 and 3 considered the 1D skirts too loose. These two subjects had an underestimation of the hip circumference using the TC2 software. Comparison of 3D-1D skirt The surface area of the 1D skirt and 3D skirt were both about 0.2 m2 (Table III). Figure 7 shows the superimposed image of the 3D and 1D skirt for a typical subject.
3D whole body scans
21
MTM skirt patterns and 3D virtual fit Depending upon the design preference (skirt length, etc.) and variable shapes of each subject, MTM skirt patterns were developed and shown in Figure 8. These skirts were virtually sewn together and tried on to test the fit. The distribution of strain, ease amount, and relative pressure were shown in Figure 9 for the objective evaluation of the fit. Discussion 3D skirt The basic idea of the 3D skirt was to convert 3D information of the body shape directly to a pattern of a skirt over the hips. The shape of the computer-generated skirt perfectly fits the body, but a scale factor has to be included in order for the skirt not to be too tight. The scale factor of 1.025 appeared to be correct in this case. The optimal scale factor is expected to be dependent on the material that is used. It is well recognized that the described technique in this study is limited to clothing that follows the body shape. However, since clothing increasingly becomes the Subject 1 Manual TC2 Stature Waist circ. Hip circ. Waist height Hip height Knee height Waist-hip
1,647 745 998 1,023 794 473 247
1,659 778 1,009 982 771 431 218
Subject 2 Manual TC2 1,748 751 1,067 1,104 863 522 252
1,745 848 1,082 1,030 833 483 204
Subject 3 Manual TC2 1,666 756 1,040 1,033 797 504 242
1,649 850 1,062 968 789 454 188
Mean Manual TC2 1,687 751 1,035 1,053 818 500 247
1,684 825 1,051 993 798 456 203
Difference Mean SD 3 275 216 60 20 44 44
15 36 6 17 11 6 13
Note: Unit – mm
3D scanned area of the hip Triangulated hip area 1.025 £ triangulated hip area Surface area of 1D skirt in hip area Note: Unit – mm2
Subject 1
Subject 2
Subject 3
Average
188,470 181,094 190,262 215,546
205,576 201,061 211,239 217,769
192,268 186,260 195,537 192,190
195,438 189,472 199,013 208,502
Table II. Manual measures and 1D measures derived from the whole body scans
Table III. Surface area of the hip and skirts
IJCST 20,1
22
Figure 7. Upper part: the triangulated skirt unfolded. Middle part: superimposed images of the triangles (red dotted) and 1D skirt (blue lines) seen from the back (left) and front (right). Lower part: separated hip block patterns of skirts
Subject 1
Subject 3 C.F.
C.B.
C.F.
C.B.
C.F.
C.B.
Figure 8. CAD patterns of MTM skirts for three different subjects seen from the back (left) and front (right) in each set of patterns
Subject 2
platform for sensors to the human body, for instance to monitor the vital capacity of soldiers, it can be foreseen that tight fitting clothing becomes increasingly important for functional clothing. For other clothing, like shirts, Sul and Kang (2006) recently described a nice draping method to interface the clothing pattern with a 3D scan to optimize fit. Putting the triangles together manually is a tedious job. However, since the dimensions of the triangles are known, computer sewing may be an option. Moreover, recent techniques became available to reorganize the triangles in clothing patterns (Kim and Kang, 2003).
11.38
11.38
–1.69
–1.69
3D whole body scans
23
50.00
50.00
1.00
1.00
0.44
0.44
0.00
0.00
Figure 9. Virtual try-on images of front and back indicating strain (top), ease amount in mm (middle) and relative pressure of skirts (bottom)
1D skirt Increasingly, 3D body scans are used to derive 1D dimensions from 3D scans (Chan et al., 2004). In a comparison of two main software packages to perform this process, Hin and Krul (2005) noted that considerable differences existed between the data of skilled measurers and the software results. Also, in the study described here, large deviations occurred between manual and scan derived measures (Table II). Daanen (1998) observed that circumferences could be derived accurately when humans directly
IJCST 20,1
24
interact with a 3D scan. ISO 20685 (2005) is a standard in preparation that defines the acceptation limits for scan derived body dimensions as compared to manual measures by skilled anthropometrists. The difference between scan derived and manual measures is dependent on the computer routine that is used. Also, the absence of skin compression and the inability to palpate the body prior to measuring has an impact. In the current experiment, the resulting errors caused the patterns to be far from optimal. Recently, Cho et al. (2005) proposed to convert a 3D whole body scan to an interactive body model suitable for pattern making. Although, this method is not automated, it may be effective in reducing errors in deriving 1D dimensions from 3D scans. Conclusions In conclusion, the 3D scan generated skirt showed a tight fit. The enlargement factor should be more than 1.025. In line with previous observations, automatic 1D derived measures from 3D scans may deviate considerably from measures of skilled anthropometrists and thus cause bad fit. References Burnsides, D.B., Files, P.M. and Whitestone, J.J. (1996), “Integrate 1.25: a prototype for evaluating three-dimensional visualization, analysis, and manipulation functionality”, AL/CF-TR-1996-0095, Air Force Material Command, Wright-Patterson Air Force Base, Dayton, OH. Chan, A.P., Fan, J. and Yu, W.M. (2004), “Prediction of men’s shirt pattern based on 3D body measurements”, International Journal of Clothing Science & Technology, Vol. 17 No. 2, pp. 100-8. Cho, Y., Okada, N., Park, H., Takatera, M., Inui, S. and Shimizu, Y. (2005), “An interactive body model for individual pattern making”, International Journal of Clothing Science & Technology, Vol. 17 No. 2, pp. 91-9. Choi, J.H. and Hong, S.A. (2005), “Pattern development of stage costume for dynamic movement”, paper presented at Seoul International Clothing and Textiles Conference, Seoul. Choi, J.H., Hong, S.A. and Kim, S.A. (2005), “3D sewing and virtual fit of stage costume”, paper presented at Seoul International Clothing and Textiles Conference, Seoul. Daanen, H.A.M. (1998), “Circumference estimation using whole body scanners and shadow scanners”, Proceedings of the Workshop on 3D Anthropometry and Industrial Products Design, Paris, June 25-26, pp. 5-1 – 5-6. Daanen, H.A.M. and van de Water, G.J. (1998), “Whole body scanners”, Displays, Vol. 19, pp. 111-20. Hin, A.J.S. and Krul, A.J. (2005), “Performance of human solutions body dimensions software”, Report 2005-A9. TNO Human Factors, Soesterberg. Hong, S.A. and Daanen, H.A.M. (2004), “3D scan related research in TNO and its application for apparel industry”, Fashion Information and Technology, Vol. 1, pp. 72-80. ISO 20685 (2005), 3-D Scanning Methodologies for Internationally Compatible Anthropometric Databases, ISO, Geneva. Kim, S.M. and Kang, T.J. (2003), “Garment pattern generation from body scan data”, Computer-Aided Design, Vol. 35 No. 7, pp. 611-8. Sul, I.H. and Kang, T.J. (2006), “Interactive garment pattern design using virtual scissoring method”, International Journal of Clothing Science & Technology, Vol. 18 No. 1, pp. 31-42.
Further reading Robinette, K.M., Blackwell, S., Daanen, H.A.M., Fleming, S., Boehmer, M., Brill, T., Hoeferlin, D. and Burnsides, D. (2002), “Civilian American and European surface anthropoometry resource (CAESAR)”, Final Report, Volume I: Summary, AFRL-HE-WP-TR-2002-0169, United States Air Force Research Laboratory, Human Effectiveness Directorate, Crew System Interface Division, 2255 H Street, Wright-Patterson AFB OH 45433-7022 and SAE International, 400 Commonwealth Dr, Warrendale, PA 15096. Corresponding author Hein Daanen can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
3D whole body scans
25
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,1
Expert-based customized pattern-making automation: Part I. Basic patterns
26 Received 10 September 2007 Revised 1 October 2007 Accepted 1 October 2007
Jing-Jing Fang and Yu Ding Department of Mechanical Engineering, National Cheng Kung University, Tainan, Taiwan Abstract Purpose – This paper aims to present a flattening method for developing 2D basic patterns from 3D designed garments. The method incorporates the techniques of professional pattern development for the purpose of pattern-making automation. The aims of the flattening method are to improve the dressing suitability and to produce pleasing figures by reversing design procedures. Design/methodology/approach – A flattening method is presented in this paper for developing 3D undevelopable NURBS surfaces in 2D. The automatic operation embeds the expertise of pattern makers by reducing total area differences between the designed garments in 3D styles and the two-dimensional patterns. Basic pattern-making invokes the boundary constraints which apply mesh alignments techniques. Findings – The global area difference between the original 3D designs and the 2D-developed pattern is controlled within 5 percent in order to reach the final outcomes of basic patterns, whose shapes are similar to the drawing patterns currently utilized in the industry. Research limitations/implications – This study currently handles simple designs, such as basal designs, and can only flatten garments in symmetric styles. The direct flattening method is developed by this study. In addition, this study is supplemented by expert-based knowledge, and it establishes basic boundary conditions for various garment patterns to increase the feasibility of flattening automation. Originality/value – This study introduces the fundamental theories and methodologies used in the automatic making of basic patterns from 3D garment designs. It proposes a flattening method with pattern expertise embedded by real-time approximations of the global area of the 3D undevelopable designs to the 2D patterns. Keywords Clothing, Dimensional measurement Paper type Research paper
International Journal of Clothing Science and Technology Vol. 20 No. 1, 2008 pp. 26-40 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810843511
1. Foreword In recent years, as personal incomes have gradually increased and living standards improve, modern people no longer base their demand for clothes only on shelter and warmth. They want to have more diversified and unique wardrobes that can show their personal styles. Only such clothing can meet the needs of modern people who wish to distinguish themselves from others and display themselves in exquisitely delicate shapes or to disguise the defects of their bodies. Along with the universal prevalence of computers and the strengthening of their capabilities, many studies and software programs have been applied in fashion designs to accelerate the design process, reduce the development costs, and reflect concrete design conceptions, with the aim to meet the various demands of modern society.
Usually, common commercial, ready-made clothing available in the markets cannot meet the widely diverse demands for modern fashion. The general public, who has not been professionally trained in garment design, cannot easily design the clothes they desire. Furthermore, garment design is often traditionally based on the designer’s visualizations, in cooperation with experienced pattern-makers and tailors, and the combined knowledge and experience in various fields to realize the designer’s perceptions. However, based on 3D scissoring concepts, this study on patterns seeks to flatten garment elements with irregular surfaces against a flat plane, using un-developable surface flattening techniques to generate slopers for sewing trials. This study intends to integrate the professional knowledge of different fields to achieve the goal of customizing garment pattern-design and making. 1.1 Research objectives Currently, the environmental incentives brought about by increases in demands and industrial competition, have led to the successive establishments of information that integrates science and information engineering into industries. In order to enhance enterprises’ competitiveness, traditional manufacturing industries introduced various CAD and CAM software programs to assist in the research and development (R&D) of newly-made products and to apply a variety of engineering analysis and simulation tools in finished product tests. This has significantly contributed to the reduction of the costs and R&D cycles needed for product design and manufacturing and effectively managing product qualities. Take the garment manufacturing industry as an example. Traditionally in the garment industry, the fashion designer needs to make a planar sketch of the garments, and then the pattern-maker works out a garment model based on the sketch to generate a planar sloper for cutting and sewing. However, in the present day, while both garment design and pattern-making increasingly rely on computers, modern commercial pattern-making software programs can notably reduce the development time needed to design and manufacture products, thus gradually attracting more attention from the market. However, since these pattern-making software programs are based only on 2D design, ordinary persons cannot achieve desirable designs without extensive pattern-making training and experience. In lieu of this situation, the current study aims to reduce the time and human resource costs involved in the process of garment development and to pursue the objectives of fittingness of clothing and the display of graceful figures. While incorporating 3D scissoring concepts, collecting and integrating traditional 2D pattern-making experience and data, and adopting expert-based pattern making knowledge and experience, the study seeks to make full use of the merits of various methodologies to develop 3D pattern-making software. As such, even those users who do not have as many years of experience as professional pattern makers can convert a 3D style design into a 2D pattern and do the scissoring and tailoring themselves. This will lead to a significant reduction of production and research costs, the minimization of design trial and error, and enhanced diversity of product development and design. 1.2 Related study Presently, many research institutions around the world have applied 3D garment surface flattening and dart development to the garment design industry. Here, we select
Customized pattern-making automation 27
IJCST 20,1
28
some of them that frequently publish papers in international journals or conferences as examples, and give brief introductions of them. In a document published by the research team led by Professor Yuen in the Hong Kong University of Science and Technology in 2001, point clouds of body scan data were obtained through body scanning, and feature seam lines that are established through the feature search method (Zhang and Yuen, 2001). Then, garment patterns were applied mechanically to make clothes, but the slopers for practical manufacture still could not be generated via this method. Though they had applied multilevel meshes to develop the method of speeding up clothing simulation, in which the process of simulation is divided into several steps, they merely made use of traditional garment patterns to display dressing effects. In another document published in 2002, Wang used a minimal energy-based model to measure the optimized 3D mesh flattening results used in minimizing the variation in total area of meshes before and after their development and the total variation in the length of meshes. Wang et al. (2002) also applied a surface flattening algorithm in areas of fashion design and the manufacturing industry. A research team from Queen’s University of Belfast, in a research document published in 1999, presented several methods of developing curved surfaces approximating triangulated meshes into 2D patterns (McCartney et al., 1999). Considering the deformation caused by darts and linings used in clothes, they suggested an energy-based model to maintain the strain energy of edges when triangulated meshes are flattened. Moreover, they (McCartney et al., 2000a) noted that the research topics at that time focused mainly on developing interactive systems, while there was little research on how garment patterns displayed their shape through software programs. Since, garment patterns were often deformed by factors such as darts, seams, and edges, they treated the left forepart of the torso as the front, left bodice of a virtual mannequin, created triangulated meshes, and assumed a preliminary blouse pattern. Then they flattened the left front slice and added darts to simulate its draping effects, enabling the system to generate a sloper for real production. In a thesis paper published in the same year (McCartney et al., 2000b), they developed a method of flattening special 3D curved surfaces into optimal 2D patterns, by employing an optimizing algorithm to calculate the minimum energy needed to develop 3D curved surfaces into planar patterns. When carrying out the calculations, the degree to which special surfaces diverge from un-developable surfaces, the tensile strains, and the shear strains of woven fabrics in all directions were given special consideration, and the energy distribution of the woven fabrics was predicted. A research team (Kang and Kim, 2000) from Seoul National University of South Korea regarded the mesh lines marked on the virtual mannequin as the garment meshes and adjusted the curved surface of the bodice with geometric constraints. They then divided the surface of a blouse, which included front and back pieces, into tens of meshes, and flattened the surface into a pattern with consideration of anisotropic textures. This formed an appearance similar to darts in traditional patterns between mesh areas. Based on the actual size of openings after developing darts, the need to develop darts was determined and the number of darts controlled. As a follow-up, the authors further published their research results in 2003, in which they developed a complete garment pattern generation system after having induced and integrated the experience of pattern makers (Kang and Kim, 2003). This system uses each mesh
as a calculation base and can predict the shear force exerted on each mesh to determine whether or not a dart should be developed in this position or whether to deform the mesh to fit in with the formal integrity of the pattern. In the document published in 2006, the authors emulated 3D scissoring concepts and applied the mesh-cutting algorithm they themselves developed to present a virtual 3D scissoring system (Sul and Kang, 2006). In a virtual field, the system puts the cloth around the virtual mannequin to generate folds when the cloth hangs from the mannequin, then slices off the superfluous cloth in a projective cutting method, trimming the cloth with the 3D scissoring method of the system to achieve virtual 3D scissoring. However, through this method, the middle waist girth of the developed pattern following 3D scissoring was smaller than the actual waist girth and could not test whether or not the pattern was suitable to be worn after clothes were sewn based on this pattern. In this research, we developed an automation strategy for pattern development by ways of rigid meshes flattening under the given constraint of non-overlapping among them. Base on the requirements in basic pattern making, we move and rotate those meshes in order to minimize the increment of the envelop area of the flattening pattern. Compared to other researches mentioned above, their ways of evaluating patterns outcomes tended to energy-based strategy rather than minimizing pattern area increment. Besides, it can be seen from the above-mentioned documents that these research teams have a similar objective – to realize garment design automation. What distinguishes our research from that of other teams the most is that our team will establish a digital human body using real human body scanning to generate garment patterns based on personalized design information for the individual’s fitting. Moreover, this study is supplemented by expert-based knowledge and experience of garment pattern-makers, and establishes basic boundary conditions for various garment patterns to increase the feasibility of manufacturing automation. The research flow chart is shown in Figure 1. The digital human body (Tsai and Fang, 2003) and the NURBS-based 3D garment restyling (Fang and Liao, 2005) are the predecessors to this study. The research results can be integrated with pattern-making software available in the market to establish pipelines for real and virtual trials, respectively, and to form the basis for an interactive fashion show. 2. System structure This study aims to develop software for automatic pattern generation and darts for garment design, which flatten the garment designed by 3D CAD (Fang and Liao, 2005) software into a plane and further divides meshes that approximate curved surfaces into several areas. These areas are then aligned locally to reduce the space gap between these meshes and decrease the difference between pattern output and the design, thus forming a bridge between garment design and actual sloper output for further sewing. The following chart (Figure 2) shows the automatic pattern generation flow in which Design3D (Fang and Liao, 2005) – a 3D garment design software – generates the garment surface from a structured digital mannequin of an actual body. This corresponds with the body features of a digital mannequin of a human for the integration of suggestions and experience from garment pattern-making experts. While Part I of this study illustrates the basic patterns that are automatically generated, Part II is concerned with reducing the distance gap between local meshes to
Customized pattern-making automation 29
IJCST 20,1
30
Measurement
Human Body Scanning
Photographing
Reconstruction of Digital Human Body
3D Garment Style Design Automatic Pattern Flattening In this research Dart Setup
Cutting, Sewing and Real Fitting
2D Garment Pattern Design Software 3D Virtual Fitting and Dressing Effect Display
Figure 1. 3D garment design project flow
Interactive Fashion Show
generate darts. The fundamental theories and methodologies of developing darts will be discussed in Part II of this study. The pattern generation results can provide a textured mapping setup and can be used to arrange printing patterns (in Part II). Alternately, these results can be supplemented with subsequent treatments, such as seams and mesh corner connection, to produce ready-made garment slopers for cutting and sewing, or can be input into 2D pattern design software for refinement purposes. 3. Expert-based knowledge for basic patterns This study defines basic patterns as those that cover body parts below the waist and hip because the lack of the hip girth often leads to a small middle waist girth or insufficient hip girth, which leads to the unsuitability of dresses. The basic patterns mentioned in this study include the bodice front and bodice back. To present a beautiful appearance and appropriateness for sewing, five conditions are set for pattern-making experience:
Customized pattern-making automation
Deign3D Garment Design Software
Bodice Surface Meshes
31
Mesh Redrawing
Mesh Flattening Structure Human body
Mesh Alignment under Restriction
Expert Knowledge and Experience
Initial Pattern Generation Dart Developing or User’s Input of Given Restriction for Local Alignment
No
Mesh Gap Analysis and Texture Mapping Distortion Evaluation
Satisfy the Gaps Threshold
Yes
Texture Coordinate Output
Finishing Treatment of Pattern Patten Output
(1) central lines shall remain straight, or central lines and waist lines shall remain straight and intersect vertically; (2) collar seam lines must intersect shoulder lines at approximately right angles; (3) shoulder lines must intersect with armhole lines at approximately right angles; (4) armhole lines must intersect with side lines at approximately right angles; and (5) collar seam lines must intersect with back central lines at approximately right angles. See attached Figure 3 for details. This study adopts the first condition as the basic requirement for pattern flattening. The remaining conditions will be automatically generated by the software in the course of alignment.
Figure 2. Flowchart of automatic pattern generation
IJCST 20,1
Side Neck Point Back Shoulder
Side Neck Point Front Shoulder
Shoulder Point
Front Neck Point
32 Front Arm Hole Bust Line
Back Neck Point
Shoulder Point
Back Arm Hole
Bust Point Side Dart Back Center Line
Front Center Line Waist Line
`
Side Line
Waist-Fitting Dart Bodice Front
Figure 3. Schematic drawing for patterns of bodice front and bodice back
Bodice Back
Hip Line Lower Hem Lower Hem
In the traditional garment-making process, to obtain the dimensions of the planar sloper components for drawing purposes, measurement tools such as tape or cloth measurers are necessary to obtain data on girths or lengths related to features of the human body so that 2D patterns can be made these measurements. However, in the computer-aided 3D garment design, 3D scanners are utilized directly to scan human bodies, and the body structures can be obtained from a feature collection (Tsai and Fang, 2003) such that the lengths of all features are obtained reliably. Nevertheless, the relative positions of the body features of a virtual mannequin, such as neck lines, waist lines, bust lines, and hip lines, are still necessary for pattern-making to meet the needs of garment design. In addition to human body features, clothes have their own feature points, feature curves, and feature lines (Tsai and Fang, 2003). Do these features correspond relatively with bodily features when these clothes are worn? Can relative curve lengths correspond? Since, these feature points or feature curves will have an influence on the fit of the clothes, these features also need to be given attention and must be given equal priority in the process of 2D pattern making.
Prior to making planar patterns, data on the 3D surfaces of bodice and information on the relative feature positions of bodies must be put in place to generate meshes that approximate these surfaces, then flatten, develop, and align the meshes to make basic patterns. Taking pattern sewing and fitting bodice needs into account, different constraints shall be given to make garment patterns based on the expert experience of pattern-makers. In the following paragraphs, we take the example of a half-bodice front and bodice back to illustrate how to set expert-based boundary conditions and the positions of feature points and feature lines in relation to 3D bodices when patterns are generated: . Basic pattern of a half-bodice front. The given condition for flattening a half front piece of bodice is that the central line and the waist line are straight lines and must intersect at right angles. Based on this requirement, the remaining meshes are aligned in order toward the central and waist lines. The surface points closest to the bodice are calculated from digital human body feature points to identify the bust point, the upper and lower vanishing points of the waist dart, and front waist line to be used for generating darts. Information on the relative positions of these points and lines is shown on Figure 4. . Basic pattern of a half-bodice back. Similarly, the condition for flattening the half back bodice is that the central and waist lines are straight and must intersect at a right angle. Based on this given condition, the remaining meshes are aligned in order toward the central line and waist line. The surface points closest to the bodice are calculated from digital human body feature points to generate the upper vanishing point of the back waist dart and back waist line to be used for developing darts. Information on the relative positions of these points and lines is shown on Figure 5. 4. Rules of flattening This study explores how to flatten the meshes approximating 3D curved surfaces of garments into a plane and then how to arrange and align independent meshes to construct the pattern. Therefore, while flattening or alignment is being carried out, the objectives of pattern-making need to be taken into account, that is, to reduce as many gaps between meshes as possible and to ensure that the total area of all meshes approximating curved surfaces is not changed as well. Correspondingly, this study proposes a 2D collision detection method for non-deformable meshes to prevent overlapping in the mesh flattening process and to maintain the closest state. In other words, the method aims to avoid mesh overlapping that may lead to a smaller surface area of the pattern than that of the garment surface, resulting in practical dressing difficulties when the flattened pattern is produced for actual sewing. Hence, the gaps between meshes generated based on this principle makes the outfit based on 2D pattern design looser than that of the 3D garment design. The flattening method developed in this study takes meshes as fundamental elements for flattening; the flattening target is a parameterized mesh surface S. Suppose the vertex points on the mesh surface are Puv, then the four adjacent vertex points Puv, Puþ 1v, Pu,vþ 1, and Puþ 1,v þ 1 on the curved surface after flattening form a 2D quadrilateral mesh on the U-V plane surface, as shown in Figure 6. The flattening steps are divided into three parts, namely, mesh coplanar, alignment, and rearrangement.
Customized pattern-making automation 33
IJCST 20,1
34
Neck Line
Armhole Front Neck Point Bust Point
Upper Vanish Point
Sideline
Waist Line Lower Vanish Point Front Central Line
Figure 4. Flattening of a half bodice front
Lower Hem
Back Neck Point Back Neck Line
Armhole
Upper Vanish Point
Sideline
Figure 5. Flattening of a half bodice back
Lower Vanish Point Lower Hem
Back Central Line
The 2D mesh construction process ensures that every triangulated mesh equals the corresponding mesh on the approximate mesh surface. In this study, we directly calculated the geometric features of the original meshes and completely restructured equal meshes from the U-V surface, where L is located without complicated calculations of a transformation matrix. The geometric features obtained from the above calculations can be used repeatedly, while position adjustment, gaps, and analysis of meshes are carried out later. The 2D meshes are generated after every mesh is restructured on the U-V plane surface. Because the region between meshes had been recorded on the original surface, it is possible to make a preliminary arrangement of the flattened meshes, based on the original region. Furthermore, the meshes that have been arranged do not overlap one another. Mesh alignment helps to further reduce the gaps between meshes. The meshes are moved and aligned along the given direction on the premise of overlapping prevention. The action of alignment, including mesh transportation and rotation, can be divided into column- and row-based alignment, where the former refers to meshes’ orderly alignment towards the given column serving as a restricting condition, while the latter refers to meshes’ orderly alignment towards the given row serving as a restricting condition.
Customized pattern-making automation 35
5. Results This section attempts to assess the results of the patterns generated in the previous two sections. It also proposes the area and sewing line lengths as the basis for assessment and demonstrates the process of transforming computer patterns into slopers for sewing and trials. First, the meshes are flattened via a column-based alignment to generate the results as shown in Figure 7(a); central lines remain vertical to produce the results as shown in Figure 7(b). Second, the longest row among all lines P1P3 on the 3D meshes is determined; this row is then used to rotate all meshes as shown in Figure 7(c). Then, the cc alignment of all meshes is carried out based on this row and the central line column, respectively, under the same boundary condition, as shown in Figure 7(d). Finally, the total area of meshes is calculated with external boundaries included and compared to the surface area of meshes that approximates the original 3D curved surface. If the difference exceeds 5 percent of the total area of approximate meshes, then further procedures are necessary to obtain a percentage difference less than five. Methods of advanced operations include those of developing darts or adopting local alignment, which aim to reduce the gaps between meshes. For this purpose, local alignment is executed towards the mesh subclass above the longest row,
pu(v+1) Buv
Duv Luv
Auv
Cuv
L
puv
Muv
p(u+1)(v+1)
p(u+1)v
S
Figure 6. Schematic drawing of mesh flattening
IJCST 20,1
Longest row
36
(a)
Figure 7. Development processes of bodice front
(b)
(e)
(c)
(d)
(f)
and then, each mesh within the right region of ubp is selected for rotation based on the column where the bust point is located, as shown in Figure 7(e). Furthermore, the column-based alignment method lines up the meshes within the region, based on the central line. The method to line up the longest row as a secondary constraint is applied to automatically develop the bodice front pattern, as shown in Figure 7(f). The result shown in Figure 8(a) is generated only after the column-based alignment based on the central line. However, if the meshes are aligned based on the waist line as the constrained column and the central line as the constrained row simultaneously, and the two lines remain straight and intersect at right angle, then the result may be as shown in Figure 8(b). It can be determined from the area difference shown in Table I that the result of column-based alignment (based on the central line) is superior, and the total area difference can be reduced effectively to about 3 percent if the two boundary conditions are maintained. The method of developing darts is necessary to further reduce the area difference and to achieve local alignment of meshes. This will be explained in details in Part II. On the other hand, in addition to area difference that can be used for the analysis of pattern results, potential sewing problems also need to be considered when patterns are transformed into practical garment slopers. For example, shoulder lines and
Uo
Customized pattern-making automation
Uo
37
Figure 8. Display of two patterns of bodice front
Vo (a)
Subjects
One half bodice
22-year-old woman
Bodice front Difference percentage Bodice back Difference percentage Bodice front Difference percentage Bodice back Difference percentage
41-year-old woman
(b)
Approximate meshes area in total
Central line intersecting with the waist line at right angle
166,389.27 0 156,252.36 0 147,183.66 0 1,532.7597 0
174,785.16 5.05 160,159.30 2.50 151,115.73 2.67 157,360.23 2.66
Table I. Comparisons between different pattern areas of half bodice front and bodice back
Note: Unit – mm2
underarm side lines in front and back pieces should be equal in length to make the patterns suitable for sewing and fitting. As indicated in Table II, the shoulder lines in the front and back piece generate a 2 mm difference in length. The result implies that different patterns are suitable for different models and that there is no one pattern that Human subjects Half bodice Boundary lines of approximated meshes Central line intersecting with the waist line at right angle Note: Unit – mm
22-years old Bodice Bodice front back
41-years old Bodice Bodice front back
94.30
94.61
106.23
103.90
100.82
100.44
120.89
101.30
Table II. Shoulder line length of different patterns for half bodice front and bodice back
IJCST 20,1
38
can reduce the gap length between the shoulder lines in the front and back pieces. On the average, there is a gap of 2 cm in length between shoulder lines in the front and back pieces, which needs improvement. Similarly, the underarm side lines of the front and back pieces should be equal, but as Table III shows, the difference in length between the side lines of the bodice front and bodice back sums to about 2 cm, possibly leading to sewing difficulties. This problem most likely occurs because the curved geometric surface of the originally designed bodice is an NURBS un-developable curved surface, so it is impossible to flatten such a surface into 2D meshes without any cracks, on the premise of non-overlapping. As a result, longer sewing lines are needed for a pattern with large cracks. Moreover, the evenness of the front piece surface is different from that of the back piece surface, and there are different boundary conditions for pattern generation (e.g. different kinds of alignments based on central lines or lower hems). Thus, the accumulated mesh space caused by the above reasons leads to mesh stacking and scattering at the most distant end. This generates the problem of gap length between side lines and results in different lengths of sewing lines in the front and back pieces. It can be observed from both Tables II and III that although different boundary conditions are adopted for the basic body structures of Asian women aged 22 and 41, respectively. The results indicate that there is no difference between the sewing lines of the front and back pieces generated from different boundary conditions. One possible reason for such results may be that the two models have different proportions, leading to varying evenness of bodice curved surfaces. It may therefore be concluded that garments made with different patterns are suitable for models of different statures, while garment patterns that are closest to the area of original curved surfaces can only be made through specific selection of boundary conditions for patterns based on individual differences. Identical to the pattern of the bodice front, the pattern of the bodice back can also derive three patterns after selecting the straight central line (Figure 9(a)), the straight lower hem, or both simultaneously, as shown in Figure 9(b). 6. Conclusions and discussions This study proposes an automatic pattern generation technique based on the expert-based experience of pattern-makers, applies mesh groups that approximate curved surfaces, establishes the principles of flattening and alignment to prevent mesh overlapping, and develops a set of automatic pattern generation software programs in order to achieve rapid and automatic 2D patterns for use. Human subjects Half bodice
Table III. Sideline lengths of different patterns for half bodice front and back
Boundary lines of approximated meshes Central line intersecting with the waist line at right angle Note: Unit – mm
22-years old Bodice Bodice front back
41-years old Bodice Bodice front back
475.36
474.05
480.56
494.88
522.06
486.87
496.30
515.38
UMax
UMax
Customized pattern-making automation 39
(a)
Vo
Figure 9. Display of two patterns of bodice back (b)
This study preliminarily proposes pattern automation generation technology based on 3D scissoring. Further, study includes moving forward on the considerations and evaluations of physic phenomena of fiber changing and knitting. In additions, the expert-based pattern generation of collars, sleeves, skirts, and trousers should be augmented to achieve the objective of direct output for sewing and fitting. Meanwhile, flattening of the trimmed NURBS Surface should be introduced to generate definitive and improved results for pattern boundary edges. Because NURBS surfaces are given for designing bodices in this study, the surfaces close to the sleeves and necks are approximated curved surfaces. As a result, deformation has occurred during practical applications. If the trimmed NURBS surface is adopted, this extraordinary instance may be reduced expectedly. Moreover, this study will be supplemented with the expert-based pattern generation of collars, sleeves, skirts, and trousers in the future. Based on the results and execution flow of this study at the present stage, we expect to design 3D garment surfaces for various kinds of styles and integrate expert-based knowledge and experience to automatically generate customized patterns, thus achieving the objective of direct output for sewing and fitting. References Fang, J.J. and Liao, C.K. (2005), “3D garment restyling based on computer mannequin model – Part I system kernel and formulas”, International Journal of Clothing Science & Technology, Vol. 17 No. 5, pp. 292-306. Kang, T.J. and Kim, S.M. (2000), “Optimized garment pattern generation based on three-dimensional anthropometric measurement”, International Journal of Clothing Science & Technology, Vol. 12 No. 4, pp. 240-54. Kang, T.J. and Kim, S.M. (2003), “Garment pattern from body scan data”, Computer-Aided Design, Vol. 35 No. 7, pp. 611-8. McCartney, J., Hinds, B.K. and Seow, B.L. (1999), “The flattening of triangulated surfaces incorporating darts and gussets”, Computer-Aided Design, Vol. 31 No. 4, pp. 249-60.
IJCST 20,1
40
McCartney, J., Hinds, B.K., Seow, B.L. and Gong, D. (2000a), “Dedicated 3D CAD for garment modelling”, Journal of Materials Processing Technology, Vol. 107, pp. 31-6. McCartney, J., Hinds, B.K., Seow, B.L. and Gong, D. (2000b), “An energy based model for the flattening of woven fabrics”, Journal of Materials Processing Technology, Vol. 107 Nos 1/3, pp. 312-8. Sul, I.H. and Kang, T.J. (2006), “Interactive garment pattern design using virtual scissoring method”, International Journal of Clothing Science & Technology, Vol. 18 No. 1, pp. 31-42. Tsai, M.J. and Fang, J.J. (2003), “A feature based data structure for computer manikin”, Taiwan Patent Pending 04083-09220535030, 2003; USA Patent Pending 10/699,640. Wang, C.C.L., Smith, S.S.F. and Yuen, M.M.F. (2002), “Surface flattening based on energy model”, Computer-Aided Design, Vol. 34 No. 11, pp. 823-33. Zhang, D. and Yuen, M.M.F. (2001), “Cloth simulation using multilevel meshes”, Computers and Graphics, Vol. 25 No. 3, pp. 383-9. Corresponding author Jing-Jing Fang can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Expert-based customized pattern-making automation: Part II. Dart design Jing-Jing Fang and Yu Ding Department of Mechanical Engineering, National Cheng Kung University, Tainan, Taiwan, and
Su-Chin Huang
Customized pattern-making automation 41 Received 10 September 2007 Revised 1 October 2007 Accepted 1 October 2007
Department of Fashion Design and Management, Tainan University of Technology, Tainan, Taiwan Abstract Purpose – Based on the knowledge of professional pattern makers, this paper aims to propose an expert-based automation technique of darts generation by aligning and drawing close meshes in basic pattern in Part I. Single dart development, such as waist-fitting dart, shoulder dart, armscye dart, side dart, and their select combination are also presented. Design/methodology/approach – In this paper, 3D garment surface is first approximated by a finite number of meshes. Patterns are developed by aligning and rotating of the flattened meshes under the constraint of overlay avoidance. The envelop areas between the developed patterns and the curved surface are dramatically reduced from 5 percent of basic pattern to below 3 percent after darts development. Findings – The development patterns are varied in their association with the subject’s body figures and the designed garment. Darts in a different location can reduce the total area difference between the flattening undevelopable surface and the original curved surface. Research limitations/implications – At the present stage the pattern development method cannot guarantee the uniqueness of pattern outline. Moreover, the pattern maker’s knowledge inputs in this paper can only apply to the subject whose waist girth is less than hip girth in circumference. Originality/value – The embedded pattern maker knowledge provides certain rules for pattern development from 3D design. Moreover, it is practical to be used and exported to modern 2D pattern software for further editing and revision. The same person is also used as a model after the patterns have been sewn into clothes. Keywords Clothing, Dimensional measurement Paper type Research paper
1. Introduction 3D finished products are often manufactured through the use of flat materials, such as sheet metal, leather ware, fabrics, etc. In traditional manufacturing industries such as fashions, shoemaking, shipbuilding, and aviation, raw materials in flat shape are used. The quantity and patterns of the materials are cut, and they undergo shaping steps such as extrusion, heat treatment, stretching, or twisting to construct the final finished products. Nowadays, even popular skincare products like the facemask could also be regarded as the development of a 3D face. Moreover, irregular 3D curved surfaces are what we know most about in our daily lives, but these are impossible to develop
International Journal of Clothing Science and Technology Vol. 20 No. 1, 2008 pp. 41-56 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810843520
IJCST 20,1
42
completely on a flat plane. In other words, if an irregular curved surface is flattened into a plane, it certainly contains those parts in which breaks or overlapping one another exist. Take the garment manufacturing industry as an example. Traditionally, in the garment industry, the fashion designer needs to make a conceptual planar sketch. The pattern maker then works out a garment pattern based on the sketch for trial production several times, and thus a usable planar sloper could be constructed. Darts are usually made in the pattern to give the garment a better fit on the subject and to make it visually appealing. Nonetheless, this traditional method of making a garment pattern is purely based on a 2D perspective. Without years of training and experience in pattern making, a worker will find it really difficult to make usable patterns. Flattening regular or irregular curved surfaces serves many applications. They can be developable (with zero Gaussian curvature) or un-developable (with non-zero Gaussian curvature). In relation to this, the current study puts forward the multi-orientation of applications. Any types of surfaces could be flattened, and the emphasis of this study is further investigation on the treatment of the development of the generating darts. The techniques for dart development take different positions, girths, or lengths into account when different algorithms or formulae are used. Based on the study of patterns from the 3D tailoring concept and the experience of pattern making experts, this study improves the formula for the crucial waist-fitting dart and its workflow, and develops automatic dart-developing techniques for shoulder darts, armscye darts, and side darts. 2. Related work The study of flattening techniques could be traced back two decades ago, and its aim has always been to obtain planar patterns for the development of a curved surface, for the convenience of scissoring, and for sewing into fit garments. Some studies use an opening process to reduce twisting and deformation after sewing. However, there are few works which discuss the techniques of how to generate darts for garment patterns. The followings describe some related work of this study for further investigations and assessments. As early as 1989, the deformation method of developable curved surfaces was studied (Redont, 1989). In contrast, Farin (1990) and Rogers and Adams (1990) proved that any developable curved surface must meet the condition in which the Gaussian curvature is zero. However, developable curved surfaces are limited only to regular curved surfaces such as cones and cylinders. Hence, limited applications cannot be inferred to un-developable curved surfaces. The method for the development of un-developable curved surfaces would inevitably produce local stretching or gaps. Therefore, decisions should be made in the development whether to use a fabric’s nature or not, which may be referred to as the energy method. The energy method can be divided into two main streams according to the development method chosen: node- and mesh-based development, respectively. 2.1 Node-based flattening Most node-based flattening methods assume that the mesh is stretchable and deformable. To make use of the limited element method in introducing the energy function for analysis, we use the method of optimization to gain the minimal stress distribution, by which the optimal flattening positions of the nodes of the curve in a 2D flattening plane are found.
In the study of Aono et al. (1996), one piece of fabric fits closely to a 3D curved surface. If the fabric has undergone curved surface analysis and it is found that the Gaussian curvature is not zero everywhere, it, of course, cannot fit the 3D curved surface perfectly. However, if the Gaussian curvature is positive, it will have overlapping fabrics after flattening, which can be solved by darts for perfect fitting. Meanwhile, if the Gaussian curvature is minus, it would contain a shortage in fabrics after flattening. In the first step of dart development, all mass points are represented by quadrilateral meshes. The authors assumed that the fabric does not stretch or retract and that the distance between mass points under stress is kept unchanged – only angles between mutually linked mass points change slightly. Aono provides a different dart-opening method at different positions to find the optimum result and makes use of visual techniques to reflect differences in curvature, including angle and deformation, which is beneficial to the study. McCartney et al. (2000) discussed the flattening process from a triangular mesh group comprised of spatial cloud points to a 2D flat plane. By means of the Delaunay Triangulation algorithm, cloud data were triangulated, a mesh curved surface was constructed, and the curved surface was developed because when the sides of the triangular mesh are closely connected, they would certainly produce mesh deformation. On such basis, an energy model was built to discuss the strain energy necessary for stretching the sides of the triangular mesh by finding the mesh point with the lowest energy. This was achieved by using the optimization method, obtaining the final flattening result through multiple iterations, and reducing the deformation energy through gaps and iterations between meshes. In this method, the initial seed mesh to flatten is designated manually, and the flattening process must be optimized. Therefore, it is labor consuming and inefficient. Wang et al. (2004) calculated the approximate Gaussian curvature on points of the triangular mesh curved surface. If one node’s Gaussian curvature was larger than the specified threshold, it implied that dart development was necessary for this node, from which the gap is produced till the borderline of the curved surface. The shortest path built by mesh sides was formed, and it was scissored and developed into a flat plane based on this path. 2.2 Mesh-based flattening Hinds et al. (1991) approximated an un-developable curved surface with quadrilateral and triangular mesh. If the curved surface is constructed by the regular constant u, c, and other methods, it was approximated by a quadrilateral mesh. If the curved surface is a relatively complicated free curved surface, it must be approximated by a triangular mesh instead. After an approximate mesh curved surface was produced, it was developed to a flat plane based on the Gaussian curvature on the curved surface. Hence, the gaps of regular shapes were produced. Nonetheless, it did not allow the mesh’s twisting and deformation. Therefore, there were densely distributed gaps in the development result. Azariadis and Asprathos (1997) triangulated the curved surface in advance, fixed the flattening direction, applied the flattening algorithm while keeping the length of mesh sides unchanged, transferred triangles to the flat plane in sequence, produced the initial flattening pattern, and then minimized distances between two adjacent rows or columns of meshes by the alignment method. In 2001, they adopted the flattening
Customized pattern-making automation 43
IJCST 20,1
44
method that kept the geodesic curvature unchanged, regarded the changing length of mesh as energy, adopted the rule of iteration maximization, and found the flattening result of minimum energy for this energy model (Azariadis and Asprathos, 2001). In mesh-based flattening, the flattening that is not overlapped produces many small pieces of mesh. By means of alignment techniques, the gaps between meshes are concentrated to produce gaps collectively. Although this causes the deformation of geometrical contour concentration, the alignment technique could preserve the total area after flattening effectively and is quite suitable to the flattening of garment patterns. As an overall review of the above literatures, the studies mentioned did not reflect the twisting deformation and distribution after flattening in general, and their effect was greatly reduced along with more meshes due to iteration optimization of the entire meshes. On the other hand, the studies mentioned above neither put those elements of positioning, length, or width into their considerations for evaluation, nor professional knowledge and experience from pattern makers is introduced into their study. Hence, in light of these shortages, this study adopts the method of stripe flattening generated at the initial development (Part I). Then in Part II of our paper, we develop a sequence of automatic dart-developing algorithms, provide the users several choices and reasonable combinations of various darts, and make the overall area of the garment pattern flattened by using the minimum energy approach, resulting in a better original area of the curved surfaces. 3. Expert-based knowledge for darts This section concludes combinations of dart development based on expert knowledge, i.e. the waist-fitting dart, the shoulder dart plus waist-fitting dart, the armscye dart plus waist-fitting dart, and the side dart plus waist-fitting dart. To reach the goal of darts generation which is a beautiful look and suit for sewing, several rules are set from pattern making experiences: . Whatever it is a waist-fitting dart, armscye dart, shoulder dart, or a side dart, the start point of a dart is usually about 2 cm away from the bust point. It is mainly for aesthetic purposes and the avoidance of sharp points in wearing. . The edges of darts are all straight lines for ease and suit for sewing. . The scissure width D in front bodies in Figure 1 should be equal to the difference, D 0 between the bust and waist girths in a half of front bodice, then: D ¼ D 0 ¼ ðbust length 2 waist lengthÞ=2;
.
ð1Þ
where the bust and waist girths belong to one-half of front bodice. For aesthetic aim, the scissure width of the pattern is usually between 2.5 and 3 cm in practice. If the difference between bust and waist lengths in one-half of front bodice is greater than 3 cm, it would usually develop a second dart combined into the pattern, such as a shoulder dart, a side dart, or an armscye dart. The start and the end points of the waist-fitting dart of bodice front located at the front princess line (Leong et al., 2007) are, respectively, 2 cm below the bust point and 2 cm above one-third of the height between the hip girth and the waist girth.
Side Neck Point Front Shoulder
Customized pattern-making automation
Side Neck Point Back Neck Point Back Shoulder Shoulder Point
Shoulder Point
Front Neck Point
45
Back Arm Hole Front Arm Hole Bust Line
Start Point
Bust Point Side Dart Start Point
Front Center Line
D' Back Center Line
d'
Waist Line
Side Line End point
Waist-Fitting Dart
End point
D Bodice Front Hip Line
d Bodice Back
Figure 1. Waist-fitting dart
Lower Hem Lower Hem
.
The scissure width d in back bodices in Figure 1 usually set to 2.5 or 3 cm if the difference between hip and waist lengths of a half of back bodice is greater than 3 cm, then: d þ d0 ¼ hip length 2 waist length; else: d ¼ hip length 2 waist length; and d0 ¼ 0
.
.
ð2Þ
where the hip and waist girths belong to one-half of back bodice. For aesthetic aim, the scissure width of the pattern is usually between 2.5 and 3 cm in practice. If the difference between bust and waist girths in one-half of front bodice is greater than 3 cm, it would usually develop a second dart combined into the pattern, such as a shoulder dart, a side dart, or an armscye dart. As the start point of the waist-fitting dart of the bodice back located at the back princess line (Leong et al., 2007) is about 2 cm above the bust line, and the end
IJCST 20,1 .
46
point of the waist-fitting dart of the bodice back is one-third of the height from the waist girth to the hip girth. The transfer directions of the shoulder dart, the armscye dart, or the side dart are shown by the arc arrows in Figure 2(a)-(c). Different darts are generated in the different parts on the pattern. We can transfer the shoulder dart (Figure 2(a)) to armscye dart (Figure 2(b)) by rotating part A. On the similar way, we also can rotate part B by transferring the armcye dart to side dart (Figure 2(c)).
4. Darts generation This study uses the scanning torso of a 40-year-old Taiwanese woman. The subject torso is reconstructed from the cloud points as a structural torso as shown in Figure 3(a)-(c) in different aspects. Subject’s digital tape-measurements of bust girth, waist girth, and hip girth are 88.4, 70.3, and 87.7 cm, respectively. The lower belly obviously protrudes, and Figure 3(b) shows that front-half of the subject’s figure lacks curves and changes. Figure 3(d) and (e) are the designed dress from self-developed 3D garment design software, Design3D (Fang and Liao, 2005), and are the example objects of this study. Based on the digital torso of the subject, a 3D garment style is initially created via the software, Design3D. Basic patterns are initially developed by the alignment method described in Part I. Two constraints are applied that the central line is kept perpendicular, and the lower hem is horizontal. All meshes are aligned according to these boundary constraints and toward the same direction of these constraints. Therefore, the gaps between meshes accumulate along with more meshes being aligned, larger gaps always generated at the corners far from the constraint boundaries. In such circumstances, gaps are developed at the corners near shoulder and also armhole.
A
A
B B
Figure 2. (a) Shoulder dart plus waist-fitting dart; (b) armscye dart plus waist-fitting dart; (c) side dart plus waist-fitting dart
(a)
(b)
(c)
Customized pattern-making automation 47
(a)
(b)
(d)
(c)
(e)
To reduce the openings and the entire area difference between the developed patterns and the 3D design, meshes in one piece of fabric are firstly divided into several zones. Each zone applies its own constraints independently. Local alignment is carried out by its defined constraints, with the aim of reducing the gaps in a local zone to a minimum. Meanwhile, in each zone, the mesh gaps are automatically concentrated toward the edge in each zone due to the alignment. Therefore, these concentrated gaps can be regarded as darts in garment patterns. 4.1 Waist-fitting dart To reduce gaps among meshes at the waist-girth, this study presented a sequence of processes to approach traditional garment pattern with waist-fitting dart. The waist-fitting dart in bodice front is the used example to describe the development procedure.
Figure 3. (a) Front view of subject torso; (b) side view of subject torso; (c) isometric view of subject torso; (d) 3D garment; (e) garment from another perspective
IJCST 20,1
48
Based on torso feature positions on the surface of the digital subject (Leong et al., 2007) and the pattern-making principles provided in the last section, we could identify the start and end points of the waist-fitting dart in bodice front, meanwhile, retain the meshes that involve start-, end-, and bust-point in the same vertical stripe (Figure 4(a)). Moreover, the meshes that involve start-, end-, and blade point are retained in the same vertical stripe. According to the rules above, the meshes in bodice front could be divided into four sub-zones, as shown by the signs in Figure 4(b). Zone 1 is the subset of all meshes above the rows of meshes across the start point in the waist-fitting dart. Zone 2 is the subset of meshes enclosed by the rows of the start point and the end point, the column of the end point, and the central line. Zone 3 is the subset of meshes enclosed by the rows of the start point and the end point, the column of the end point and the sideline. Zone 4 is the meshes with the exception of the above zones in bodice front. Along with the constraints shown in sequence as below, the alignment method is applied to some sub-zones for the purpose of automatic generation of the waist-fitting dart: (1) The central line retains straight alignment and the meshes in Zone 2 are close up. (2) The aligned row of end point is perpendicular to the straightened central line, and the meshes in Zone 4 are close up. (3) Rotate the meshes in Zone 3, move horizontally, and then align to generate a waist-fitting dart. This part will be described in detail in the following paragraphs.
Bust Point 1
Bust Line
Start Point
P1'
P1
Ω1
L1
L2 D'
Waist Line P2, P3
2
3 Ω2
End Point
P2',P3' L4
L3
P4'
P4 Hip Line
4
D
Figure 4. (a) Basic pattern; (b) zones diagram; (c) ideal draw of waist-fitting dart (a)
(b)
(c)
(4) Finally, the meshes in Zones 1 and 4 are individually close up to the constraint boundary by the correspondent rows of the start point, end point, and the central line. The followings describe the whole process of the waist-fitting dart development in Zone 3 in details. Before dart development, the initial positions of L1 and L2, L3 and L4 should be the same. This is means that the scissure of the ideal waist-fitting dart is a combination of two equilateral triangles. As shown in the edge contour in Figure 4(a), the basic pattern developed in Part I of this study contains no darts inside. The scissure width D of the ideal waist-fitting dart in front piece would exhibit the body figure to make the dress wearing aesthetically appealing. In other words, the aim of generation of the waist-fitting dart from the basic pattern is to try to evenly distribute the difference between the bust girth and the waist girth, or the difference between waist girth and the hip girth to the left and right sides of Zone 3. Since, Zone 3 is the crucial part of forming triangle-shape dart in pattern, its operations of meshes re-alignment are described by the following steps: (1) Zone 3 is firstly divided into sub-zones V1 and V2 in basic pattern in Figure 5(a). Lines P 1 P 2 and P 3 P 4 are their associated boundaries along sideline in basic pattern. In final stage after dart development, points P2 and P3 should at the same position. (2) Line up the external strip in V1 and V2 in order to be perpendicular to the horizontal bust girth (Figure 5(b)). (3) Move the rest meshes in V1 and V2 to close up the straightened meshes strip (Figure 5(c)). Move V2 accordingly from P 04 to P4 (Figure 5(d)). Rotate the remaining meshes in V2 by: D0 21 u1 ¼ tan ð3Þ yðP 3 0 Þ 2 yðP 4 Þ where D 0 is gained by equation (1). It then draws close to the bust line to avoid partial meshes penetration between Zone 4 and V2 (Figure 5(e)). (4) Move meshes in V1 accordingly from P 02 on the position of point P 03 (Figure 5(f)). Rotate meshes in V1 by: 0 21 xðP 2 Þ 2 xðP 1 Þ u2 ¼ tan ð4Þ yðP 2 0 Þ 2 yðP 1 Þ meshes in V1 are drawn close to the straighten stripe meshes P 01 P 02 to avoid penetration between V1 and V2 (Figure 5(g)). (5) Finally, the meshes in Zone 1 are drawn close to V1 in order to reduce the gaps in between. At this point, the waist-fitting darts are completed (Figure 5(h)). Figures 6 and 7 show the breaks distribution chart between the meshes before and after the development of the waist-fitting dart. The break lengths among these meshes are shown by grey ruler, from white (0 mm) to black (5 mm). It is obvious that larger accumulation gaps position at the region of side breast and the edge of armhole in basic pattern. After development of waist-fitting dart, gaps lengths near the side breast are significant lessened. Comparisons of envelop areas of one-half of bodice front and bodice back are listed in Table I. As shown by their associated envelop area, the front-half
Customized pattern-making automation 49
IJCST 20,1
50
Figure 5. Sequences of developing waist-fitting dart
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
q1
piece’s envelope area is 3.84 percent larger than the original curved surface, and the back-half piece is larger by 3.51 percent. After the development of a single waist-fitting dart, the front-half piece drops to 3.83 percent larger, and the back-half piece drops to 2.91 percent larger. The dropping percentages of envelop area are not what we predict significantly. The reasons of that may be due to the subject’s figure lacks curves and changes, circumferences ratios of both waist-hip and waist-bust are about 0.8. Moreover, the designed virtual dress (Figure 3(d)-(e)) are able to conceal subject’s figures imperfection by free-handed 3D garment creation. In this paper, we present a brand new
Customized pattern-making automation 51
Figure 6. Breaks distribution among meshes before and after the waist-fitting dart applied on one-half of the bodice front
Figure 7. Breaks distribution among meshes before and after waist-fitting dart applied on the bodice back
Envelop area (mm2)
Approximated area of the 3D curved surface Basic pattern Development of waist-fitting dart
Difference percentage Front-half piece Back-half piece (percent) (percent)
Front-half piece
Back-half piece
177,996.86 183,324.02
185,863.80 193,294.06
0 3.84
0 3.51
184,068.50
191,265.78
3.83
2.91
Table I. Comparisons of the envelop areas of one-half of bodice front and bodice back
IJCST 20,1
52
pattern developing method without using the traditional pattern developing formulas which is in widespread use in Asian garment industry. 4.2 Dart combination Besides, the waist-fitting dart described in previous section, the armscye dart, side dart, and shoulder dart can be also applied to reduce the break summations among those meshes in pattern. Compare to the operations of generating waist-fitting dart, operations of the other three single darts developing are rather easy. Simply assign the start point of the dart, its associated single dart is able to be developed by strip meshes alignment operation. Based on the expertise experience of making garment patterns, it is concluded that there are four basic darts and their combinations. They are: (1) waist-fitting darts plus side darts (Figure 8(b)); (2) waist-fitting darts plus armscye darts (Figure 8(c)); (3) waist-fitting darts plus shoulder darts (Figure 8(d)); and (4) side darts only (Figure 8(e)). According to the figures of the mid-age subject’s we used, smoother break level pictures in pattern with darts show that the combination of waist-fitting dart plus side dart may be a suitable front-half pattern for the select subject. Associative quantified details of these four combination darts are given in Table II. Envelope area of basic pattern slightly decreases after darts developing. According to the table, pattern with side dart provides the closest envelop area to the original area of the 3D designed surface. In Figure 8(e), the small breaks gather in the side region of bust. Besides, from the aspect of garment manufacturing, curved side line causes the difficulty of sewing between front bodies and back bodies patterns. Therefore, to compromise the break level pictures and the envelop area assessment, it is concluded that pattern with both side dart and waist-fitting dart may be the best choice for the subject.
Figure 8. Break evaluation of bodice front: (a) basic pattern; (b) waist-fitting dart plus shoulder dart; (c) waist-fitting dart plus armscye dart; (d) waist-fitting dart plus side dart; (e) single side dart
(a)
(b)
(c)
(d)
(e)
5. Derivational benefits In addition to the advantages of easy control over mesh movement and formation of mutual position correlations, the technique of meshes alignment with overlapping avoidance proposed in this study can also be applied in follow-up computer-aided pattern printing. Common computer-aided design (CAD) software trends to produce deformed and distorted textures (Figure 9(a)) on 3D geometric modeling easily. The results of this study could therefore serve as the base for un-distortable texture printing in common CAD software. With the use of the developed meshes in this study, we are able to create un-distorted texture printing (Fang and Liang, 2007) as shown in Figure 9(b).
Customized pattern-making automation 53
6. Practical try-on In view of the transformation from a virtual design to a physical textile, the follow-up operation provides outputs in both HP-GL and DXF format into full scale printer which links up with the processes of computerized manufacturing in garment industry. Figure 10 shows the resulting output in DXF format importing by AutoCAD for the front and back pieces with combination darts. Figure 11 shows the workflow of the use of a computerized 3D deep V-collar basal design in Figure 11(a), and the use of the pattern development method to generate darted patterns in Figure 11(b), which would then be sewn for a trial put-on to the subject (Figure 11(c)). The entire process begins
Approximated area of the 3D curved surface Basic pattern Waist-fitting dart Waist-fitting dart þ shoulder dart Waist-fitting dart þ armscye dart Waist-fitting dart þ side dart Side dart
Envelop area (mm2)
Area difference (percent)
177,996.86 183,324.02 184,168.50 183,976.44 183,142.42 182,874.17 182,482.64
– 3.84 3.83 3.36 2.89 2.74 2.52
Table II. Developing darts in one-half bodice front and back
Figure 9. (a) Pattern printing on NURBS surfaces; (b) un-distorted printing based on the development method (a)
(b)
IJCST 20,1
54
Figure 10. Pattern export to AutoCAD in DXF format
(a)
Figure 11. Practical try-on from 3D design
(b)
(c)
from body scan, body features extraction (Leong et al., 2007), 3D garment creation (Fang and Liao, 2005), patterns generation, and finally, trial-on. Pattern development in this study is one of the crucial milestones of this “forward” engineering. The applications could be investigated for its potential usage in garment industry. By investigating fitting outcomes in practical try-on, an infrastructure was built up for improving further studies in computerized 3D garment design and pattern developing in automation. 7. Conclusions and discussions In this study, we present an expert-based garment pattern making knowledge system by computer technology. By lessening the envelop area of the developed pattern from its associated basic pattern, a potential pattern is formed. Such meshes alignment method is capable not only be used in darted pattern generation, but also provide a better solution of un-distorted printing in 3D which is a useful tool in CAD system. Further, fitting assessment is able to be carried out by sewn these full-scale developed patterns and then try-on the subject. In present stage, we present an export-based method of how to generate practical patterns, in which, however, cannot guarantee the uniqueness of pattern outline. Although the pattern developmental method lacks of uniqueness in current stage, most importantly, we propose a new idea of “forward” engineering in garment manufacturing. Different from modern computer-aided pattern drawing based on Bunka pattern which is popular in Asia, we present an intuitive design workflow from 3D garment creation to 2D pattern generation. In the near future, physic phenomena include fiber material and textile techniques will be taken into account to reveal optimal patterns for customization. References Aono, M., Denti, P., Breen, D.E. and Wozny, M.J. (1996), “Fitting a woven cloth model to a curved surface: dart insertion”, IEEE Computer Graphics in Textures and Applications, Vol. 16 No. 5, pp. 60-70. Azariadis, P. and Asprathos, N. (1997), “Design of plane development of doubly curved surfaces”, Computer-Aided Design, Vol. 29 No. 10, pp. 675-85. Azariadis, P. and Asprathos, N. (2001), “Geodesic curvature preservation in surface flattening through constrained global optimization”, Computer-Aided Design, Vol. 33 No. 8, pp. 581-91. Fang, J.J. and Liang, K.S. (2007), “Flattening method for undeveloped surfaces and distortion elimination method for texturing irregular surface”, Taiwan Patent I284289, 2007/7/21 , 2025/10/27. Fang, J.J. and Liao, C.K. (2005), “3D garment restyling based on computer mannequin model – Part II results and applications”, International Journal of Clothing Science & Technology, Vol. 17 No. 5, pp. 307-19. Farin, G. (1990), Curves and Surfaces for Computer Aided Geometric Design, A Practical Guide, 2nd ed., Academic Press Inc., New York, NY. Hinds, B.K., McCartney, J. and Woods, G. (1991), “Pattern development for 3D surfaces”, Computer-Aided Design, Vol. 23 No. 8, pp. 583-92. Leong, I.F., Fang, J.J. and Tsai, M.J. (2007), “Automatic body features extraction from marker-less scanned human body”, Computer-Aided Design, Vol. 39 No. 7, pp. 568-82.
Customized pattern-making automation 55
IJCST 20,1
56
McCartney, J., Hinds, B.K., Seow, B.L. and Gong, D. (2000), “Dedicated 3D CAD for garment modeling”, Journal of Materials Processing Technology, Vol. 107 Nos 1-3, pp. 31-6. Redont, P. (1989), “Representation and deformation of developable surfaces”, Computer-Aided Design, Vol. 1 No. 21, pp. 13-20. Rogers, D.F. and Adams, J.A. (1990), Mathematical Elements for Computer Graphics, 2nd ed., McGrawHill, New York, NY. Wang, C.C.L., Wang, Y., Tang, K. and Yuen, M.M.F. (2004), “Reduce the stretch in surface flattening by finding cutting paths to the surface boundary”, Computer Aided Design, Vol. 36 No. 8, pp. 665-77. Corresponding author Jing-Jing Fang can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Investigation on the seam’s quality by sewing of light fabrics
Investigation on the seam’s quality
Diana Germanova-Krasteva and Hristo Petrov Technical University of Sofia, Sofia, Bulgaria
57
Received 28 February 2007 Revised 14 September 2007 Accepted 14 September Purpose – This paper aims to define the dominating factors having influence on the quality of the 2007 seam by sewing of fine fabrics and their optimum values.
Abstract
Design/methodology/approach – Ten factors defining a seam’s quality are chosen on the base of literary research. There was a check for concurrence of specialists’ opinions, as the Spearman’s coefficient of concordance was determined. A designed experiment with variation of the first three of the arranged factors was made. Mathematical models for the tensile and the aesthetical properties of the seam were devised, and also optimization made. Findings – A classification of the properties, defining seam’s quality, is made. A cause-effect diagram of Ishikawa with aiming parameter – quality of the seam – has been developed. On the basis of a survey the factors that have a great deal of influence on it are presented. Mathematical models for seam’s strength, elongation and smoothness are produced by changing the following factors: straining of the upper thread, size of the needle and load on the pressing foot are worked out. Received models are optimized. Research limitations/implications – The research was conducted using a sewing-machine class 301. Received results and conclusions refer to seams made from base material – fabric from PES and sewing thread from 100 percent PE. Practical implications – Optimum values have been established for the straining of the upper thread, the size of the needle and the load on the pressing foot by sewing of fabric from synthetic silk with mass applicable machines, needles, and sewing threads. Originality/value – The research has been done in several directions: systemizing the seam’s properties for evaluation of its quality and the factors defining it, inquiry into the significance of the different factors and implementation of a designed experiment. Consultations were made with a broad circle of specialists and these results are given in visual systems (schemes and graphs). Keywords Quality, Optimization techniques, Modelling, Fabric testing Paper type Research paper
Introduction The complex evaluation of the quality of a sewing product includes the evaluation of the construction and the pattern, the structure, the composition, and the properties of the used material and not on the last place – the quality of the passed seams. Seam’s quality is difficult for research object, because on it many factors have influence – kind of the seam; kind, structure and properties of the sewing thread and of the stitched material, seam’s parameters, technological adjustments, etc. It can be seen from different aspects – as from the point of view of the aesthetics, as from the point of view of its tensile and exploitation properties. The appearance of the seam forms the aesthetical properties of the sewing product. It is inadmissible curving of the seam’s line, poorly tightening of the seams ends leading to incompactly joining, torn threads or disturbing of the seam’s wholeness and unevenness of its density.
International Journal of Clothing Science and Technology Vol. 20 No. 1, 2008 pp. 57-64 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810843539
IJCST 20,1
58
During the exploitation of the product the seam bears different loadings and some zones of it are put under intensive stretching or bending. Tensile properties of the seam define the steadiness of product’s construction to deformation and are characterized through the strain strength in cross direction, elongation in direction of the seam and the bending stiffness. Zones of joining the details have to be soft enough (well-bending) and with satisfying elasticity. At the same time, the seam and the details need to have the necessary stiffness to put up resistance against the deformation created by the needle due the sticking. During the exploitation, as the product so and the seams are put under wearing up, which is often evaluated by the friction resistance. At the time of exploitation the products are under the impact of other physical and chemical factors, too – moisture, temperature, sweat, light, different chemical products, etc. The underwear and upper clothes are washing up and overcoats are proceeding through dry cleaning. The resistance of the product against washing and dry cleaning is a basic factor for evaluation of the exploitation properties of the seam. Describing factors are summarized in the scheme, shown on Figure 1. Seam’s quality in considerable degree is defined by the adjustment of the sewing machine. In about 30 percent of the different cases the main reason for gathering of the textile materials in the seam’s zone is wrong adjustment of the machine. In about 10-15 percent of the cases, the causing factor is the sewing needle and with the same percent – the sewing thread (Mitova and Kostova, 1996). Optimizing the work of the sewing machine includes suitable choice of: strain of the upper thread and of the under thread, seam’s density, sewing speed, load on the pressing foot, sort of the transport mechanism, highness of the transport teeth, sizes of the needle’s plate hole, kind of the pressing foot, shuttle and elastic spring (stitch class 300), brake. SEAM’S QUALITY
ESTETICAL PROPERTIES
BRIGHT PROPERTIES
Evenness of seam’s line
Strength in cross direction
Evenness of seam’s density
Seam’s elongation
Completeness of the seam
Bending stiffness of the seam
EXPLOATATIONAL PROPERTIES
Friction resistance Washing resistance Dry cleaning resistance
Figure 1. Factors for seam’s quality evaluation
Steadiness of natural aging Loosing of the seam
Factors, defining seam’s quality Ishikawa’s diagram is developed with aim parameter “seam’s quality” (Figure 2). In it are included the basic factors, having direct or indirect influence on the parameter, and then are systemized in categories. As factors are included properties of the materials (fabric and sewing thread), seam type (it depends on it purpose), technical means of production (machine, needle, pressing foot), personnel and the influence of environmental factors. In “tuning of the machine” are included the above described parameters and characteristics of the stitch-formation elements.
Investigation on the seam’s quality 59
Evaluating of the factors having influence on the seam’s quality by sewing of fine fabrics Under fine fabrics are meant materials with mass between 120-150 g/m2. Such mass per unit area have fabrics made from natural or chemical silk, but it is possible to be produced and from other materials. For determination of influence’s degree of the different factors is developed a inquiry, where are included ten factors suggested by the firm Gu¨ttermann (Heckner, 2003). Every factor is given a number, showed in Table I.
PRESSING FOOT
STICH KIND - lockstitch
THREAD - kind
- form
- overlock
- number of parallel seams
- linear density
- chain stitch
- angel between seam’s and threads’ direction
- kind - kind
- size
- tuning of the machine
- noise level
SEAM’S QUALITY
- state of the machine
- qualification
- top - structure
- environment temperature
and distance between them
- strength
- material
- time of the day
- presence of face seams
- material
- surface treatment
EXTERNAL FACTORS
SEAM - density
- motivation
- number of layers FABRIC
NEEDLE
MACHINE
Figure 2. Diagram for analyzing of seam’s quality
PERSONNEL
No.
Factor
1 2 3 4 5 6 7 8 9 10
Straining of the upper thread Straining of the under thread Stitch density Sewing speed Load on the pressing foot Kind of transporting mechanism Hole of needle’s plate Kind of the pressing foot Brake for thread straining Needle’s size
Table I. Factors’ numbering
IJCST 20,1
60
The inquiry cart is filled in by ten specialists – technologists, mechanics of machines in sewing firms and teachers in sewing technology. Every expert (identified from A to J) has given a mark from 1 to 10 concerning the degree of influence for each factor on the seam’s quality (mark 1 – least influence, mark 10 – strongest influence). The arrangement of the factors of the different specialists is introduced in Table II. Done is check for the concurrence of specialists’ opinions. Calculated is the Spearman’s coefficient of concordance on a matrix with coincide rows, and the received value is W ¼ 0.57. Its significance is checked through Pierson’s criterion at significance level of 0.01. The check shows that there is a concurrence between specialists’ opinions. On Figure 3 is showed the influence of the factors, defined on the basis of a common expert opinion. Designed experiment It is made a designed experiment and mathematical models for the following characteristics, defining seam’s quality, are worked out:
Table II. Matrix of the ranges
Factor
1
2
3
4
5
6
7
8
9
10
A B C D E F G H I J
10 10 10 10 9 8 10 8 8 10
5 10 10 6 9 8 5 8 6 10
4 10 7 5 7 3 4 7 5 8
6 8 9 7 8 3 6 3 6 4
9 8 10 8 9 6 8 8 7 7
10 6 9 8 9 8 8 9 6 6
8 10 8 8 6 5 10 8 7 5
5 6 9 8 8 4 6 8 7 5
6 6 7 8 8 8 3 6 6 4
10 10 9 9 10 6 10 7 6 9
100 90
Factors 87.5
Straining of the upper thread
Degree of influence, %
80
Figure 3. Factors determining seam’s quality by sewing of fine fabrics
70
64
61
60
20 10
Sewing speed Load on the pressing foot 35
40 28.5
34
Straining of the under thread Stitch density
55.5 46
50
30
71.5
67
Kind of transporting mechanism Hole of needle’s plate Kind of the pressing foot Brake for thread straining Needle’s size
0
. . .
Investigation on the seam’s quality
seam’s strength in cross direction; elongation in seam’s direction; and appearance (gathering).
Determination of seam’s strength is done according to EN ISO 13 935-01 and-02, for seam’s elongation – according to EN ISO 13 934-1 and for gathering – with photo standards according to AATCC 88B (Reumann, 2000). There factors are varied, that according the specialists showed the best influence on the seam’s quality in fine fabrics, namely: . straining of the upper thread – x1; . needle’s size – x2; and . load on the pressure foot – x3.
61
It is used an optimal compositional plan with basic level and intervals of varying of the factors, presented in Table III. Experiment’s circumstances The preparation of the samples is realized on sewing machine JUKI DDL 5550-4 with sewing speed of 4,000 stitches per minute. The base cloth is fabric super silk from 100 percent PES with mass per unit area 102 g/m2.The stitches are passed using a sewing thread COATS-EPIC from 100 percent PE, number 180. Sewing needles are trade-mark GROZ-BECKERT. About 14 variants of samples are produced, as for determination of strength and elongation of the seam are made five trials and for appearance – three trials. Mathematical models After processing of the received results are worked out the following models: . for strength in cross direction: y ¼ 16:175 2 0:26 · x1 þ 0:62 · x2 þ 0:71 · x3 2 1:475 · x21 þ 0:625 · x22 .
ð1Þ
for elongation in seam’s direction: y ¼ 25:3 þ 1:59 · x1 þ 1:38 · x2 2 0:975 · x1 · x2 þ x2 x3 2 4:05 · x21
Factors’ levels Basic Intervals of varying Upper level Lower level
ð2Þ
Denotation Unit
x1 cN
x2 No.
x3 N
Xo,i Ji Xu,i Xl,i
40 20 60 20
70 10 80 60
25 15 40 10
Table III. Levels of the variable factors
IJCST 20,1
62
.
for seam’s gathering: y ¼ 4:146 þ 0:33 · x2 þ 0:5 · x3 2 0:292 · x1 · x2 2 0:48 · x21
ð3Þ
The made after Fisher’s criterion check shows that all three mathematical models are adequate with level of significance 0.01. Optimization of the models After optimizing the mathematical models it was determined that the maximum values of the investigated characteristics for evaluation of the seam’s quality are received when the ruling parameters have values, as the presented in Table IV. They show that for achievement of an optimal quality for the chosen materials is recommended: . straining of the upper thread between 35 and 40 cN; . needle’s size 80; and . load of the pressing foot 40 N. i.e. a little straining of the needle’s thread and high pressure of the pressing foot, such as be ensured reliable leading of the fabric by the high-sewing speed. More interesting is the fact that seam’s best quality is approached using a needle with the largest size. It should follow to larger deformation of the sewed material and destruction of more warp and weft threads. The reason for this effect probably has to be found in better stability of the thicker needles against deformation during the sewing process. Graphical presentation of the results The influence of the factors on the characteristics for evaluation of seam’s quality is shown in Figures 4-6. Because the graphics are 3D in their presentation take part two factors: straining of the upper thread and load on the pressing foot. They are drawn for the optimal value of the needle’s size (No. 80). The factors are given in coding values. Conclusion A research and an analysis were made and as a result of that are systemized and classified the characteristics for elavuation of seam’s quality. There has been developed a cause-effect diagram for analysis of the reasons for appearance of quality problem. Survey is made among specialists, which aim was the arrangement of ten factors, determining the seam’s quality by sewing of fine fabrics. On its basis are chosen three
Value Table IV. Optimum values
Straining of the upper thread, cN Needle’s size Load on the pressure foot, N
Strength Pmax ¼ 17.94 daN
Elongation 1max ¼ 27.7 percent
Gathering AATCC ¼ 5
38.24 80 40
41.52 80 40
34 80 40
Investigation on the seam’s quality
Strength, daN 17 16
63
15 14 13 1 0.5 0.5
0
Thread straining
–0.5 –1 –1
–0.5
1
0 Pressure load
Figure 4. Influence of the thread straining and the load on the pressing foot on the seam’s strength
Elongation, % 28 26 24 22 20 18 1 1
0.5 0 –0.5
Thread straining
–1 –1
–0.5
0.5 0 Pressure load
Figure 5. Influence of the thread straining and the load on the pressing foot on the seam’s elongation
AATCC 5 4.5 4 3.5 3 2.5 1 0.5
0.5
0 Thread straining
– 0.5 –1 –1
– 0.5
0 Pressure load
1
Figure 6. Influence of the thread straining and the load on the pressing foot on the seam’s gathering
IJCST 20,1
64
factors and a designed experiment varying these factors is made. Set are their optimal values for receiving of the best seam’s strength, elongation and appearance. As a result of the optimization recommendational parameters for work with synthetic fabrics, having such mass per square unit, are given. References Heckner, R. (2003), “Optimization of the sewing machine as basic precondition for non-gathered and smooth seam by light fabrics”, Textile and Apparel, Vol. 51 Nos 9/10, pp. 58-61. Mitova, B. and Kostova, D. (1996), “Evaluation of seam’s quality”, paper presented at the International Scientific Conference of EMF, Sofia. Reumann, R-D. (2000), Pru¨fverfahren in der Textil-und Bekleidungstechnik, Springer-Verlag, Berlin. Further reading Pavlinic, D., Gersak, J., Demsar, J. and Bratko, I. (2006), “Predicting seam appearance quality”, Textile Research Journal, Vol. 76 No. 3, pp. 235-42. Stylios, G. and Sotomi, O.J. (1996), “An intelligent environment for flexible apparel production”, IMACS/IEEE Discrete Events and Manufacturing Systems, pp. 173-7. Stylios, G., Sotomi, O.J.R., Zhu, R., Xu, Y.M. and Deacon, R. (1995), “The mechatronic principles for intelligent sewing environment”, Mechatronics, Vol. 5 Nos 2/3, pp. 309-19. Web sites www.amann.com/pdfs/en/s_t_broschueren/naehfadenbedarf.pdf www.coatsddb.com/industrial/apparel/sewingsolutions
Corresponding author Diana Germanova-Krasteva can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,2
Evaporative cooling and heat transfer in functional underwear Markus Weder and Rene´ M. Rossi
68
Empa Materials Science and Technology, St. Gallen, Switzerland, and
Cyril Chaigneau and Brice Tillmann Damart, Despature et Fils S.A., Roubaix Cedox, France
Received 8 May 2007 Revised 12 September 2007 Accepted 12 September 2007
Abstract Purpose – The purpose of this investigation is to measure seven different underwears on a sweating torso with differing relative air humidity (30, 50, 80 and 95 per cent RH) and at a fixed ambient temperature of 308C to determine the influence of the water vapour partial pressure of the environment on the moisture transport properties of various materials. Design/methodology/approach – All measurements in this investigation were accomplished with the authors’ sweating torso which simulates the thermal- and humidity release of the human body. Four different sweating rates (50, 75, 100 and 150 g/h *torso) were selected for this investigation. Findings – It was established that the partial pressure difference did not correlate directly with the evaporative cooling. In general, higher evaporation rates were observed in the dry climate conditions. However, with low-sweat rates, the highest relative humidity (95 per cent) generally resulted in greater evaporative cooling than the lowest surrounding humidity conditions (30 per cent). In this investigation, a blended fabric made of PES/Vinal exhibited the most efficient evaporative cooling for all the sweat rates, as well as for the four relative humidity conditions chosen. Research limitations/implications – All received results are based on a surrounding temperature of 308C (summer climate), for other temperatures the results may be different. Originality/value – The investigation shows that both the relative humidity and the sweat rate have a major influence on the heat loss. Keywords Evaporation, Cooling, Clothing, Heat transfer, Humidity Paper type Research paper
International Journal of Clothing Science and Technology Vol. 20 No. 2, 2008 pp. 68-78 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810850450
Introduction The heat and moisture transport characteristics of textiles are indispensable factors for the determination of the use for defined activities and surrounding conditions. The metabolic heat and humidity production of the human body and the environmental climatic conditions (air and radiant temperature, relative humidity, and wind velocity), are the other parameters to quantify the transfer of energy through clothing layers. In order to obtain good thermal physiological (wear) comfort, there must be a balance between heat production and heat dissipation. If the heat production is higher than the heat loss, the body usually starts to produce liquid sweat. The body can only prevent a hyperthermia effectively, if this moisture can be evaporated quickly in the vicinity of the skin. Therefore, in this situation, materials are required that enable the liquid sweat to evaporate very quickly and efficiently in order to maximize the cooling performance. In this investigation, diverse functional underwear materials were compared with an underwear made of pure cotton at different relative humidities and sweat rates
measured on a sweating torso (Figure 1) to determine the overall heat transfer. The sweating torso was used to study the heat and moisture behaviour of fabric layer combinations for different humidities and sweat rates, rather than using a whole-body manikin (Fan and Chen, 2002; Richards and Mattle, 2001), designed to test ready made garments with the additional effects of air layers between the clothing layers and even forced ventilation of the microclimate due to repetitive body movements. Several studies have already investigated the influence of the relative humidity of the surrounding air on the effective water vapour resistance, the moisture transport characteristics of membranes alone or membranes laminated on textile-supporting materials. When using hydrophilic polymer layers, it was established that the water vapour transport was less affected by the temperature than by the moisture concentration (Gibson, 1993, 2000; Osczevski, 1996). The influence of the temperature on water vapour transport with nine different polymer materials was investigated (Gibson et al., 1997) with the dynamic moisture permeation cell (ASTM F2298). Polymers with hydrophilic surfaces absorbed more humidity and increased the water-vapour transport rate. As some investigations showed, the temperature gradient and the hydrophilicity of the materials examined play an important role with regard to the condensation of water vapour (Gretton et al., 1998; Osczevski, 1996; Rossi et al., 2004). Nefzi et al. (2004) found that with
Functional underwear
69
Upper Guardring
Test sample
Sweat glands (54)
Layers of Torso: Aluminium Polyethylene PTFE
Lower Guardring
Dripped off Water
Water Feeding and Power supply Scale
Figure 1. Sweating torso
IJCST 20,2
70
non-hydrophilic materials, the water vapour transport did not change appreciably with the relative humidity in the microclimate as the change in the diffusion rate caused by the relative humidity increasing from 40 to 70 per cent was only a few per cent. The kind of material and construction of the textile, which affect the water vapour transport, were found to be more important than the ambient conditions. Like Osczevski (1996), Nefzi et al. (2004) established a slight increase in the water vapour diffusion with increasing temperature. Most investigations were based on a vaporous moisture transport, Farnworth and Dolhan (1985) and Farnworth et al. (1990) examined the water vapour transport with different relative humidity ratios and found hardly any humidity-related correlation with micro-porous PTFE and PU-membranes. On the other hand, the hydrophilic coatings showed clearly a lower water vapour transport with low humidity than with high humidity. The selection of the method to be used for testing the water-vapour resistance is a matter of great importance. For instance, with measuring methods that employ a more or less thick air layer between the water and specimen (cup methods, Hu et al., 2001; Kumaran, 1998a, b; Nilsson and Hansen, 1981), hydrophilic materials – in particular – clearly display higher resistance values with an increasing air gap thickness (Gibson, 1993). According to Gretton et al. (1998), hydrophilic polymers or clothing systems with relatively small transmission rates transport the water vapour under isothermal conditions much better than micro-porous polymers with an existing temperature gradient. The presence of a temperature gradient, especially at low-environmental temperatures, will most probably induce condensation of part of the water vapour in the textile layers, which may be beneficial for moisture transport in hydrophilic materials, but will usually have a negative effect in hydrophobic materials. Wang and Yasuda (1991) investigated not only the water vapour transport but also the liquid transport in hydrophilic and hydrophobic materials in comparison with non-finished materials. They found that the materials with hydrophilic treatment showed a higher wicking effect but that the water vapour resistance was not influenced by the surface treatment. Kwon et al. (1998) examined the effect of hydrophilic and hydrophobic characteristics of clothing on the thermal physiology of seven female subjects and showed that the hydrophilic samples clearly exhibited more favourable physiological parameters (lower heart frequency, deeper skin and rectal temperatures, and less micro-climatic humidity) for the reduction of a heat stress than the hydrophobic materials. The goal of this work was the systematic variation of the environmental relative humidity and the sweat rate for different materials blends with changing hydrophilic and hygroscopic properties in order to determine the heat transfer due to evaporative cooling effects. Test materials For this investigation, seven samples with different material compositions but with similar weight per unit area were selected, except the PES sample, which was lighter than the others. The thermal resistance (Rct) was also similar for most of the samples, with two exceptions: sample chlorofibre/acrylic (ClA) had a resistance of about 40 per cent higher than the others and sample PES/Vinal (PeV) of about 30 per cent lower than the others. All samples were knitted fabrics.
Measurement method (sweating torso) The measurements of the evaporative cooling were made with a heatable and sweating cylinder, which has a similar size as a human trunk. This torso is made of different material layers having similar properties as the human skin. A total of about 54 separately controlled sweat outlets are distributed over the cylinder surface to control the sweat rate, which can be liquid or vaporous. In this study, however, we did only use sweat in liquid form. A detailed description of this apparatus is provided in Zimmerli and Weder (1997). The seven underwear materials were fitted crease-free to the vertical-standing sweating torso and subjected to a horizontal wind flow with a speed of 1 m/s. The samples were applied with a defined pretension of approximately 10 per cent around a cylinder. The sweating torso was regulated at a constant temperature of 358C, with different sweating rates released during the measurement. All seven samples were measured three times in the four different climatic conditions with four different sweat rates with each test separately in an acclimatisation phase, a sweating phase and a drying phase. As the thermal resistance Rct of the seven materials examined was not exactly the same and in order to determine the overall influence of sweat (Qw) on the total heat loss (Qtot), the dry heat loss determined during a first phase without water release was subtracted from the total heat loss of the sweating torso during the second phase (sweating phase). The total heat loss is the sum of the dry heat loss (Qc), the evaporative heat loss (Qe) and the heat loss due to the higher conductivity of the wet materials (Qcw). In this study, Qe and Qcw were not considered separately (equation below) and for questions of simplicity, we refer to the two effects of moisture collectively as being a “wet heat loss.” Qtot ¼ Qc þ Qe þ Qcw ¼ Qc þ Qw Qw ¼ Qtot 2 Qc Results and discussion Influence of the specific weight on the thermal insulation The thermal insulation of a clothing is the combined insulation provided by the clothing (intrinsic thermal resistance of the clothing) and the resistance of the air surrounding the clothing. This thermal resistance Rct is dependent on the thermal conductivity of the material. In the presence of humidity from either the ambient air with a high-relative humidity or the influence of sweating, hygroscopic or hydrophilic materials absorb all or part of the moisture which increases the thermal conductivity. Sample CO exhibits a heat transmission coefficient (measured with a calorimeter plate) of 77 W/m2K, sample MTM – 45 W/m2K, sample MoM – 47 W/m2K and sample PeV 65 W/m2K. The samples CO and MTM have exactly the same Rct, when measured in the sweating guarded hot plate (Table I) but sample CO has an approximately 70 per cent higher thermal conductivity at 30 per cent as well as at 80 per cent relative humidity. In general, a 10 per cent higher thermal conduction was observed at higher relative humidity (95 per cent, compared with 30 per cent) for the measured samples. The fibre type and construction of the fabric affects the thermal insulation much more than the specific weight of the materials, for instance the sample PES exhibits a relatively high-thermal insulation of 31 £ 102 3 m2K/W compared to the other samples, despite its lower weight.
Functional underwear
71
IJCST 20,2
Sample Material
72
ClA CO PES MTM
Table I. Description of the samples
PeM MoM PeV
85%Chlorofibre / 15%Acryl 100%Cotton 100%Polyester 46%Modacryl / 43%TEK / 8% Modal / 3% Polyamide 85%PES / 15%Modal 85%Modacryl / 15%Modal 82%PES / 18%Vinal
2
Specific weight g/m 210 210 145 200 210 215 190
Rct £ 102 3 Construction (m2K/W) Interlock Interlock Pique´ Pique´ Interlock Interlock Pique´
Rct/weight
50 35 31
0.238 0.167 0.214
35 36 40 24
0.175 0.171 0.186 0.126
Note: The Rct values were measured with the sweating guarded hot plate according to ISO 11092
Influence of the fibre material and the construction As to be expected, a reduction in the wet heat loss with increasing relative humidity of the ambient air was only observed by the higher sweat rates. As the driving force for the water vapour transfer is the water vapour partial pressure gradient between the surface of the torso and the atmosphere, one would expect that the evaporative heat loss and therefore also the wet heat loss should decrease proportionally with increasing relative humidity in the environment. However, this could not be observed, as can be seen for the samples PES (Figure 2a) and CO (cotton) (Figure 2b). On the contrary, it is interesting to note that with a relative humidity of 80 per cent (34 mbar), the wet heat loss with sample PES was even higher than with a relative humidity of 50 per cent (21 mbar) for all sweat rates with the exception of 50 g/h. A reduction in the wet heat loss of sample PES was observed when the relative humidity was changed from 30 to 50 per cent and with a sweat rate of 150 g/h. However, sample PES generates a practically humidity-independent wet heat loss. The sample PeV exhibits the most efficient wet heat loss of all seven underwear materials examined (Figure 3a), especially with lower relative humidity (30 and 50 per cent).This high-wet heat loss can partly be explained by the hydrophilicity of this sample which favoured the evaporative cooling, but not only as samples like MTM or PEM showed a better wicking effect than PeV. The two samples PES and PeV evaporated 85 per cent (þ /2 2 per cent) of the sweat water supplied during the sweating phase at a relative humidity of 80 per cent when 150 g/h was supplied. The classification of the different samples according to their wet heat loss was not the same for all sweat rates. For instance, with a low-sweat rate, the sample ClA showed a higher wet heat loss than CO or PeM, but it cooled the least with a higher sweat rate. This may be explained by the fact that the material was relatively hydrophobic, and thus did not absorb or wick liquid well. Therefore, with high-sweat rates, part of the sweat water supplied dripped down the sweating torso and was therefore no longer available to contribute to the wet heat loss. The cotton sample CO exhibited a low-wet heat loss for all the sweat rates and, all the relative humidities used. Owing to its hygroscopic nature, cotton first absorbs and stores humidity, preventing evaporation of water. Evaporation will only start when the water storage in the cotton fibres has reached a certain level; however from this moment, the cooling efficiency may be as high as for some synthetic materials. However, during the measuring time chosen (1 h), this evaporative rate of cotton was not high enough to achieve the same cooling performance as the other samples.
Functional underwear 150 140
Evaporation cooling in (W)
130
73
120 110 100 90 80 70 60 50 40 30
10
15
20
30% r.H.
25
30
50% r.H.
35
40
80% r.H.
45
95% r.H.
Partial pressure (mbar) Sweat rates in g/h:
50
75
100
150
(a) 110 100 Evaporation cooling in (W)
90 80 70 60 50 40 30 20 10 10
15
20 50% r.H.
30% r.H.
25
30
35 80% r.H.
40 95% r.H.
Partial pressure (mbar) 50
75
100 (b)
150
45
Figure 2. Mean wet heat loss as a function of the partial pressure for the four sweat rates: (a) PES; (b) cotton
IJCST 20,2 170 160 150 140 Evaporation cooling (W)
74
130 120 110 100 90 80 70 60 50 40 10
15
20 50% r.H.
30% r.H.
25
30
35 80% r.H.
40 95% r.H.
45
Partial pressure (mbar) Samples:
ClA
Co
PES
MTM
PeM
MoM
PeV
(a) 60
Evaporation cooling (W)
50
40
30
20
10 10
15
20 50% r.H.
30% r.H.
Figure 3. Wet heat loss at the highest and the lowest sweat rates at: (a) 150 g/h; (b) 50 g/h
25
30
35 80% r.H.
40 95% r.H.
Partial pressure (mbar) Samples:
ClA
Co
PES
MTM (b)
PeM
MoM
PeV
45
Influence of the sweat rates on the wet heat loss At a relative humidity of 30 per cent, the wet heat loss increased almost linearly with the sweat rate for all the samples except ClA (Figure 4a). Figure 4b shows the mean increase of wet heat loss for all samples dependent on of the sweat rate and demonstrates that this almost linear increase was only found with a relative humidity of 30 per cent. At higher relative humidities (80 and 95 per cent), there was a noticeable reduction of the increase in wet heat loss with increasing sweat rates. The mean wet heat loss of all the samples was very similar for a sweat rate of 50 g/h, but it was about 30 per cent lower at a relative humidity of 95 per cent compared to 30 per cent, when the sweat rate was 150 g/h, even reaching 50 per cent for the sample PeV. These differences between the single samples could be attributed to different thermal conductivity coefficients when wet. At a low-sweat rate and with a low-relative humidity, the influence of this “wet” thermal 140
Functional underwear
75
R.H.:
130
30%
120 Evaporation cooling (W)
110
50%
100 80%
90 80
95%
70 60 50 40 30 20 40
60
80
100 Sweat rates in (g/h) (a)
120
140
160
Evaporation cooling (W)
180 160
ClA
140
Co
120
PES
100
MTM
80
PeM
60 40
MoM
20
PeV
0 40
60
80
100 120 Sweat rates in g/h (b)
140
160
Figure 4. (a) Wet heat loss in dependency of the four different sweat rates and the four different relative humidities (mean value of all seven samples); (b) wet heat loss in dependency of the sweat rate for a relative humidity of 30 per cent for all seven samples
IJCST 20,2
76
conductivity was low as the moisture evaporated before it could really moisten the sample. But also with higher sweat rates (100 and 150 g/h) and despite the wetter material, the thermal conductivity was found not to be as important as evaporative cooling for the total wet heat loss. With the sample PeV, for example, the thermal conductivity increased by only about 7 per cent as a consequence of the rise in the relative humidity from 30 to 80 per cent, whereas the wet heat loss increased by more than 30 per cent. The thermal conductivity of the samples used increased by approximately 6-13 per cent when the relative humidity changed from 30 to 95 per cent. Influence of the environment relative humidity on the wet heat loss With an increase in the partial pressure difference (corresponding to a reduction of the relative humidity), the wet heat loss shows a lower increasing rate for the higher sweat rates than for the low rates, even though the water vapour permeability through a fabric is proportional to the partial pressure difference as shown in the equation below. WVP ¼
1 · 0:672 · Dp Ret
with, WVP – water vapour permeability; Ret – water vapour resistance of the fabric. For instance, when the partial pressure ratio increases by a factor of 2.5, the wet heat loss only increases by a factor of about 1.5 for a relative humidity of 30 per cent (Figure 5). With the lowest sweat rate of 50 g/h, the wet heat loss even reduced with increasing partial pressure difference. This is attributable to the already mentioned dominance of the thermal conduction in relation to the evaporative cooling. As a result of the increased partial pressure with higher relative humidity, the higher thermal conductivity compensates the reduction in the heat loss through evaporative cooling. Therefore, the wet heat loss will be highly influenced by the wet heat conductivity when the relative humidity is high and but mostly by the evaporative heat loss when the relative humidity is low, as shown in Figure 6. 1.8
ratio Qw = Qw(pa)/Qw(95%)
1.7
Figure 5. Wet heat loss quotient (Qw 95% ¼ 1) at four different humidities and four sweat rates
1.6
30%
1.5
SW=50 g/h
1.4
50%
SW= 75 g/h
1.3 SW=100 g/h 1.2
95%
80%
SW=150 g/h
1.1 1 0.9 0.8 1
1.5
2 ratio: pa(95%)/pa
Note: Average results of all 7 samples are shown
2.5
3
Functional underwear
Q evaporation and Q dry: mean value of all 7 samples R.H.= 30% R.H.
R.H.= 95%
160 140 120 Q [W]
100
Qcw
80
Qe
60
Qc
40 20 0 sweat rate:
50 g/h
150 g/h
50 g/h
150 g/h
The dry heat loss Qc and the wet heat loss could be measured on the torso. However, the differentiation in the wet heat loss between wet heat conductivity and evaporative heat loss could not be made. For the determination of these two factors as shown in Figure 6, we used the assumptions that the wet heat conductivity was the same for both climates with the sweat rate of 150 g/h and that the heat conductivity at 50 g/h and 95 per cent relative humidity was equal to the dry heat conductivity. Conclusions We assessed the wet heat loss of different underwear fabrics for different relative air humidities and found not linear relationship with the partial pressure difference. For low-sweat rates (50 and 75 g/h), no significant differences in the wet heat loss could be found for different relative humidities in the environment, as all the moisture supplied could be evacuated to the atmosphere. With high-relative humidities and low-sweat rates, the humidity could not fully evaporate, and was stored in the fabric, inducing a higher heat loss by wet thermal conductivity. For high-sweat rates and high-relative humidities in the environment, this wet thermal conductivity becomes the dominant factor for wet heat loss. At a low-relative humidity, the wet heat loss increased proportionally to the increase in sweat rate. However, when the relative humidity in the environment was raised, the wet heat loss increased with a much lower rate in dependency of the sweat rate. The investigations show that both the relative humidity and the sweat rate have a major influence on the heat loss. These two factors as well as the material composition will determine whether the thermal conductivity or the evaporative cooling is the dominant factor in wet heat loss. Using results from such investigations, it should now be possible to design sportswear with an optimised sweat management for defined applications and activities. References Fan, J. and Chen, Y.S. (2002), “Measurement of clothing thermal insulation and moisture vapor resistance using a novel perspiring fabric thermal manikin”, Measurement Science & Technology, Vol. 13 No. 7, pp. 1115-23. Farnworth, B. and Dolhan, P.A. (1985), “Heat and water transport through cotton and polypropylene underwear”, Textile Research Journal, Vol. 55 No. 10, pp. 627-30.
77 Figure 6. Estimation of the distribution of the wet heat loss at two sweat rates (50 and 150 g/h) for a relative humidity of 30 and 95 per cent
IJCST 20,2
78
Farnworth, B., Lotens, W.A. and Wittgen, P. (1990), “Variation of water-vapor resistance of microporous and hydrophilic films with relative-humidity”, Textile Research Journal, Vol. 60 No. 1, pp. 50-3. Gibson, P.W. (1993), “Factors influencing steady-state heat and water-vapor transfer measurements for clothing materials”, Textile Research Journal, Vol. 63 No. 12, pp. 749-64. Gibson, P.W. (2000), “Effect of temperature on water vapor transport through polymer membrane laminates”, Polymer Testing, Vol. 19 No. 6, pp. 673-91. Gibson, P.W., Elsaiid, A.E., Kendrick, C.E., Rivin, D. and Charmchi, M. (1997), “A test method to determine the relative humidity dependence of the air permeability of woven textile fabrics”, Journal of Testing and Evaluation, Vol. 25 No. 4, pp. 416-23. Gretton, J.C., Brook, D.B., Dyson, H.M. and Harlock, S.C. (1998), “Moisture vapor transport through waterproof breathable fabrics and clothing systems under a temperature gradient”, Textile Research Journal, Vol. 68 No. 12, pp. 936-41. Hu, Y., Topolkaraev, V., Hiltner, A. and Baer, E. (2001), “Measurement of water vapor transmission rate in highly permeable films”, Journal of Applied Polymer Science, Vol. 81 No. 7, pp. 1624-33. Kumaran, M.K. (1998a), “An alternative procedure for the analysis of data from the cup method measurements for determination of water vapor transmission properties”, Journal of Testing and Evaluation, Vol. 26 No. 6, pp. 575-81. Kumaran, M.K. (1998b), “Interlaboratory comparison of the ASTM standard test methods for water vapor transmission of materials (E 96-95)”, Journal of Testing and Evaluation, Vol. 26 No. 2, pp. 83-8. Kwon, A., Kato, M., Kawamura, H., Yanai, Y. and Tokura, H. (1998), “Physiological significance of hydrophilic and hydrophobic textile materials during intermittent exercise in humans under the influence of warm ambient temperature with and without wind”, European Journal of Applied Physiology and Occupational Physiology, Vol. 78 No. 6, pp. 487-93. Nefzi, N., Jouini, M. and Ben Nasrallah, S. (2004), “Water vapor transfer through textile under a temperature and humidity gradient”, Journal of Porous Media, Vol. 7 No. 2, pp. 133-41. Nilsson, E. and Hansen, C.M. (1981), “Evaporation and vapor diffusion resistance in permeation measurements by the cup method”, Journal of Coatings Technology, Vol. 53 No. 680, pp. 61-4. Osczevski, R.J. (1996), “Water vapor transfer through a hydrophilic film at subzero temperatures”, Textile Research Journal, Vol. 66 No. 1, pp. 24-9. Richards, M.G.M. and Mattle, N.G. (2001), “A sweating agile thermal manikin (SAM) developed to test complete clothing systems under normal and extreme conditions”, paper presented at Human Factors and Medicine Panel Symposium – Blowing Hot and Cold: Protection Against Climatic Extremes, RTO/NATO, Dresden. Rossi, R.M., Gross, R. and May, H. (2004), “Water vapor transfer and condensation effects in multilayer textile combinations”, Textile Research Journal, Vol. 74 No. 1, pp. 1-6. Wang, J.H. and Yasuda, H. (1991), “Dynamic water-vapor and heat-transport through layered fabrics. Part 1. Effect of surface modification”, Textile Research Journal, Vol. 61 No. 1, pp. 10-20. Zimmerli, T. and Weder, M.S. (1997), Performance of Protective Clothing, ASTM STP 1273, American Society of Testing and Materials, Philadelphia, PA. Corresponding author Markus Weder can be contacted at:
[email protected] To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Optimized pattern grading
Optimized pattern grading
Elizabeth Bye, Karen LaBat, Ellen McKinney and Dong-Eun Kim Department of Design, Housing and Apparel, College of Design, University of Minnesota, St Paul, Minnesota, USA
79 Abstract Purpose – To evaluate current apparel industry Misses grading practices in providing good fit and propose grading practices to improve fit. Design/methodology/approach – Participants representing Misses sizes 6-20 based on ASTM D 5585 were selected. The fit of garments from traditionally graded patterns was assessed. Garments were fit-to-shape on participants. Traditionally graded patterns were compared to fit-to-shape patterns using quantitative and qualitative visual analysis. Findings – Current apparel industry grading practices do not provide good fit for consumers. The greatest variation between the traditionally graded patterns and the fit-to-shape patterns occurred between sizes 14 and 16. For size 16 and up, neck and armscye circumferences were too large and bust dart intakes were too small. Research limitations/implications – This study was limited to a sheath dress in Misses sizes 6-20. Future research should assess the fit of garments from traditionally graded patterns for other size ranges. Practical implications – Multiple fit modes are needed in a range of more than five sizes. The fit model should be at the middle of a sizing group that does not range more than two sizes up or down. Originality/value – There are few studies on apparel grading that test fit of actual garments on the body. The analysis documents the real growth of the body across the size range and suggests that changes in body measurements and shape determine the fit of a garment. These findings impact future research in apparel and the practices of apparel manufacturers.
Received 5 June 2007 Revised 31 October 2007 Accepted 31 October 2007
Keywords Garment industry, Clothing, Measurement, Women, United States of America Paper type Research paper
Introduction The US apparel industry is challenged in meeting consumer’s needs for well-fitting apparel. About 49 percent of women have difficulty finding clothes that fit (Kurt Salmon Associates, 2004). Every year 12 percent of all clothes sold is returned due to poor fit, 36 percent of women’s apparel is returned (Barbaro, 2006). Finding well fitted ready-to-wear clothing takes a considerable amount of consumers’ time (LaBat, 1987). Large size women identified size and fit as the most common problem when questioned concerning garment satisfaction (Chowdhary and Beale, 1988). Fit is the relationship of body to garment. The standard practice of mass-production is to create a range of sizes by increasing and decreasing from a sample size garment that fits the sample size model (Price and Zamkoff, 1996). For example, Misses sizes 4-20 are graded up or down from a sample size 8 (Schofield, 2000). However, the human body does not grow proportionally (O’Brien and Shelton, 1941) as suggested in size charts that guide grading practices. Current size charts do not accurately reflect body measurements across sizes or changes in body shape. As a result, grading practices contribute to fit problems.
International Journal of Clothing Science and Technology Vol. 20 No. 2, 2008 pp. 79-92 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810850469
IJCST 20,2
80
There are few studies on apparel grading that evaluate fit across a size range and test fit of actual garments on the body. In order to identify the optimum grading practice, two types of pattern grading methods, traditionally graded patterns and fit-to-shape patterns were tested on the body and compared. As the body growth is not proportionate, this study highlights the role of body shape variation in individuals with the same bust, waist, and hip circumference and height measurements in the comparison of two pattern sets. Review of literature Size surveys O’Brien and Shelton (1941) conducted an anthropometic survey of 10,042 women divided into four age groups, 18-29, 30-44, 45-59, and 60 years and over. The survey results formed the basis for women’s sizing systems (Bye et al., 2006). In 1958, the US Department of Agriculture published US Voluntary Product Standard CS 215-58 by O’Brien and Shelton, the first US sizing standard based on anthropometric data (US Department of Commerce, 1958). Revision of CS 215-58 was made to more accurately reflect the population of women and a new edition of the standard, PS 42-70 was published (US Department of Commerce, 1970); however, it was not based on anthropometric research (Schofield and LaBat, 2005a). The current Misses size standard, ASTM D 5585-95, provides body measurements for the adult Misses figure type, sizes 2-20. It was developed by the American Society for Testing and Materials in 1995 and the measurements were derived from the original PS 42-70 database (D 5585-95 Standard, 2001). The size USA survey used 3D body scanning technology to measure over 10,000 men and women whose age range was from 18 to 80 to provide new information regarding body measurements and shapes. For example, less than 10 percent of participants with a size 8 bust met the current size 8 standards for waist and hips: What we’ve learned is that it’s not good enough to take a size 10 fit model and add a half inch up and down the line. People’s body shapes don’t change in a straight line up or down (Bond, 2004).
Grading background After World War I, ready-to-wear production increased, requiring more efficient pattern and grading methods (Cooklin, 1990). The ideal grading technique produces a range of patterns with the same style proportions and fit as the sample size. The proportional and the traditional method are the most common. Both use standard and variable grade rules. Standard grades are the same for all sizes while variable grades change according to the difference in circumference between sizes (Price and Zamkoff, 1996). Proportional grading produces patterns that change proportionally in circumference and length (Cooklin, 1990). However, people larger in circumference are not necessarily taller, which results in fitting problems (Cooklin, 1990). Traditional grading uses a standardized set of relationships to guide the treatment of lines, shapes, and forms on patterns (Bye, 1990). For example, in traditional grading the yoke line is not graded in length to maintain neckline to yoke proportion. Bye and Delong (1994) compared the visual difference between the images of Misses sizes 8-20 representing traditional and proportional grading. Neither method maintained the visual effect across the entire size range.
Relationship of size charts to grading Based on 40 US size charts, Schofield and LaBat (2005a) found the following assumptions: . the use of constant intervals between sizes within a grade; . the increase of all vertical measurements as size increases; . a constant difference between principle girths for all sizes; and . the use variable grades within a size range. These practices simplify pattern grading (Kunick, 1967) and reduce number of sizes. Schofield and LaBat (2005a) also found that most measurements in size charts are not useful for developing grade rules. A comparison of the 1970 US Sizing Standard intervals with Price and Zamkoff (1974) grade rules for the front bodice showed that 87 percent were not based on the sizing standard. Seven grading assumptions were compared to the 1988 Anthropometric Survey of US Army Women with no assumptions representing real body data (Schofield and LaBat, 2005b, p. 149). This was especially true for larger women as an inappropriate increment accumulates as size increases, causing more fitting problems. For example, the body measurement increase at the side neck point was smaller than was suggested by grade rules, resulting in a wide neckline. To understand the effect of grading assumptions, Schofield and LaBat (2005b) suggested fit-testing graded garments. Loker et al. (2005) conducted a fit analysis of a ready-to-wear pant on 156 participants Misses sizes 4-16 whose age range was 34-55 years. The pant patterns had been graded from a size 8 fit model. Participants wore the pant size that provided best fit at hip level. 3D scans were used to provide visualizations in a digital fit analysis procedure. Overall, fit was rated higher for sizes 6, 8, and 10 and lower for sizes 12, 14, and 16. Researchers concluded because the smaller sizes were closer in proportion to the fit model than the larger sizes they received higher ratings. Method Research objectives and design Based on research indicating a relationship between grading practices and poor fit, the following research objectives were developed: . evaluate traditional industry grading practices; and . propose optimum grading practices. Research was designed to assess the fit of a sheath dress developed from traditionally graded patterns and compare them to fit-to-shape patterns. By comparing these two sets of patterns, optimum grading practices were developed. The sheath dress was chosen as the test garment because it conforms closely to the body circumferences of the torso. It hangs from the shoulder and reflects the major portion of body growth between sizes. The resulting optimum grading practice can be applied to any styles based on a sheath dress. The method included selecting participants representing a size range, perfecting a garment for the participant representing the sample size, grading from sample size to all sizes using traditional methods, developing custom fit garments for all participants, and conducting fit tests.
Optimized pattern grading
81
IJCST 20,2
82
Participant selection Participant recruitment began after receiving Human Subjects Committee approval. About 34 potential participants were screened for participant selection. Potential participants were scanned with a Human Solutions VITUS 3D scanner to quickly extract measurements. The VITUS 3D scanner uses eye safe laser sensors and eight cameras to capture the surface dimension of the body. Most 3D body scanners are laser or light based with slight differences in equipment and data output (Istook and Hwang, 2001). Light scanners usually take longer to process data than the laser-based systems. The VITUS 3D scanner with more cameras positioned on four columns captures more body contours including top of the head than the other system. Participants’ scans were processed through the ScanWorXe AutoMeasuree program to extract height, bust, waist, and hip measurements. The goal was to find eight participants representing Misses sizes 6, 8, 10, 12, 14, 16, 18, and 20 with size 8 representing the sample size. The size dimensions were defined using the ASTM D5585-95 Standard (2001). Eight participants were selected. Each participant matched the bust, waist, and hip measurements of a designated size. The age range of the participants was 19-36. A limitation of this study is that one garment was tested on one participant in each size. For generalization to a broader population, testing on a larger sample is necessary. Data collection Dress for sample size. A sleeveless sheath dress with fit ease was developed to fit the size 8 participant (designated sample size). The dress was developed without a waistline seam, and with fullness controlled by side seam front bust darts, two front and back waist darts, and back shoulder darts. A center back opening was used. A panel of expert judges, with more than 25 years each of professional experience and research in fit analysis, evaluated the garment fit. Major circumference ease amounts were 2.00 in. at the bust and the hip and 1.00 in. at the waist. Ease amounts were consistent across the size range. Participants were scanned wearing the dresses to provide a permanent record of the dress on the body. Traditional graded patterns. The size 8 pattern was the sample size for the traditional grading method. The pattern was graded down to size 6 and up to sizes 10, 12, 14, 16, 18, and 20. Grading intervals were based on ASTM D 5585-95 and Grading Techniques for Fashion Design by Price and Zamkoff (1996). The size 6-10 patterns had a 1.00 in. grade interval, the size 10-18 patterns had a 1.50 in. grade interval, and the grade interval from 18 to 20 was 2.00 in. The size 8 pattern was digitized and graded using OptiTexe PDS software. Fit-to-shape graded patterns. To develop fit-to-shape graded patterns, the expert judges altered the fit of the traditionally graded dress on each participant. Judges used Armstrong’s (2000a, b) fit criteria: . center front and center back aligns with the body center; . armscye fits smoothly; . the waist level aligns with body waist; . no stress or gapping at neckline; . side seam hangs vertically; . shoulder seam centered on the shoulder;
.
.
skirt hangs straight from the hip to the hem and cross grain parallel to the floor; and no strain at bust, waist, or hip.
Patterns were adjusted to reflect dress alterations. This process was repeated until fit judges determined that garment fit was as near to perfect as possible. Fit-to-shape patterns were digitized into the OptiTex PDS software for comparison to the traditionally graded patterns.
Optimized pattern grading
83
Data analysis Quantitative and qualitative analyses were used to interpret differences between traditionally graded and fit-to-shape patterns. Quantitative analysis included total number of pattern adjustments and dimensional differences between patterns. Qualitative analysis included visual assessment of body and pattern shapes. The number of adjustments needed for the dress to achieve good fit was counted. The counts were organized by body measurement and grade location for each size. Total adjustments were summed and averaged per size to evaluate traditional grading practices. Front and back traditionally graded and fit-to-shape patterns were nested (Figure 1). These nests were examined as full-size print-outs and digitally. Changes in measurements from size-to-size were analyzed. Distances between largest and smallest sizes at cardinal grading points were measured using OptiTex and compared between the two nests.
Figure 1. Traditionally graded nest and fit-to-shape pattern nest
IJCST 20,2
84
Quantitative changes from size 6 to 20 were determined with small to large differences noted. Key pattern segments were measured using OptiTex software. For example, the length of the armscye seam was measured, as well as the length and depth along the x and y planes. Traditionally graded patterns for each size were nested with the corresponding fit-to-shape patterns. Corresponding pairs of sizes 16, 18, and 20 patterns were also analyzed as full-scale allowing a detailed view of differences. The paired patterns were visually evaluated to identify variations in shape and to identify features that required further analysis. Body scans of participants in scan suits and garments provided permanent records for convenient analysis of body shape difference and garment fit. Body scans were analyzed digitally in 3D rotation, on paper and on transparency sheets, in four poses (front, left-side, back, and right-side) (Figure 2). Body scans (front- and side-view) printed on transparency sheets were used to compare adjacent sizes and the sample size 8 participant to explain differences in fit-to-shape patterns. To provide additional body analysis, front-views of body scans were outlined using Adobe Illustratorw. The silhouettes, with each size traced in a different color, were printed on paper. This method was useful to compare changes in body shape for the full size range (Figure 3). Multiple methods were used to evaluate cardinal pattern points (Table I). Results and discussion Traditional grading practices The first objective was to evaluate traditional industry grading. Based on the assumption that the ASTM sizing standard and traditional grading practices accurately represent body differences from size-to-size, dresses made using information from these sources should have provided acceptable fit. Participants were selected who met ASTM circumference and height requirements. The number of adjustments needed to achieve
Figure 2. Four-view pose of three dimensional scan
Optimized pattern grading
85 Size 8 Size 14 Size 20
Figure 3. Layered body outlines
Cardinal pattern points Bust, waist, and hip circumference measurements/cross-body and waist dart grades Vertical measurements/height grades Shoulder length and neck base measurements/cross-shoulder, shoulder-level and neck grades Armscye circumference and depth measurements/armhole grade (depth and width) Bust point-to-bust point and neck-to-bust point measurements/front side bust dart grade Overall, body shape
Quantitative data Qualitative visual data Pattern Body change 2D patterns 2D patterns shape count (OptiTex) (OptiTex) 3D scans outlines X
X
X print and transparency
X
X
X X
X
X
X transparency
X
X
X
X transparency
X
X
good fit varied (Table II) with the average number of adjustments of 3.4 per size. Rasband and Liechty (2006, p. 64) indicates that neckline and shoulder alterations “set up a chain reaction of other fitting problems” and strongly advises against purchasing garments that require these alterations. More than half of the adjustments for these participants were in the neck and shoulder areas. She further recommends against
Table I. Data analysis methods
86
garments requiring multiple alterations as they will “likely never fit properly.” Therefore, we conclude that traditional grading practices do not provide good fit across the size range. Discussion of these adjustments is arranged by cardinal pattern points. Optimum grading practices The second objective was to identify optimum grading practices. Size charts, based on body circumferences and lengths, are not easily translated to grading. Size chart measurements and the incremental size change at cardinal pattern points were evaluated. The cardinal pattern points for the grades are shown in the Figure 4. Bust, waist, and hip circumference measurements/cross-body grade. In ASTM D 5585-95, the circumference measurements incrementally increase. Visual data of the fit-to-shape 2D pattern nest showed some increase in bust and hip grade intervals in three of the seven sizes, even though the participant’s circumference measurements Adjustment
Size 10
Size 12
Size 14
Size 16
Size 18
Size 20
Total
1 1 1 0 3
1 0 0 1 2
0 0 0 0 0
2 0 0 1 3
3 1 1 0 5
2 1 1 2 6
3 0 1 0 4
12 3 4 4 24
Neck Shoulder Armscye Side seam Total
Armhole Width Grade
Armhole Depth Grade
Table II. Number of adjustments required for each size
Size 6
Shoulder Level Grade
IJCST 20,2
Cross Shoulder Grade Neck Grade
Height Grade
Bust Dart Grade
Shoulder Point
Under Arm Point Dart Intake
Side Neck Point Center Front Neck Point
Front Bust Point Upper Dart Point
Waist Dart Grade Lower Dart Point Front
Figure 4. Key pattern points Cross-Body Grade
met the standard. The size 10 participant required additional bust circumference, the size 14 participant required additional hip circumference, and the size 18 participant required additional bust and hip circumference. No adjustments were made to waist dart intakes. Based on the visual data of body scans on transparency sheets, we attribute the increase in circumference to variance in body shape. Visible differences in the shape of side-to-side and front-to-back profiles across the size range were noticeable. When compared to the size 8 participant, the size 10 participant, with a D-cup bust, had more body mass distribution on the front of the bust, the size 14 participant had a more prominent hip from front-to-back, and the size 18 participant, with a D-cup bust, had more fat distribution on the front of the bust and front of the waist. We found no conclusive trend in cross body grade. The adjustments are likely based on individual body variation that we would not expect to be accommodated by a standard size range. We observed change to the body shape outlines. However, this was not visible in the 2D fit-to-shape pattern nest. Previous research shows that body outline and pattern side seam shape are not directly related (Gazzuolo, 1985). Vertical measurements/height grades. With the ASTM Standard there is an assumption of 0.50 in. increase in stature per size. Quantitative data of 2D fit-to-shape patterns showed that there were no fit adjustments needed at the bust, waist, or hip level. This may be because the participants, except size 20, matched the ASTM height requirement. The shorter height difference did not affect the fit of the sheath dress with its undefined waist. Shoulder length and neck base measurements/cross-shoulder, shoulder-level and neck grades. Grading of the neck and shoulder areas is inter-related. The ASTM shoulder length measurement is related to the cross-shoulder and increases variably by grade interval. ASTM provides a neck base circumference measurement for each size. However, pattern growth is based on the horizontal neck width grade and the vertical shoulder level grade. The traditional front and back neck grades result in a total circumference increase of 0.50 in. per size, however, the final neck circumference is dependent on the shape of the neck curve. Shoulder. Traditional and fit-to-shape patterns for each size were compared. The shoulder area required minor changes. Visual analysis of the 3D transparency sheets was used to relate change in pattern shape to change in body shape. The size 6 participant required a reduction in shoulder length which was probably due to more angular shoulders than the size 8 participant. The size 16 and 18 participants required movement of the shoulder seam for correct positioning due to their rotated shoulder points and curved upper backs. It is not clear if this is an individual body variation or a change in body shape due to increased size. In the visual analysis of the body shape outlines, the shoulders did not exhibit a regular pattern of growth. We found no conclusive trend with regard to shoulder growth across the size range. Neck. Neck circumference and neck level required the greatest number of alterations. The ASTM D 5585-95 Standard, participant neck circumferences, the traditionally graded patterns, and fit-to-shape patterns were evaluated quantitatively (Table III). Traditional grade rules result in a pattern that is larger than the growth represented by ASTM and participant neck circumferences were smaller, even though they matched the bust, waist, and hip circumferences.
Optimized pattern grading
87
IJCST 20,2
88
Table III. Neck measurement comparison
The size 6 participant required a 1.00 in. back neck intake. Excess neck circumference was a consistent problem in the larger sizes. The size 14-20 participants required a total circumference reduction ranging from 1.00 to 1.75 in. Excess front neck circumference was transferred to the side bust dart, while back neck excess was transferred to the shoulder dart. Body scans on transparency sheets were analyzed for shape differences from the size 8 participant that contribute to neck adjustments for 2D fit-to-shape patterns. The size 10 participant required a lower center front level due to her sharply angled neck and lower neckline. The size 16 and 20 participants required higher center front neck levels. Comparison of the side view neck shapes to the size 8 yielded little insight as to why these adjustments were required. It may be related to their curved upper backs. However, the size 16 participant had a protracted neck and shoulders, while the size 18 participant had a neck angle similar to the size 8. There were no adjustments to back neck level. We found no consistent trend in neck level adjustments. Changes may be due to individual posture that would not be accommodated across a standard size range. Armscye circumference and depth measurements/armhole grade (depth and width). The armscye circumference typically measured around the shoulder joint and under the arm point. This circumference is difficult to relate directly to the horizontal and vertical pattern grade. So the on-the-body circumferences measurement is abstracted to one vertical and one horizontal measurement to define the size of the armscye. The armscye depth is a vertical measurement with a standard grade. The armhole width is graded variably. The armscye required alterations on four of the seven patterns. Full size print-outs of the traditional and fit-to-shape nests were compared. There were no differences between the two sets at the back armscye. The sizes 6-14 patterns required slight adjustments to the front armscye. However, excess fabric in the front armscye was a consistent problem for the size 16-20 participants, and chest width was too wide at armscye level in the traditional pattern. Body scans on transparency sheets were analyzed for shape differences to the size 8 participant that may have contributed to excess fabric at the armscye. Bust shape was examined because breast tissue extends under the arm and thus impacts fit. The size 16 participant had a fuller, rounder bust and a lower bust apex. The size 18 participant had a fuller, rounder bust, lower bust apex, and little definition between bust and waist. She also had a rounded upper back. The size 20 participant had a full, round bust and slightly lower bust apex. A fuller bust requires a larger dart intake, which was the result when excess from the armscye was transferred to the bust dart. The length, width, and depth of the front armscye for traditional and fit-to-shape patterns were measured and compared. For sizes 6-14, the grade was consistent and there was little difference in length. The armscye length shortens slightly at size 14, but then follows a regular interval of growth. A dramatic difference was seen in the
ASTM neck base measurement Participant neck base measurement Neck circumference traditional patterns Neck circumference fit-to-shape patterns
Size 6 (in.)
Size 20 (in.)
Range (in.)
14.25 14.63 15.64 14.34
16.88 16.50 18.82 16.50
2.63 1.88 3.18 2.16
size 16-20 patterns. The fit-to-shape patterns were each 1.00 in. shorter in length, 2.00 in. shorter in depth, and 0.63 in. narrower in width. This resulted in a more curved armscye on the fit-to-shape patterns (Table IV). In research by Schofield and LaBat (2005b), the difference between the largest and smallest sizes for back armscye length for test and current patterns was more than 2.00 in. They did not compare the front armscye length due to missing data. However, findings from this study and Schofield and LaBat’s (2005b) study indicate that traditional grade rules result in a poor fitting armscye for larger size women. Bust point to bust point and neck to bust point measurements/dart grade. In traditional grading, darts increase in length and the dart position follows the crossgrade interval, based on the bust point to bust point and neck to bust point measurements from ASTM. Qualitative visual analysis of 2D pattern nests showed that no changes were needed in horizontal or vertical placement of the darts. Dart intake. There is no ASTM body measurement related to dart intake and in traditional grading, dart intake remains constant. Price and Zamkoff (1996, p. 40) suggest the following:
Optimized pattern grading
89
The width of the dart – known as the dart pick-up – remains the same from size to size, because the areas that shape the dart at its widest and narrowest ends grow in proportion to each other.
The fit-to-shape patterns for larger size participants had an increased front bust dart intake, and demonstrates that foreshortening of the dart is not enough to accommodate a fuller bust. The participants’ bust circumferences met the ASTM standard, however, bust shape varied. Increase in bust circumference in the larger sizes is due to an increased amount of tissue in the breast rather than underlying skeletal structure (Schofield and LaBat, 2005b). Conclusions Traditional industry grading practices do not provide good fit. This study evaluates the fit of a graded sheath dress on participants that represent the range of Misses sizes 6-20, based on height and circumference. Body differences across the size range is documented. To achieve optimum grading, the results indicate that grading practices should include measurement and shape variations. This study used a multi-method approach, combining visual and quantitative analysis of 2D and 3D data allowing a holistic interpretation. In the future, these techniques may be used to investigate body measurements and shapes within and between target markets. The greatest variation between traditionally graded and fit-to-shape patterns occurred between sizes 14 and 16. Size 6, 10, and 12 patterns required few changes, as they were closer to the size 8 sample size. This supports findings by Loker et al. (2005) who found that the greatest number of acceptable ratings were for Misses pants closest to the sample size.
Front side bust dart intake Front neck circumference on one-fourth pattern
14 (in.)
16 (in.)
18 (in.)
20 (in.)
þ 0.38 2 0.38
þ2.38 20.25
þ 2.63 20.38
þ 2.38 2 0.38
Table IV. Adjustments from graded pattern to fit-to-shape pattern for sizes 14-20
IJCST 20,2
90
From size 16 up, the neck grade in the traditional patterns resulted in a circumference that was too large. A similar conclusion was found in Schofield and LaBat’s (2005b) research where the neck of their test patterns had smaller differences than the pattern based on current grade rules. The results indicate neck grade length and width should be reduced, especially for size 16-20 patterns. Armscye shape and length in the fit-to-shape patterns changed dramatically from size 16 up. An increased bust dart intake and a strongly curved armscye were required to fit a fuller, more rounded bust. Cup size for the Misses size range is not indicated by ASTM, but should be considered as a major factor in new grading methods. The results suggest the need for more than one fit model in a range of sizes to achieve optimum grading results. The sample size should be the base size for no more than two sizes up or down. For sizes 16-20, a size 18 fit model would be optimum. The optimum fit model for sizes 6-14 is a size 10. These results extend the finding that visual details of a graded garment more than two sizes away from the fit model are noticeably different (Bye and Delong, 1994). More knowledge about body measurements and shape variations may lead to redefined size groupings. Methods of incorporating body shape into grading practice should be developed to optimize fit of mass produced apparel. References Armstrong, H.J. (2000a), Draping for Apparel Design, Fairchild Publications, New York, NY. Armstrong, H.J. (2000b), Patternmaking for Fashion Design, Prentice-Hall, Upper Saddle River, NJ. Barbaro, M. (2006), “Clothes that fit the women, not the store”, New York Times, March 31, pp. 1 & 6. Bond, P. (2004), “Sizing it up: new data could revamp sizing standards, but is the industry ready to get out its measuring tape?”, Women’s Wear Daily, March 17. Bye, E.K. (1990), “A visual sensory evaluation of two pattern grading methods”, unpublished doctoral dissertation, University of Minnesota, St Paul, MN. Bye, E.K. and Delong, M.R. (1994), “A visual sensory evaluation of the results of two pattern grading methods”, Clothing and Textiles Research Journal, Vol. 12, pp. 1-7. Bye, E., LaBat, K.L. and DeLong, M.R. (2006), “Analysis of body measurement systems for apparel”, Clothing & Textiles Research Journal, Vol. 24, pp. 66-79. Chowdhary, U. and Beale, N.V. (1988), “Plus-size women’s clothing interest, satisfactions and dissatisfactions with ready-to-wear apparel”, Perceptual and Motor Skills, Vol. 66, pp. 783-8. Cooklin, G. (1990), Pattern Grading for Women’s Clothes, Blackwell Scientific Publications, Oxford. D 5585-95 Standard (2001), Annual Book of ASTM Standards. Section 7, Textiles, American Society for Testing and Materials, Philadelphia, PA, pp. 777-80, D 5585-95 standard table of body measurements for adult female Misses figure type, sizes 2-20. Gazzuolo, E.B. (1985), “A theoretical framework for describing body form variation relative to pattern shapes”, unpublished master’s thesis, University of Minnesota, St Paul, MN. Istook, C.L. and Hwang, S-J. (2001), “3D body scanning systems with application to the apparel industry”, Journal of Fashion Marketing & Management, Vol. 5 No. 2, pp. 120-32. Kunick, P. (1967), Sizing, Pattern Construction, and Grading for Women’s and Children’s Garments: A Treatise and Standard Textbook for All Who are Engaged in the Production and Distribution of Women’s and Children’s Garments, Philip Kunick Ltd, London.
Kurt Salmon Associates (2004), From Mindshare to Market Share: Using Solution Selling to Drive Business, Kurt Salmon Associates, New York, NY. LaBat, K.L. (1987), “Consumer satisfaction/dissatisfaction with the fit of ready-to-wear clothing”, unpublished doctoral dissertation, University of Minnesota, St Paul, MN. Loker, S., Ashdown, S. and Schoenfelder, K. (2005), “Size-specific analysis of body scan data to improve apparel fit”, Journal of Textile and Apparel, Technology and Management, Vol. 4, pp. 1-15. O’Brien, R. and Shelton, W.C. (1941), Women’s Measurements for Garment and Pattern Construction, Misc. Pub. 454 Textiles and Clothing Division of Bureau of Home Economics, US Department of Agriculture in Coop with Work Projects Administration, US Government Printing Office, Washington, DC. Price, J. and Zamkoff, B. (1974), Grading Techniques for Modern Design, 1st ed., Fairchild Publications, Inc, New York, NY. Price, J. and Zamkoff, B. (1996), Grading Techniques for Modern Design, 2nd ed., Fairchild Publications, New York, NY. Rasband, J. and Liechty, E.G. (2006), Fabulous Fit: Speed Fitting and Alteration, Fairchild Publications, Inc, New York, NY. Schofield, N.A. (2000), “Investigation of the pattern grading assumptions used in the sizing of U.S. women’s clothing for the upper torso”, unpublished doctoral dissertation, University of Minnesota, St Paul, MN. Schofield, N.A. and LaBat, K.L. (2005a), “Defining and testing the assumption used in current apparel grading practice”, Clothing and Textiles Research Journal, Vol. 23, pp. 135-50. Schofield, N.A. and LaBat, K.L. (2005b), “Exploring the relationships of grading, sizing, and anthropometric data”, Clothing and Textiles Research Journal, Vol. 23, pp. 13-27. US Department of Commerce (1958), Commercial Standard: Voluntary Product Standard, CS 215-58: Body Measurements for the Sizing of Women’s Patterns and Apparel, Department of Commerce, Washington, DC. US Department of Commerce (1970), NBA (National Bureau of Standards) Voluntary Product Standard, PS 42-70: Body Measurements for the Sizing of Women’s Patterns and Apparel, US Department of Commerce, Washington, DC. Further reading Cooklin, G. (1995), Master Patterns and Grading for Women’s Outsizes: Pattern Sizing Technology, Blackwell Science Inc, Oxford. The US National Size Survey (2006), [TC]2, NC, available at: www.sizeusa.com/ (accessed October 17, 2006). About the authors Elizabeth Bye is an Associate Professor in the Clothing Design program in the College of Design at the University of Minnesota, Minneapolis. She received the BS degree in Textile Science and the MS degree in Apparel Design from Virginia Tech in 1982 and 1984, respectively. She completed the PhD degree in Design, Housing, and Apparel at the University of Minnesota in 1990. She worked in the apparel industry as a systems engineer for a major CAD company and as a technical designer for a uniform company. Her professional interests include apparel technology, product development, sizing and fit, visual analysis, and design education. Elizabeth Bye is the corresponding author and can be contacted at:
[email protected]
Optimized pattern grading
91
IJCST 20,2
92
Karen LaBat, PhD is a Professor in the Department of Design, Housing, and Apparel, at the University of Minnesota. She has held academic positions at the Minneapolis College of Art and Design and the University of Wisconsin-Stout. She’s held positions in industry with the Industrial Fabrics Association International and with a major sportswear producer. Her research program at the university focuses on clothing design for human health and safety with research projects often sponsored by industry. A sample of current research topics includes sizing for women 55 and older, body form and shape assessment using full body scans, and a longitudinal study of adolescents’ sun protective behaviors with implications for sun protective apparel designs. She is a member of the federally sponsored research committee on “Mediating Exposure to Environmental Hazards Through Textile Systems.” Ellen McKinney, PhD is a graduate of the Design, Housing, and Apparel program at the University of Minnesota. She received a MA in Fashion Design from Texas Woman’s University and a BS in Fashion Design from Texas Christian University. She has held academic positions at the University of Minnesota, the Art Institute of Dallas, Texas Woman’s University, Texas Christian University, and Tarrant County College. Industry positions have been as a women’s magazine fashion editor, and with a women’s specialty suit producer. Her dissertation research focused on the application of data from 3D body scans to pant sloper drafting. Dong-Eun Kim is a PhD student with apparel emphasis in the Department of Design, Housing, and Apparel, at the University of Minnesota. She received Bachelor of Home Economics from Department of Clothing and Textiles at Ewha Womans University in Korea. She received Master of Science from Fashion Design program in the College of Media Arts and Design at Drexel University, Philadelphia. She worked as Technical Designer at the G-Unit, Marc Ecko Enterprises and Fubu the Collection, New York.
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Improvement of hemoglobin washability of grafted nylon-6
Hemoglobin washability of grafted nylon-6
Soo-Min Oh Research Institute for Applied Science and Technology, Sogang Univeristy, Seoul, South Korea, and
In-Young Kim and Wha-Soon Song Sookmyung Women’s University, Seoul, South Korea
93 Received 5 August 2007 Revised 13 October 2007 Accepted 13 October 2007
Abstract Purpose – To explore the effect of acrylic acid polymerization and NaOH treatment of nylon-6 on hemoglobin washability. Design/methodology/approach – The nylon-6 was chemically grafted with acrylic acid and treated with NaOH for the purpose to improve the washability of hemoglobin as a blood protein soil. The structural change before and after graft polymerization was analyzed by X-ray photoelectron spectroscopy and scanning electron microscopy. The moisture regain, the contact angle, and the washability were each measured. Findings – Graft polymerization and NaOH treatment of nylon-6 changed the surface energy and structure of nylon-6 causing the washability of hemoglobin to improve. Compared to ungrafted nylon-6, the hydrophilic properties were increased remarkable by graft polymerization and NaOH treatment, which reulted in the improvement of washability. Practical implications – Hemoglobin is one of the most difficult soils to remove from the fabric. The paper might be of interest to those who would consider purchasing fabrics that are good at both hydrophilic properties and washability. Originality/value – The study on washability of hemoglobin as a blood protein soil for grafted fabric has not been reported so far. The results of this research may be used in a basic research for the development of new process which is capable of improving of hemoglobin washability. Keywords Fabric testing, Clothing Paper type Research paper
1. Introduction In recent years, the demand for nylon-6 as clothing material has increased because of its excellent mechanical properties such as elongation, tenacity and flexibility. However, great mechanical properties, nylon-6 is not suitable for clothing material which requires higher wettability because its moisture regain, anti-static characteristics, and soil removal are poor compared with cellulose fabrics. The graft polymerization method that grafts hydrophilic monomer on to the surface of nylon-6 has been investigated in order to improve the wettability of nylon-6 (El-Naggar et al., 1995; Bogoeva-Gaceva et al., 1993; Trivedi and Mehta, 1975). The surface modification of nylon-6 after being grafted by chemical treatment, therefore, would improve washability because soil deposition and soil removal on the surface of fiber is highly correlated to wettability of that fiber (Powe, 1972, p. 46). The graft polymerization is initiated by chemical method, by radiation or by plasma. Among them, the chemically initiated graft polymerization method has been
International Journal of Clothing Science and Technology Vol. 20 No. 2, 2008 pp. 93-103 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810850478
IJCST 20,2
94
reported to be not only an effective but also a simple method because it does not need any complicated equipment. Proteins are soluble in water and are best removed with water. However, over time they are denatured by oxidation as time goes by and become insoluble in water. If water solubility of proteins becomes low by denaturing it is very difficult to remove them using classical techniques. It has been reported that graft polymerization increases the wettability, dyeability, thermal resistance and mechanical properties of nylon ( El-Naggar et al., 1995; Bogoeva-Gaceva et al., 1993; Trivedi and Mehta, 1975), but reports on the washability of grafted nylon are few. Especially, study on washability of hemoglobin as a blood protein soil for grafted fabric, which is the one of difficult soil to remove on the fabric, has not been reported so far. In this study, nylon-6 was chemically grafted with acrylic acid (AA) as a hydrophilic monomer and then it was treated with NaOH, with the expectation that the hydrophilic property of the substrate would be changed by these processes, resulting in the improvement of the washability of hemoglobin as a blood protein soil. The change before and after graft polymerization was analyzed by X-ray photoelectron spectroscopy (XPS) as well as scanning electron microscopy (SEM). The changes of the hydrophilic properties such as moisture regain and contact angle were examined and the washability change of hemoglobin was also investigated. As a consequence, the effect of graft polymerization and NaOH treatment of nylon-6 and the subsequent improvement of washability were discussed. 2. Materials and methods 2.1 Materials Nylon-6 fabric supplied by Korea Apparel Testing & Research Institute was used in this study. Nylon-6 fabric of 60 g/m2 and 70 £ 70/ in. (warp £ weft) yarn density was used as the fabric sample, which was scoured for 1 h at 1008C in 0.1 N aqueous sodium carbonate (liquor ratio 30:1). Monomer and AA of 99 percent purity ( Junsei Chemical Co.) were used as received. All other chemicals were of pure grade and were used without further purification. 2.2 Methods 2.2.1 Grafting procedure. In the grafting procedure AA was used as the reactive chemical and ammonium persulfate (APS) was used as an initiator for the graft polymerization. A flask with three holes, which was equipped with a reflux refrigerator, nitrogen inflow tube and thermometer was used as the polymerization apparatus. This flask was submerged in an oil bath in order to keep the temperature constant during the reaction. AA concentrations were adjusted to 5, 10, 15, 20 and then 25 percent, respectively, and the APS concentration remained fixed at 0.05 percent. However, the temperature was increased to 908C and maintained for 1 h under nitrogen bubbling, because as it was reported by Ohguchi et al. (1980), the temperature needed to be higher than 808C in order to sufficiently expand initiator and monomer and the radical disjoint in the graft-polymerization process. The injecting nitrogen gas was used for removing oxygen inside the flask. The ungrafted nylon-6 (ungrafted nylon) of 10 £ 10 cm was dipped into the aqueous solution, containing AA, APS and distilled water for 30 min. The graft polymerization process was initiated when heat was added to the aqueous solution.
After finishing the graft polymerization, the AA grafted nylon-6 (nylon-g-AA) was extracted with a large amount of boiling distilled water for 2 h to remove the homopolymer, which was incidentally formed by a side reaction and the unreacted monomer. Afterwards the sample was dried. The graft ratio was determined as the percentage increase in weight over the original weight of the sample, where the weights of the sample were measured at the absolute dry condition. The grafted and NaOH treated nylon-6 fabric (nylon-g-NaAA) was attained by treating nylon-g-AA in NaOH 3 percent solution for 60 min at 208C. 2.2.2 SEM analysis. The morphology of the grafted fiber was observed by SEM. The SEM micrographs were taken with a JSM 8401 instrument by Jeol. 2.2.3 XPS analysis. The chemical analysis was studied by XPS. C 1s, O 1s levels of grafted nylon and C 1s, O 1s and Na 1s levels of base-treated nylon were studied using an XPS spectrometer (SSI 2803-S) with Al/Mg source at the condition of 12 KV and 20 mA. The XPS spectrum measured was standardized by adjusting C 1s peak to 284.7 eV. The contributions, due to the various chemical groups in the spectra were obtained by deconvoluting the peak envelope into elementary peaks. A fitting technique at known binding energies was developed using a computer program. 2.2.4 Moisture regain measurement. The moisture regain was evaluated by following the JIS L 1096 (1990) method. 2.2.5 Contact angle measurement. The contact angle between the distilled water (0.05 cm3) as a measuring liquid and the fabric sample was directly measured using a contact angle meter (Erma, gonio-meter type). Measurements were repeated using at least ten drops each time. The standard deviation of measurement was less than 28C measurements were done at 258C. 2.2.6 Detergency measurement. Pig hemoglobin (Acumedia manufactures Inc.) was used as the blood protein soil in washing experiments. In order to make hemoglobin-soiled fabric, a hemoglobin solution of 400 ml at 4.0 percent (w/v) was uniformly added by spotting to the surface of 5 £ 10 cm sized ungrafted nylon and nylon-g-NaAA, which resulted in 0.32 mg/cm2 hemoglobin-soiled fabric. Afterward, this hemoglobin-soiled fabric was denatured by steam at 758 or 30 min to make the temperature-degenerated protein. Washing was carried out to study the effect of graft polymerization on the washability change in water and washing solution with LAS, respectively. The washing in water was carried out using the distilled water and shaking water bath. Conditions for the washing experiment included 100 ml of distilled water at a temperature of 25 ^ 28C, an agitating time of 120 min. The washing in LAS solution was carried out using sodium dodecylbenzenesulfonate (LAS) as a detergent, the distilled water and Terg-O-meter were used. Conditions for the detergent washing experiment included 500 ml of 0.1 percent detergent liquid, at a temperature of 208C, a stir speed of 100 ^ 2 rpm, and an agitating time of 20 min. The hemoglobin was extracted from the fabric before and after washing with a solution of 0.1 N NaOH 100 ml, temperature of 90 ^ 28C, and time of 120 min. The absorbance of Fe in 750 nm was measured by a UV-spectrometer (Shimadzu, UV-1201) to evaluate the amounts of the hemoglobin, following Tokoro’s report (Tokoro et al. (1984)) on the measuring process which was based on the copper-Folin method: DðpercentÞ ¼
As 2 Aw £ 100 As 2 Ao
ð1Þ
Hemoglobin washability of grafted nylon-6 95
IJCST 20,2
96
D, detergency; Ao, absorbance of the liquid which is extracted from original fabric; As, absorbance of the liquid which is extracted from soiled fabric before washing; Aw, absorbance of the liquid which is extracted from soiled fabric after washing. 3. Results and discussion 3.1 Effect of monomer concentration on graft polymerization Figure1 shows the AA concentration dependence on the graft ratio for nylon-6 when nylon-6 is grafted at 908C for 60 min. The effect of hydrophilic monomer concentration on the graft ratio was studied by varying the monomer within the concentration range of 5-25 percent. It is clear from Figure 1 that the graft ratio is proportional to monomer concentration with a near straight line from zero. The more the concentration of AA increases the more the swelling of nylon-6 accelerates. AA, therefore, can penetrate the surface and into interior of nylon-6. As a result, the graft ratio increases linearly because chemical bonding increases according to the increase of initiation rate and propagation rate at the active sites of the polyamide molecule. That is, the increase of concentration of AA monomer makes the graft ratio for nylon-6 increase by accelerating polymerization that results from the activation of exhaustion of first radical. 3.2 Morphological structure of grafted nylon-6 with AA The structural morphology of the grafted nylon-6 fabric with AA was investigated. Figure 2 shows the morphology of ungrafted nylon and nylon-g-NaAA as to graft ratio. It is clear that the AA polymer does not form a uniform coating on the nylon’s single fibres, as shown in the Figure 2(b)-(f). The surface of nylon-6 was, however, covered with the grafted layer; even though the surface was not coated uniformly. As a result, 30
Graft Ratio (%)
20
10
Figure 1. Effect of the concentration of AA on the graft ratio for nylon
0
0
5
10 15 20 Concentration of AA (%)
25
Notes: Condition of graft: concentration of ammonium persulfate 0.05 percent, temperature 90°, time 60 min
30
Hemoglobin washability of grafted nylon-6 97 (a) ungrafted
(b) 5% Nylon-g-NaAA
(c) 10% Nylon-g-NaAA
(d) 15% Nylon-g-NaAA
Figure 2. SEM micrographs of the surface morphology for nylon-g-AA as changing graft ratio (e) 20% Nylon-g-NaAA
(f) 25% Nylon-g-NaAA
compared with that of ungrafted nylon the surface of nylon-g-AA was rough. The visible granular polymeric material embedded on the nylon-6 fibrils makes the smooth texture of the fibers of ungrafted nylon-6 (Figure 2(a)) gradually rough as the graft ration increase. The amount of polyacrylic acid attached to the surface of the grafted nylon-6 likewise increased with graft ratio This indicates that, the change of the surface morphology resulted from the AA graft polymerization on the nylon-6. The surface structures and the atom binding statement of ungrafted nylon, nylon-g-AA and nylon-g-NaAA, respectively, were analyzed by XPS. With respect to the degree of surface sensitivity, XPS is able to detect thinner outer layer of the ˚ ) comparised with ATR characterization. The resulting spectra of C interface (40-50 A 1s, O 1s, and Na are shown in Figures 3-5. Figures 3-5 also show the deconvolution of the C 1s, O 1s peaks into each elements individual components. In case of grafted nylon (nylon-g-AA and nylon-g-NaAA), intensities and areas of C 1s peak in 290 eV and of O 1s peak in 531.6 eV increased compared with ungrafted nylon. These peaks appeared with OvCZO binding, which means that OvCZO binding was introduced to nylon-6 by graft polymerization. In the case of nylon-g-NaAA, on the other hand, a new peak was introduced with 1,071 eV. This peak was identified as Na 1s peak, which was
IJCST 20,2 C-C
shoulder
C-N C-O
98
C=O-O
Intensity (arbitrary unit)
(a)
C-C
C=O-O
shoulder
C-N C-O
(b)
shoulder
C-C C-N C=O-O
Figure 3. XPS spectra of C 1s lines of the specimen: (a) ungrafted nylon; (b) nylon-g-AA; (c) nylon-g-NaAA
C-O
(c) 292
288 284 Binding energy (eV)
280
276
reported to be located in between 1,070 and 1,073 eV (Chastain, 1992). It is suggested, therefore that the replacement of ZCOOH with ZCOONa by NaOH treatment made this peak appear. 3.3 Hydrophilic property Figure 6 shows the moisture regain of ungrafted nylon, nylon-g-AA, and nylon-g-NaAA. The moisture regain of nylon-g-AA increased slightly compared with that of ungrafted nylon. Moisture regain of nylon-g-NaAA, however, increased remarkably with the graft ratio. To get a high-hydrophilic property, a sample of ungrafted nylon was then taken and subjected to hydrosysis with 3 percent NaOH. If the graft ratio of nylon-g-NaAA which had high-hydrophilic property by NaOH treatment was higher than 10 percent, the moisture regain of nylon-g-NaAA was about 8 percent, same as the moisture regain of cotton. It is supposed that the hydrophilic properties of
Hemoglobin washability of grafted nylon-6
shoulder C-O
99
C=O-O
Intensity (arbitrary unit)
(a)
shoulder C-O C=O-O
(b)
C-O
shoulder
C=O-O
(c) 540
536
532 528 Binding energy (eV)
524
nylon-g-NaAA increased remarkably due to the replacement of ZCOOH with ZCOONa by NaOH treatment, as can be confirmed in XPS results. These results suggest that nylon-6 ought to be treated with NaOH after graft polymerization in order to achieve as high moisture regain as cotton. Figure 7 shows the contact angle change of nylon-g-NaAA for water, where the contact angle of nylon-g-NaAA decreases with the graft ratio. It is supposed that the difference of contact angle between ungrafted nylon and nylon-g-NaAA was due to a greater hydrophilicity increased by the nylon/water surface energy owing to graft polymerization and 3 percent NaOH treatment. As a result, the moisture regain of the nylon-6 improved and the contact angle decreased after graft polymerization and NaOH treatment. This suggests an
Figure 4. XPS spectra of O 1s lines of the specimen: (a) ungrafted nylon; (b) nylon-g-AA; (c) nylon-g-NaAA
IJCST 20,2
100
Intensity (arbitrary unit)
(a)
Figure 5. XPS spectra of Na 1s lines of the specimen: (a) ungrafted nylon; (b) nylon-g-AA; (c) nylon-g-NaAA
(b)
(c) 1,085
1,080
1,075 1,070 Binding energy (eV)
1,065
1,060
improvement in the hydrophilic properties, which are related to the washability. On the other hand, it has been reported that KOH (Hegazy et al., 1990) or Na2CO3 (Howard et al., 1969) as well as NaOH can be used in the solution of neutralization treatment after graft polymerization, and the hydrophilic property of the grafted substrate is improved remarkably by post-treatment in any case. 3.4 Washability Figure 8 shows the washability of hemoglobin on nylon-g-NaAA with graft ratio. The hemoglobin washability of nylon-g-NaAA was improves remarkably with the proportional increase of polymerization (the graft ratio) compared with that of ungrafted nylon. The sample of ungrafted nylon was taken and was subjected to the hydrosysis with 3 percent NaOH to improve washability because the hydrophilic
Hemoglobin washability of grafted nylon-6
10
Moisture Regain (%)
8
101 6
4
2
0
Nylon-g-AA Nylon-g-NaAA
0
5
10
15 20 Graft Ratio (%)
25
30
Figure 6. Effect of the graft ratio on the moisture regain for nylon-g-AA and nylon-g-NaAA
80
Contact Angle (degree)
60
40
20
0
0
5
10
15 20 Graft Ratio (%)
25
30
Figure 7. Effect of the graft ratio on the contact angle for nylon-g-NaAA
IJCST 20,2
Hemoglobin Removal (%)
102
100
80
60
40
20
0
Figure 8. Effect of graft ratio on the hemoglobin removal from AA grafted nylon
5
10
15
20
25
30
Graft Ratio (%) Notes: Washing condition: instrument Terg-O-Tometer, conc. of detergent 0.1percent LAS, revolution speed 100 rpm, washing temperature 25°C, washing time 20 min
properties of nylon-g-NaAA could be increased by NaOH treatment. Washibility decreases, however, when the graft ratio exceeded 15 percent. It is thus supposed that the improvements in washability resulted from the introduction of hydrophilic OZC ¼ O in the nylon-6 by graft polymerization and from the polarity increase of the substrate owing to the replacement of ZCOOH with ZCOONa by NaOH treatment. This suggests that, the property change of the substrate is closely related to moisture regain and contact angle changes, both of which resultes from the increase of anion ion value for zeta potential. If surface energy increases; that is, washability also increases. Therefore, the washability of hemoglobin was improved because the surface energy and structure of nylon-6 were changed by graft polymerization and NaOH treatment. Washability decreases, however, if graft ratio exceeded than 15 percent. It is thus assumed that the rough surface created by superabundant graft polymerization resulted in low washability. In other words, if the surface of the fabric is too rough, soil can attach easily to it, and it is difficult to remove the attached soil. 4. Conclusion To develop washability of the hemoglobin as a blood protein soil, nylon-6 was chemically grafted by AA and treated with NaOH. The correlation between structural change and hydrophilic property change of the substrate and its effect on washability were also investigated. The graft ratio increased linearly with AA concentration. The surface of nylon-g-AA became rougher compared with that of ungrafted nylon.
It was found that ZCOOH was introduced by graft polymerization, parts of which were replaced with ZCOONa by NaOH treatment. The moisture regain increased and contact angle decreased by graft polymerization and NaOH treatment, which meant an increase of hydrophilic property. It was found that nylon-6 achieves a hydrophilic property as high as cotton, if treated by NaOH after graft polymerization. As a result, the washability of nylon-g-NaAA adulterated with hemoglobin was improved remarkably. References Bogoeva-Gaceva, B., Pimonenko, N.Y. and Petrov, G. (1993), “Photo-induced acrylamide graft polymerization onto polyamide-6”, Textile Res. J., Vol. 63, pp. 51-7. Chastain, J. (1992), Handbook of X-Ray Photoelectron Spectroscopy, Perkin-Elmer Corporation, Eden Prairie, MN. El-Naggar, A.M., El-Salmawi, K., Ibraheim, S.M. and Zahran, A.H. (1995), “Direct radiation grafting of acrylic acid to nylon-6 fabric and the behaviour of the resulting graft copolymer”, Polym. Int., Vol. 38, pp. 125-39. Hegazy, E-A., Taher, N.H. and Ebraid, A.R. (1990), “Radiation-initiated graft copolymerization of individual monomer and comonomer onto polyethylene and polytetrafluoroethylene films”, J. Appl. Polym. Sci., Vol. 39, pp. 1029-43. Howard, G.J., Kim, S.R. and Peters, R.H. (1969), “Graft polymerization of methacrylic acid on nylon 6 film”, J. Soc. Dyers. Colour., Vol. 85, pp. 468-73. JIS L 1096 (1990), Testing Methods for Woven Fabrics. Ohguchi, M., Ikeda, K. and Yasumura, T. (1980), “One-step grafted onto polyethylene terephthalate using benzonyl peroxide”, SEN I-GAKKAISHI, Vol. 136, p. 435. Powe, W.C. (1972), Surfactant Sciences Series, Vol. 5, Marcel Dekker, New York, NY, p. 46. Tokoro, Y., Fujii, T. and Minagawa, M. (1984), “Studies on removal of blood protein stains from fabrics (Part 2)”, Jpn. Res. Assn. Textile End-Uses, Vol. 25, pp. 125-32. Trivedi, I.M. and Mehta, P.C. (1975), “Gamma ray-induced graft copolymerization of acrylamide and acrylic acid to nylon 6 fabric”, J. Appl. Polym. Sci., Vol. 19, pp. 1-14. Corresponding author Wha-Soon Song can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Hemoglobin washability of grafted nylon-6 103
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,2
104 Received 12 June 2007 Revised 11 September 2007 Accepted 11 September 2007
Modelling the woven fabric strength using artificial neural network and Taguchi methodologies Mithat Zeydan Department of Industrial Engineering, Erciyes University, Kayseri, Turkey Abstract Purpose – Jacquard woven fabrics are widely used in various sections of upholstery industry, where mattress cover is one of them. Strength of jacquard woven mattress fabric depends on several factors. The objective of this study is to model the multi-linear relationship between fibre, yarn and fabric parameters on the strength of fabric using artificial neural network (ANN) and Taguchi design of experiment (TDOE) methodologies. Design/methodology/approach – TDOE was applied to determine the optimum design values and the contribution of each parameter. Robustness (performance) of models is measured by root mean squared error (RMSE). These tools will enable the user to predict the fabric strength from number of given inputs. It also provides the knowledge related to the contribution of fibre, yarn and fabric parameters on fabric strength. Fabrics tested in this study made from different fibre types and max/min level for several fabric and yarn-related parameters. The models generated with TDOE and ANN methodologies were compared with the actual experimental data. Findings – It was found that ANN model gives better approximation with the minimum RMSE. Research limitations/implications – The data taken from factory are related with jacquard woven fabric. Practical implications – This study has many practical implications that brings up a general approach for textile industry. During manufacturing, waste or scrap ratio can be reduced and production planning become more efficient. Originality/value – Firstly, before starting manufacturing in factory, we can easily predict the strength of woven fabric using the defined factors. This makes the model usable at the planning stage of the fabric. Secondly, the contribution of factors affecting fabric strength was determined. The ANN model generated in this study helps the engineers of planning department at the company easy to plan the manufacturing of fabric with a good estimation of fabric strength before the production order. Keywords Fabric testing, Neural nets, Taguchi methods Paper type Research paper
International Journal of Clothing Science and Technology Vol. 20 No. 2, 2008 pp. 104-118 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810850487
This paper is based upon supported by Erciyes University Foundation. The author appreciates the support and understanding of Dr Sheik Meeran from the School of Management of Bath University and Dr Levent Onal from the Textile Engineering Department of Erciyes University. In addition, the author would like to thank Mr Orhan C¸alıs¸kan, General Manager of the BOYTEKS Corporation and his Plant Manager Mr Hakan Bagdas, for providing the required data.
Introduction The application of artificial neural network (ANN) to textiles is relatively a new issue which goes back to a few decades. ANN in textiles is associated with fibre-fault classification and detection, chemical processing, prediction of fabric properties and textile manufacturing (fibre, yarn and fabric) processes. A vast review for the application of ANN to textiles was summarized by Chattopadhyay and Guha (2003). ANN was applied to analyze as a predictive tool for clothing processes (Barrett et al., 1996; Liu and Zhang, 2004; Park and Ha, 2005). Park and Ha (2005) also used Taguchi method for optimizing sewing conditions to minimize seam puckering problem in woven garments. They chose sewing speed, stitch length, sewing thread tension and presser foot pressure as parameters in the analysis. Chen et al. (2005) compared and established physical, statistical and ANN models for predicting the fibre diameter of melt blown non-wovens from the processing parameters. The results showed that ANN is very suitable for modelling the process. Liasi et al. (1999) used Taguchi design of experiment (TDOE) method for finding optimum needle temperature by using factors affecting needle heating in sewing heavy materials such as upholstery fabrics. Majumdar et al. (2004) used ANN for predicting spinning quality index and fibre micronaire from the given fibre properties. Keshavaraj et al. (1996) considered ANN as a serious design tool used in determining air permeability of woven fabrics for airbags. Effect of yarn count, warp/weft counts, calendaring (only for polyester fabrics) and weave type on permeability of airbag fabrics was modelled. Pilling propensity of woollen knitted fabrics, which is a major physical problem for limiting the service life of a knitted fabric, has recently been modelled using ANN. With the help of ANN, they could model the multi-linear relations between the fibre (diameter, curvature, hauteur, bundle strength), yarn (count, unevenness, twist factor) and fabric (tightness) parameters to generate a predictive tool before manufacturing (Beltran et al., 2005). ANN methodology was also used for predicting strength of textile structures. Cheng and Adams (1995) determined the yarn strength using neural network. Ertugrul and Ucar (2000) used ANN to predict bursting strength of cotton plain knitted fabrics. Ogulata et al. (2006) predicted elongation and recovery test results of woven stretch fabric for warp and weft direction using different test points with regression and ANN models. However, any research has been conducted on the strength prediction of woven fabric from fibre, yarn and fabric parameters using ANN and TDOE approach so far. In this study, the effect of some fibre, yarn and fabric parameters on the strength of jacquard woven mattress fabric was analyzed using TDOE and ANN methodology. Initially, parameters are chosen from experimental design perspective and then fabric strength is modelled based on the given parameters with both TDOE and ANN approach. The performance criteria assessing suitable model for the two approaches is based on root mean square error (RMSE). Ishikawa cause-effect diagram There are several parameters affecting fabric strength in manufacturing. Operator, materials, environment and process have effects on the fabric strength which are schematized in Ishikawa cause-effect diagram of Figure 1. Among them, this study specifically emphasizes on factors associated with materials, which can be accepted as the major input.
Modelling the woven fabric strength 105
IJCST 20,2
Operator
Materials Experience
Fiber type Yarn parameters Fabric parameters
Skill Knowledge Motivation
106 Fabric Strength Air humidity Temperature
Figure 1. Ishikawa cause-effect diagram for strength of woven mattress fabric
Jacquard mechanism
Weaving m/c sencronization
Warping Environment
Process
The following parameters were identified as potentially important parameters affecting the strength of woven fabric for TDOE: . number of warp yarns at fabric width; . weft density; . weft yarn count; . fibre type of weft yarn; . warp density; . warp yarn count; and . fibre type of warp yarn. While determining parameters of this study, parameters related to weaving process have been considered rather than yarn-based parameters. The firm that the research was carried out purchases the yarn from its suppliers. But, it performs all weaving and treatment processes in its plant. Because of this reason, production process parameters of the firm were adjusted according to desired conditions during the sample production. Thus, parameters significantly affecting fabric strength such as yarn strength and twist could not be taken into consideration. Consequently, it was not possible to manufacture additional (extra) samples that is suitable for the aforementioned parameters. Taguchi design of experiment Design of experiment has been used as a powerful technique to improve the product and process of manufacturing systems since Sir Ronald Fisher determined the effect of factors affecting the final condition of crop in the area of agriculture in the early 1920s. TDOE has been applied many different areas, even in service systems (Taner and Anthony, 2006; Kumar et al., 1996). Quality control is divided into two stages as off-line and online quality control. While online quality control is concerned with production operations and the relations with the customer after shipment, off-line quality control is
concerned with both product and process design issues. Taguchi claims that product deficiencies should be prevented before existing in the process or after produced. Therefore, high-quality product comes into existence with a good system design (off-line) rather than online quality control system. The Taguchi method is a statistical off-line quality control method that aims to reduce minimum level of product or process variation at the design stage (Tong et al., 1997). In the early of 1950s, Dr Genichi Taguchi developed and used some statistical experimental design techniques in the manufacturing industry. In Taguchi design approach, there are three design stages, which are system, parameter and tolerance designs. In system design, scientific and engineering knowledge is used to determine the basic configuration of the product (process) by engineer. In the parameter design stage, the specific values for the system parameters are determined. Parameter design is a methodology between the system design and the tolerance design. Parameter design gives us the best values for the parameters of the system. The goal of parameter design is the identification of settings that minimize variation in the performance characteristic and adjust its mean to an ideal value. Tolerance design is used to determine the best tolerances for the parameters (Montgomery, 1997). Taguchi recommends that statistical experimental design methods are employed to assist in quality improvement, particularly during parameter design and tolerance design to minimize total product manufacturing and lifetime costs. In the Taguchi analysis, there are three types of quality characteristics with respect to the target design. These are “smaller is better” “nominal is better” “bigger is better”. Standard tables known as orthogonal arrays are used for the design of experiments in the Taguchi method. In our study, we adopt the “bigger is better” approach. S/N ratio (SNR) is a measure of the performance variability of products/processes in the presence of noise factors. SNR is a performance criteria, defined as the signal to noise ratio, in that, S stands for mean and that is called signal and also N stands for standard deviation and that is called noise The higher the SNR, the better the quality of product is. The idea is to maximize the SNR and thereby minimising the effect of random noise factors has significant impact on the process performance (Antony et al., 1999). SNR is formulized with the following equation: !1 0 n X i C B B i¼1 y2i C C B ð1Þ S=N ¼ 210 log B C n A @ where n is the number of repetitions for an experimental combination and yi is a performance value of the ith experiment. Detailed information about Taguchi method can be found in Taguchi (1987) and Phadke (1989). Materials and method Materials Jacquard woven fabrics were manufactured at Picanol rapier weaving machine equipped with Stauble electronic jacquard using industrial settings. Warp and weft yarns made from staple cellulosic (i.e. cotton) and continuous polymeric (i.e. polyester) fibres, where cellulosic fibres (CF) represent minimum value and polymeric fibre (PF)
Modelling the woven fabric strength 107
IJCST 20,2
108
represent maximum value. Continuous polyester yarns were texturized to make them bulkier and natural fibre-like handle. Warp and weft yarn counts are accepted as the yarn parameters. Number of warp yarn along fabric width, warp density and weft density are accepted as fabric parameters affecting fabric strength. Two levels are considered for each parameter namely levels 1 and 2, which are the possible maximum and minimum levels for manufacturing bedding woven fabrics at the factory. Factors considered in this study included the factor levels are tabulated in Table I. Fabric strength was tested at Titan fabric strength tester machine according to the ASTM D5035 testing method. Strip method (ASTM D5035) was adopted for the evaluation of breaking force of narrow fabrics in Titan fabric strength tester. Any fabric slippage from the tester jaws was recorded. Breaking force was recorded for evaluation. Five samples were tested at each group, which were 20 cm in length and 25 mm in raveled width. Gage length was set to 75 mm and none the samples were failed at or close to the grip region. Only warp-wise testing was performed. Testing machine was set for a loading rate of 300 mm/min. The results are given in Table II included the standard deviations in parenthesis. All tests were performed under standard atmospheric conditions (per cent 65 ^ 2 relative humidity and 20 ^ 28C temperature) and the samples were conditioned hours under such conditions for 24 before testing. TDOE analysis was performed with Minitab Version 14.0 software package that is a computer program designed to perform basic and advanced statistical functions. In addition to that, Qwiknet Version 2.3 software package was used to perform ANN analysis.
Levels Factors
Table I. Control factors and levels for the experimental design
A: Number of warp yarns at fabric width B: Weft density (weft/cm) C: Weft yarn count (denier) D: Fibre type of weft yarn E: Warp density (warp/cm) F: Warp yarn count (denier) G: Fibre type of warp yarn
Order
Table II. Factor sequence within groups
1 2 3 4 5 6 7 8
Parameters D
A
B
C
7,040 7,040 7,040 7,040 8,658 8,658 8,658 8,658
8 8 16 16 8 8 16 16
300 300 600 600 600 600 300 300
PF CF PF CF PF CF PF CF
E
F
G
33 38 33 38 38 33 38 33
150 354 354 150 150 354 354 150
PF CF CF PF CF PF PF CF
1
2
7,040 8 300 PF 33 150 PF
8,658 16 600 CF 38 354 CF
Average fabric strength (N/m) 1,026 1,313 1,057 1,350 1,148 1,161 1,669 1,117
(21.6) (32.7) (26.9) (32.7) (34.2) (38.0) (36.3) (24.9)
The experimental design In order to perform this work, working team was formed in the factory composing of a quality engineer and a manufacturing engineer in addition to the academic team. After determining the main target, following list of sequential steps together with the team members was decided: . conducting brainstorming sessions to identify main factors affecting fabric strength in process and its levels; . selecting an appropriate orthogonal array for Taguchi method; . collecting the data or use the data available on the past performance of the process; . performing ANOVA for identifying significant factors in the process; and . finding the optimal combination of factor levels to maximize bedding fabric strength.
Modelling the woven fabric strength 109
Identification of parameters (factors) and their levels The most important stage in the TDOE is the selection of control factors. By controlling these factors, standard deviation of the process can be reduced. Design parameters affecting strength of woven fabrics together with factory team were evaluated by determining minimum-maximum values of each factor and factor levels for statistical analysis. Two levels were chosen with seven degrees of freedom. L8 experimental design is appropriate because of seven factors predetermined in this study. If full factorial experimental design for L8 experiment is used, 27 ¼ 128 sample size should be preferred. However, analyzing the effects with such sample size is costly and time consuming, even though it is the best for factor characterization. That is why, Taguchi approach is applied in order to work with less sample size. Appropriate orthogonal array Selecting orthogonal matrix is the next stage of the methodology. Orthogonal matrix is used to specify the sample groups. Minitab 14.0 software package is employed for obtaining the orthogonal matrix given in Table III. This matrix enables us to determine the group specification during sample production. According to orthogonal matrix in Table III, experimental layout given in Table II was arranged. Results and discussions Test results included the sequence of factors within the groups are given in Table II. Test results were assessed to find main effects for SNR and means. SNR values as a performance measurement criteria are calculated in Table IV according to the bigger-the better. 1 1 1 1 2 2 2 2
1 1 2 2 1 1 2 2
1 1 2 2 2 2 1 1
1 2 1 2 1 2 1 2
1 2 1 2 2 1 2 1
1 2 2 1 1 2 2 1
1 2 2 1 2 1 1 2
Table III. Orthogonal matrix for sample production
IJCST 20,2
110
Table IV. Calculated SNR values
ANOVA and optimal parameter design values For both levels 1 and 2, it was obtained average SNR as shown in Table V. Main effects plotted for SNR is shown in Figure 2. It implied that the largest impact in process consists of E, F, G, B, respectively. Maximum strength levels configuration for SNR are A2B2C1D2E2F2G1, which means that A (8658), B (16), C (300/DN), D (PF), E (38), F (30/2 DN), G (PF). This result is consistent from the stand point of mechanics of textile structures. Since, test is performed along warp direction, warp yarn-related parameters should be at maximum level. Maximum level of weft density of fabric can be as a result of bias effect during test. Minimum level for weft yarn count also makes sense, because finer weft yarn leads less undulation along warp direction, which is desired for higher Young’s modulus and strength of fabric. In addition, PF considered in this study have higher Young’s Modulus and strength than that of CF. Pooled ANOVA was preferred for testing reality of given parameters as shown in Table VI, where parenthesis represents pooled parameters. The idea behind pooling is that any effect that is not statistically significant can be eliminated from the model and the model can be refitted. Eliminated terms belonging to sums of squares and degrees of freedom are added into the residual sum of squares and degrees of freedom. The table value of F statistic at 90 and 95 per cent confidence levels are 5.54 and 10.13, respectively. Among the factors considered in the pooled ANOVA for the SNR, warp density (factor E) is the only factor, which is statistically significant at 90 and 95 per cent confidence level, while weft density (factor B), warp yarn count (factor F) and fibre type of warp yarn (factor G) are only significant at 90 per cent confidence level. Factor E contributes almost 49 per cent to the probability. Factor effects are shown for the mean response in Table VII. Here, maximum strength level configuration for mean response is found as A2B2C1D2E2F2G1, which is similar to what found in Table V. Experimental run
SNR
1 2 3 4 5 6 7 8
60.21 62.35 60.47 62.60 61.19 61.28 64.44 60.95
Factors
Table V. Average SNR
A B C D E F G
Average SNR at level 1
Average SNR at level 2
Effect of the factor
61.41 61.27 62 61.58 60.73 61.24 62.14
61.97 62.12 61.39 61.80 62.65 62.14 61.25
0.56 0.86 20.61 0.22 1.92 0.90 20.89
Modelling the woven fabric strength
Main Effects Plot (data means) for SN ratios A
B
C
62.4 61.6
111
60.8 Mean of SN ratios
1
2
1
D
2
1
E
2 F
62.4 61.6 60.8 1
2
1
2
1
2
G 62.4 61.6 60.8 1
Figure 2. Main effects plot for the SNR
2
Signal-to-noise: Larger is better
Source of variation A B C D E F G Pooled error Total
Factors A B C D E F G
Degree of freedom
Sum of squares
Mean square
[1] 1 [1] [1] 1 1 1 3 7
0.62 1.46 0.73 0.09 7.34 1.61 1.59 1.44 13.46
0.62 1.46 0.73 0.09 7.34 1.61 1.59 0.48 1.92
F-ratio
P (per cent)
3.04
10.84
15.29 3.35 3.31 – –
54.50 12.01 11.83 10.77 100
Mean response at level 1
Mean response at level 2
Effect of the factor
1,187 1,162 1,281 1,225 1,090 1,160 1,302
1,274 1,298 1,179 1,235 1,370 1,300 1,159
87 136 2 102 10 280 140 2 143
Table VI. Pooled ANOVA for the SNR
Table VII. Mean response values and effects of factors
IJCST 20,2
112
ANOVA was performed to verify whether the main effects are true or not (Table VIII). Main effect plotted for mean response is in Figure 3. The table value of F statistic at 90, 95 and 99 per cent confidence levels are 2.75, 3.92 and 6.85, respectively. The tabled value of F statistic at 99 per cent confidence levels is 6.85. Number of warp yarns at fabric width (factor A), weft density (factor B) and warp density (factor E) of fabric, weft yarn count (factor C), warp yarn count (factor F) and fibre type of warp yarn (factor G) are found significant on fabric strength at 99 per cent confidence levels according to ANOVA for the mean response. Among them factor E is the most significant parameter which carries 49.8 per cent of probability alone. In the meantime, factor D is insignificant at 99 per cent confidence level.
Source of variation
Table VIII. ANOVA for the mean response
Degree of freedom
Sum of squares
Mean square
F-ratio
P (per cent)
1 1 1 1 1 1 1 112 119
228,377 556,922 313,652 3,152 2,347,802 585,902 611,327 65,260 4,712,393
228,377 556,922 313,652 3,152 2,347,802 585,902 611,327 583
391.94 955.80 538.29 5.41 4,029.33 1,005.53 1,049.17 – –
4.83 11.80 6.64 0.05 49.80 12.42 12.96 1.47 100
A B C D E F G Error Total
Main Effects Plot (data means) for Means A
B
C
1,300 1,200 1,100 Mean of Means
1
2
2
1
E
2 F
1,300 1,200 1,100 1
2 G
1,300
Figure 3. Main effects plot for the mean response
1
D
1,200 1,100 1
2
1
2
1
2
Confirmation test If the real production value under optimum conditions is within confidence interval value, it could be say that confirmation for the test value is realized. Table IX gives the predicted values deduced from TDOE methodology. The best value is 1,679.25. The predicted mean fabric strength at the 99 per cent confidence limits (intervals) is given by the following equation: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi y e · F a; 1; n ð2Þ CL99 per cent ¼ ^ Ne
Modelling the woven fabric strength 113
where, ne is error variances (MSE); a is the significance level (1-confidence level); Ne is the effective number of replications. For the present study, MSE ¼ 583, Ne ¼ 15, table value of F0.01,1,112 ¼ 6.85 and hence, approximately, the result obtained at the optimum condition varies within 1,679.25 ^ 17 (expected results at optimum) at the 99 per cent confidence level (CI99 per cent ). Since, the production goes on under the optimal conditions, three sample from results were taken for verification, whether the predicted mean response depending on the optimal combination of factor levels is within the confidence intervals or not. These values are 1,696, 1,689, 1,670. Because of this situation, it could be accepted that the predicted results are within the confidence limits. Therefore, this test shows that it is within acceptable limits. Another method for finding optimum strength (Yopt ) is given with the following equation: þ ðD2 2 T Þ Y opt ¼ T þ ðA2 2 T Þ þ ðB2 2 T Þ þ ðC 1 2 TÞ þ ðF 2 2 T Þ þ ðG1 2 TÞ þ ðE 2 2 TÞ
ð3Þ
where T is the average of all experiments. Therefore, Yopt is found as 1,679.25 ^ 17 which is similar to the one obtained from Minitab 14.0. System modelling approach with ANN ANN is a computational tool that has similar running nature like the neurons in the brain. In the ANN, each neuron has multiple inputs and a single output. The output of the neuron is given as follows: ! n X wi xit 2 u yt ¼ f ð4Þ i¼1
A
B
C
D
E
F
G
Predicted fabric strength (N/m)
1 1 1 1 2 2 2 2
1 1 2 2 1 1 2 2
1 1 1 1 1 1 1 1
1 2 2 2 2 2 2 2
1 2 1 2 2 1 2 1
1 2 2 1 1 2 2 1
1 2 2 1 2 1 1 2
1,026 1,313 1,169.5 1,452.25 1,260.5 1,263.25 1,679.25 1,117
Table IX. Predicted Taguchi values
IJCST 20,2
114
where t is the time, n is the number of inputs, yt is the output, xit is the ith input of tth time, wi is a weight, u is the bias of the neuron and f(z) is the activation function. The activation function can be linear or nonlinear. Weight wi takes the positive value in the case of excitation and takes the negative value in the case of inhibition. In an ANN, neurons can be organized in two forms: recurrent net and a feed-forward net. In a recurrent net, all the neurons are interconnected to each other. In a feed-forward net neurons are organized in layers: one input layer, hidden layers and one output layer. Data flow through input layer to output layer. The structure of ANN enables them to learn, approximate functions, and classify patterns. Such capabilities of ANN make it a powerful tool for modelling control systems. Mostly feed-forward nets with sigmoidal, signum or Gaussian activation function are used in order to model control systems. The weights in these nets are commonly updated using back-propagation learning algorithm. Defining an error as the difference between the desired output of the network and the actual output of the network, back-propagation learning algorithm minimizes the sum of the mean square error using a gradient search technique (Fukuda, 1992). Back propagation error function has the following form: Erri ¼
1X ðT i 2 Oi Þ2 2
ð5Þ
where Oi is the actual ith output, Ti is the desired ith output (teacher signal), Erri is the error (Ti 2 Oi). In this study, “feed forward with a single hidden layer perception” form of ANN is adopted, in that the form is made up of one layer and the data propagate forward through the network from input to output. The first layer is called the input layer and contains one node (or neuron) for each of the inputs in the training data file. The last layer is called the output layer and contains one neuron for each of the network outputs. For the ANN model of woven fabric, a back-propagation neural network with one hidden layer was used. It must be emphasized that a single hidden layer of neurons is often mentioned to be adequate in representing input-output relationship using a back-propagation learning algorithm. The number of epochs required in the training set, N, depends upon the accuracy desired in the predictions of the output values by the neural network. For a good generalization by the network, it is recommended that the number of epochs be determined based on the following criterion (Rai and Pitchumani, 1997): N.
W e
ð6Þ
Where W is the number of synaptic weights (connections) in the network, and e is the prediction error of the network. If values given below are taken into consideration, according to the equation (6), N must be greater than 4,000. The input variables to the network are number of warp yarns at fabric width, weft density, weft yarn count, fibre type of weft yarn, warp density, warp yarn count and fibre type of warp yarn. The output variable of the network is fabric strength. Between the input and output layers are an arbitrary number of hidden layers each containing an arbitrary number of neurons. Each neuron is connected to every other neuron in adjacent layers by a set of weights. The weights define as the “strength” of the flow of information from one layer to the next through the network. “Training” a neural network is simply the process of determining an appropriate set of weights so that the network accurately approximates the
input/output relationship of the training data. In a back-propagation algorithm the weights are updated as follows: Dwi ðtÞ ¼ 2h
›EðtÞ þ aDwi ðt 2 1Þ ›wi ðtÞ
ð7Þ
where h is the learning rate, a is the momentum, and E(t) is the error. Figure 4 shows the real back-propagation neural network of fabric strength (Su and Wu, 2001). In this study, 120 training patterns and ten testing patterns are used to train and test the network, respectively. Learning rates of 0.01, 0.05, 0.1, 0.15, 0.20 and 0.30 were used and each network was run for 100,000 epochs. The following neural network parameters were used for computation: . Learning rate of 0.3. . Number of output layer neurons is 1. . Momentum is 0. . Activation function (for input, hidden and output) is linear. . Error margin is 0.01. . Epochs is 100,000. . Hidden layer is 1. . Number of input layer neurons is 7. . Number of hidden layer neurons is 5.
Modelling the woven fabric strength 115
As a result of testing, the best result was obtained according to the input parameters above. Neural network test data and output results are shown in Figure 5. It was selected five hidden nodes by trial and error attempts as there is no commonly accepted method for determining the number of hidden nodes to use in a back propagation neural network model (Chiu et al., 1997). Factor A
Factor B
Factor C Hidden Layers
Factor D
Fabric Strength
Factor E
Factor F
Factor G Bias
Bias
Figure 4. The back-propagation neural network for fabric strength
IJCST 20,2
1,800.70
116
Output
1,440.56 1,080.42 720.28 360.14
Figure 5. Comparison of neural network test data targets and outputs
0.00
1
2
3
4
5 Pattern #
6
7
8
9
Testing Data Targets NH Outputs
Comparison of TDOE and ANN models According to both TDOE and ANN, it was predicted the values of fabric strength for five condition (order). It is given in Table X. When TDOE and ANN models for the strength of jacquard woven mattress fabric are compared for their RMSE values, it can be said that ANN model has better model the case than TDOE in Table XI. Conclusions Significant outputs of this study can be summarized in two. Initially, the generated design model enables to predict the strength of woven fabric with negligible deviation. This makes the model usable at the planning stage of the fabric. Secondly, the
Table X. Comparison of experimental results with the predicted values from ANN and TDOE
Table XI. Comparison of TDOE and ANN models in terms of RMSE values
Order
A
B
C
D
E
F
G
Experimental
Predicted ANN
Predicted TDOE
1 2 3 4 5 6 7 8 9 10
1 2 1 2 1 2 1 2 1 2
1 2 1 2 1 1 2 1 2 2
2 1 2 1 2 2 1 1 1 1
2 1 1 2 2 1 1 1 1 2
2 1 1 2 2 1 2 2 2 1
1 1 2 2 1 2 2 1 2 2
1 2 2 2 2 2 1 2 2 1
1,250 1,098 1,005 1,428 1,302 1,052 1,637 1,301 1,544 1,498
1,266.72 1,172.79 1,036.31 1,639.41 1,189.69 1,130.94 1,596.25 1,249.99 1,497.16 1,578.05
1,213.75 1,106.75 920.75 1,536.50 1,071 1,008 1,581.75 1,250.25 1,439 1,399.5
Models
RMSE
TDOE ANN
143.86 91.24
contribution of factors affecting fabric strength was determined. The ANN model generated in this study helps the engineers of planning department at the company easy to plan the manufacturing of fabric with a good estimation of fabric strength before the production order. In addition, they will have the knowledge of important parameters affecting fabric strength, when new fabric designs are created. If we use TDOE in this manufacturing process for making parameter design, it is clear that the strength value of jacquard woven bedding fabric can be improved. The strength of woven fabric improves approximately 1 per cent, when the optimum process conditions deduced from TDOE is applied during manufacturing stage. If the ANN model is used, fabric strength prediction can be obtained better than TDOE model. The best fabric average is 1,669 which is given in Table II. However, the best average value under optimal conditions is 1,685. In this case, improvement in fabric strength is approximately 1 per cent.
References Antony, J., Hughes, M. and Kaye, M. (1999), “Reducing manufacturing variability using experimental design technique: a case study”, Integrated Manufacturing Systems, Vol. 10 No. 3, pp. 162-9. Barrett, G.R., Clapp, T.G. and Titus, K.J. (1996), “On-line fabric classification technique using a wavelet-based neural network approach”, Textile Research Journal, Vol. 66 No. 8, pp. 521-8. Beltran, R., Wang, L. and Wang, X. (2005), “Predicting the pilling propensity of fabrics through artificial neural network modelling”, Textile Research Journal, Vol. 75 No. 7, pp. 557-61. Chattopadhyay, R. and Guha, A. (2003), “Artificial neural networks: applications to textile”, Textile Progress, Vol. 35 No. 1. Chen, T., Li, L. and Huang, X. (2005), “Predicting the fibre diameter of melt blown nonwovens: comparison of physical, statistical and artificial network models”, Modelling and Simulation in Materials Science and Engineering, Vol. 13, pp. 575-84. Cheng, L. and Adams, D.L. (1995), “Yarn strength prediction using neural networks”, Textile Research Journal, Vol. 65 No. 9, pp. 495-500. Chiu, C.C., Su, C.T., Yang, G.S., Huang, J-S., Chen, S-C. and Cheng, N-T. (1997), “Selection of optimal parameters in gas-assisted injection moulding using a neural network model and the Taguchi method”, International Journal of Quality Science, Vol. 2 No. 2, pp. 106-20. Ertugrul, S. and Ucar, N. (2000), “Predicting bursting strength of cotton plain knitted fabrics using intelligent techniques”, Textile Research Journal, Vol. 70 No. 10, pp. 845-51. Fukuda, T. (1992), “Theory and applications of neural networks for industrial control systems”, IEEE Transactions on Industrial Electronics, Vol. 39 No. 6, pp. 472-89. Keshavaraj, R., Tock, R.W. and Haycook, D. (1996), “Airbag fabric material modeling of nylon and polyester fabrics using a very simple neural network architecture”, Journal of Applied Polymer Science, Vol. 60, pp. 2329-38. Kumar, A., Motwani, J. and Otero, L. (1996), “An application of Taguchi’s robust experimental design technique to improve service performance”, International Journal of Quality & Reliability Management, Vol. 13 No. 4, pp. 85-98. Liasi, E., Du, R., Bujas-Dimitrejevic, J. and Liburdi, F. (1999), “An experimental study of needle heating in sewing heavy materials using infrared radiometry”, International Journal of Clothing Science and Technology, Vol. 11 No. 5, pp. 300-14.
Modelling the woven fabric strength 117
IJCST 20,2
118
Liu, K. and Zhang, W. (2004), “Prediction of the performance of the fabrics in garment manufacturing by artificial neural network”, Journal of Dong Hua University (English Edition), Vol. 21 No. 59, pp. 22-6. Majumdar, A., Majumdar, P.K. and Sarkar, B. (2004), “Selecting cotton bales by spinning consistency index and micronaire using artificial neural networks”, AUTEX Research Journal, Vol. 4 No. 1, pp. 1-8. Montgomery, D.G. (1997), Design and Analysis of Experiments, 4th ed., Wiley, New York, NY. Ogulata, S.N., Sahin, C., Ogulata, T.O. and Balci, O. (2006), “The prediction of elongation and recovery of woven bi-stretch fabric using artificial neural network and linear regression models”, Fıbres & Textiles in Eastern Europe, Vol. 14 No. 2(56), pp. 46-9. Park, C.K. and Ha, J.Y. (2005), “Process for optimizing sewing conditions to minimize seam pucker using the Taguchi method”, Textile Research Journal, Vol. 75 No. 3, pp. 245-52. Phadke, M.S. (1989), Quality Engineering Using Robust Design, Prentice-Hall International, Englewood Cliffs, NJ. Rai, N. and Pitchumani, R. (1997), “Rapid cure simulation using artificial neural networks”, Composites Part A: Applied Science and Manufacturing, Vol. 28, pp. 847-59. Su, C. and Wu, C. (2001), “Intelligent approach to determining optimal burn-in time and cost for electronic products”, International Journal of Quality & Reliability Management, Vol. 18, pp. 549-59. Taguchi, G. (1987), System of Experimental Design, Vol. 1/2, ASI, Dearborn, MI. Taner, T. and Anthony, J. (2006), “Applying Taguchi methods to healthcare”, Leadership in Health Services, Vol. 19 No. 1, pp. 26-35. Tong, L.I., Su, C.T. and Wang, C-H. (1997), “The optimization of multi-response problems in the Taguchi method”, International Journal of Quality & Reliability Management, Vol. 14 No. 4, pp. 367-80. Corresponding author Mithat Zeydan can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Influence of A study of the influence of different different clothing materials clothing materials on heat and moisture transmission through 119 clothing materials, evaluated using a sweating cylinder Damjana Celcar
Received 23 April 2007 Revised 20 August 2007 Accepted 20 August 2007
Mura European Fashion Design, Murska Sobota, Slovenia
Harriet Meinander SmartWearLab, Tampere University of Technology, Tampere, Finland, and
Jelka Gersˇak The Faculty of Mechanical Engineering, Institute of Engineering Materials and Design, University of Maribor, Maribor, Slovenia Abstract Purpose – The purpose of this study was to investigate the heat and moisture transmission through different types of textile materials or material combinations used for male business clothing. Design/methodology/approach – In this study, eight different material combinations, which simulate four-layer male business clothing system were tested using the sweating cylinder under two environmental conditions (108C/65% RH and 258C/65% RH), and at two sweating levels (100 and 200 gm2 2h2 1), in order to evaluate the heat and moisture transmission properties of material combinations. Findings – The results show how combinations of clothing materials that simulate male business clothing system influence on the dry and evaporative heat loss between the cylinder surface and two different environment conditions as well as to different sweating levels. Practical implications – The sweating cylinder can be used for measuring the heat and moisture transmission through clothing materials or material combinations in order to find out the best combination of textile materials, which simulate clothing system. Measured thermal comfort properties of material combinations evaluated with a sweating cylinder can provide valuable information for the textile and clothing industry by manufacturing/designing new textiles and clothing systems. Originality/value – The paper investigated the heat and moisture transmission through combinations of clothing materials that simulate male business clothing system. In the past few years, clothing materials containing microencapsulated phase-change materials (PCMs) have appeared in outdoor garments, particularly sportswear; therefore, we decided to investigate the combinations of standard used textile materials as well as of materials, containing PCMs, which simulate male business clothing system. Keywords Textiles, Clothing, Heat transfer, Moisture Paper type Research paper
This research was performed in SmartWearLab of the Tampere University of Technology. We would like to thank the University for performing this research in SmartWearLab with the aid of the Erasmus exchange programme.
International Journal of Clothing Science and Technology Vol. 20 No. 2, 2008 pp. 119-130 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810850496
IJCST 20,2
120
1. Introduction The heat and moisture transmission properties of clothing materials, which displaying ability to transport heat and moisture from the human body’s surface into the environment, are dominating determinants of clothing’s thermal comfort. The measuring values that reflect this ability to assess the heat exchange of the human body with the environment, and are related to human perceptions of comfort, are thermal resistance (Rc) and water vapour resistance (Re) of clothing materials. Thermal comfort properties of clothing materials have been of great interest for researchers, since they are among the major characteristics that determine clothing’s thermal comfort. It is thus possible to find papers in the literature focused on the thermal comfort properties of textile materials and clothing. Research work by Meinander (1985, 1988), Meinander and Richards (2005) and Meinander and Bro¨de (2006) focused on the measurements of heat and moisture transmission through winter workwear, cold and protective clothing materials, evaluated using a sweating cylinder. Several research work focused on the thermal insulation and thermal contact properties of fabrics and fabric assemblies determined by using the Alambeta instrument (Hes et al., 1996; Frydrych et al., 2002; Gunesoglu et al., 2005). Another experimental device for measuring heat absorption of textile materials in a transient state was described by Pac et al. (2001). A transient technique to measure the effective thermal conductivity of moist fabrics is developed by Schneider et al. They have shown that the effective thermal conductivity of moist materials is substantially higher than that for dry materials (Schneider et al., 1992). Three studies concerned with laboratory measurement of heat and water vapour transfer through clothing materials are presented in Gibson’s work. The results of three studies illustrate important factors to be considered when evaluating the thermal and moisture vapour transport properties of textile materials (Gibson, 1993). Experiments in a study of heat and moisture transfer in textile assemblies involve the steaming of various fabric beds by using a mathematical model (Le et al., 1995). The work presented by Rossi et al. (2004) describes the water vapour transfer and condensation effects in multilayer textile combinations studied with sweating arm. In this paper, we investigate the thermal properties, respectively, heat and moisture transmission through different combinations of clothing materials that simulate male business clothing system. We decided to investigate clothing materials that are used for male business clothing as well as textiles, containing phase-change materials (PCMs) that could be used as liner or outerwear materials for business clothing. PCMs, also called latent heat storage materials, are materials that are able to store, release or absorb thermal energy as latent heat whilst they oscillate between solid and liquid forms, giving off heat as they change to a solid state, and absorbing it as they return to a liquid state (Zhang, 2001). In the past few years, clothing materials containing microencapsulated PCMs have appeared in outdoor garments, particularly sportswear. There are no references about heat and moisture transmission through combinations of clothing materials used for male business clothing; therefore, we decided to investigate the combinations of conventional used textile materials as well as of materials, containing PCMs, which simulate male business clothing system. A sweating cylinder (Meinander, 1985), which produces heat and moisture in a way similar to the human body, was used in this study for measuring the simultaneous
transmission of heat and moisture through combinations of clothing materials that simulate four-layer male business clothing system. By using a sweating cylinder, it is possible to test different types of materials or material combinations in a climate chamber under different climate and sweating conditions.
2. Heat and moisture transmission through clothing materials Heat transfer through clothing materials is an important topic related to thermal comfort of clothing, which is the result of a balanced process of heat exchange between the human body, the clothing system and the environment. Heat is transmitted from the body into the environment by the respired air, and from the skin. The heat loss through respiration is fairly small (about 10 per cent) and is, thus, mostly ignored in clothing comfort studies (Meinander, 1985). The effect of the clothing and the air trapped both within it and around the body can be evaluated by the parameters thermal resistance or insulation Rc and water vapour resistance Re, which provide the possibility of estimating the influence of clothing on thermal balance within a certain environment. The thermal resistance and water vapour transmission (WVT) properties of textile materials are therefore the measuring values that are dominating determinants of thermal comfort of the wearer (Mecheels, 1998). There are several physical and physiological testing methods to estimate the thermal comfort properties of both clothing systems and textile materials. Physical tests use some kind of device to simulate the skin’s heat and/or water vapour production and can be carried out either on textile materials or on completed clothing systems. Most of them are concerned with only one property, either the resistance to dry heat loss or to WVT. The best-known standard test method (ISO 11092, 1993) is the sweating guarded hot plate instrument, which allows for measurements of simultaneous dry and evaporative heat loss through horizontally placed textile materials. A couple of other methods for testing heat and/or moisture transmission through textile materials exist and have been reported, e.g. sweating cylinder (Meinander, 1985), the Alambeta and Permetest instrument (Hes and Dolezal, 1989; Hes, 1999; Hes et al., 1996), Thermo Labo measuring system (Yoneda and Kawabata, 1983), Hot plate apparatus (ISO 5085-1, 1989), Gore cup method (Gohlke, 1980) and other methods. A sweating cylinder from Tampere University of Technology, Figure 1, which produces heat and moisture in a way similar to the human body, was used in this study for measuring the simultaneous transmission of heat and moisture through combinations of clothing materials that simulate clothing systems. In comparison to the standard measurements of thermal properties, which are performed at horizontally placed test specimen and with either a temperature or a water vapour pressure gradient across the specimen, with sweating cylinder it is possible to evaluate the thermal resistance and WVT of textile materials or combinations of textile materials under different environmental and sweating conditions. Measurements are made at three-dimensionally placed test specimen or combination of textile materials, which simulate clothing system in a shape of a human trunk (Meinander, 1985). The thermal resistance of the textile materials or combination of materials (Rc) is determined from the heat supplied to the cylinder and the temperature values, by (Meinander, 1985):
Influence of different clothing materials 121
IJCST 20,2
122
Figure 1. A sweating cylinder for measuring heat and moisture transmission through clothing materials
Rc ¼
ts 2 ta · A ð8C m2 W21 Þ H dry
ð1Þ
where Hdry is the heat supplied to the cylinder in the dry test, ts is the cylinder’s surface temperature which corresponds to the skin’s temperature (358C), ta is the ambient temperature (8C), and A is the surface area (0.29 m2 ). Taking into account that heat supply is partly used to evaporate the water, the corrected thermal resistance value is calculated by: Rc;corr ¼
ts 2 ta · A ð8C m2 W21 Þ H sw 2 H e
ð2Þ
where Hsw is the total heat supply in the sweat test and He is the evaporative heat loss, which is calculated by:
H e ¼ ðms 2 mc Þw ¼ me w ðWm22 Þ
ð3Þ
where ms is the water fed into the cylinder (g), mc is the condensed water in the textile materials (g), me is the evaporated water (g), and w is the specific heat of the evaporated water (0.684 Wg2 1 at 258C). The amount of evaporated water Me as a percentage of the water input, gives a value for the WVT of the tested material’s combination. The amount of evaporated water Me is calculated by (Meinander, 1985): me Me ¼ · 100 ð%Þ ð4Þ ms
Influence of different clothing materials 123
3. Experimental part In this investigation, eight different combinations of materials that simulate four-layer male business clothing system were tested on the sweating cylinder under two environmental conditions: 108C/65% RH and 258C/65% RH, and at two sweating levels: 100 and 200 gm2 2h2 1. In the first part of the investigation, complementary tests of material, thermal and WVT properties were carried out according to standardized test methods. 3.1 Investigation model and clothing materials The first part of the experimental work included the selection and investigation of different textile materials, which could be used in male business clothing and its properties. Table I shows a review of the selected materials and their description. The determination of basic material, thermal and WVT properties in steady-state conditions were carried out according to standardized test methods, as follows: . material’s thickness (h) according to ISO 5084 (1996) and mass per unit area of textile materials (W) according to ISO 3801 (1977); . air permeability (Q) of textile materials was determined according to ISO 9237 (1995); . thermal resistance (Rct) and thermal conductivity (l) of textile materials were determined with hot plate apparatus according to ISO 5085-1 (1989); and . WVT of textile materials was determined according to Gore cup method modified by Gore-Tex (Gohlke, 1980). Clothing Fabric system layer sample Underwear Shirt Liner
UN1 S1 L1 L2
Male suit
MS1 MS2 MS3 MS4
Fabric content 100% CO 78% CO, 22% PES 100% CV 1. layer: 100% CV, 2. layer: Outlast-Acryl with PCMs 100% WO 88% WO, 12% PA 98% WO, 2% EL 68% Outlast-Acryl with PCMs, 28% WO, 4% EL
Weight Thickness Air permeability W/gm2 2 h/mm Q/lm2 2 s2 1 221.0 85.0 76.0
1.59 0.21 0.11
618.0 322.0 596.0
93.0 179.0 206.0 189.0
0.21 0.51 0.49 0.49
151.0 323.5 75.2 223.0
168.0
0.49
277.0
Table I. Description of test materials
IJCST 20,2
124
The second part of the investigation included testing of different combinations of textile materials, which simulate male business clothing systems. Eight different combinations of materials were chosen for testing on the sweating cylinder under different environmental and sweating conditions, Table II. The tested combinations of materials, which simulate four-layer business clothing system, consisted of one underwear layer, one shirt layer, two different liner layers and four different outerwear-suit layers. All tests were done under two environmental conditions (108C/65% and 258C/65%) and at two sweating levels (100 and 200 gm2 2h) and two parallel tests were carried out for all combinations of materials. If the difference in results was . 5 per cent, a third test was done and the outer was ignored. The test procedure of sweating cylinder measurement was as follows: . Weighing of individual test materials and cylinder (dry weight). . Dressing the cylinder with materials and fastening of 6 PT 100 surface temperature sensors onto the each surface of the tested materials. The surface sensors measure the tested materials’ temperature. One sensor measures air temperature in the climate chamber. . Placing the cylinder on a precision scale in the climate chamber to assess the amount of condensed water. . Switching on heating. . Dry testing. Duration of the dry test was 1 h. . Sweating test. Duration of wet test was 2 h. . Switching off heating and sweating. . Undressing of the cylinder and immediate weighing of the individual test materials (wet weight). 3.2 Test method – sweating cylinder measurement A sweating cylinder, Figure 1, was constructed to measure the simultaneous transmission of heat and moisture trough clothing materials or material combinations. Measurements using the cylinder are made three dimensionally, and with this method, it is possible to test different types of materials or material combinations under different climate conditions, in the climate chamber. The basic idea is that the cylinder produces heat and moisture in a similar way to the human body. The cylinder construction and a cross-section of one sweat gland are shown in Figure 2.
Clothing layer
Fabric sample
Underwear Shirt Liner
UN1 S1 L1 L2 MS1 MS2 MS3 MS4
Male suit Table II. Tested material combinations
c1
c2
c3
* * *
* * *
* * *
Combination c4 c5 * * *
* *
c6
c7
c8
* *
* *
* *
* *
* *
*
*
*
* *
* *
*
Influence of different clothing materials 11
125
10
12
2 3 4 1
11
(a)
heat + water vapour
5 6 7
8 9
(b)
Notes: 1 - water supply, 2 - cylinder wall, 3 - heating wire, 4, 5, 6, 9 - protective layers, 7 - wicking layer, 8 - microporous layer, 10 - test area, 11 - edge with separate heating, 12 - sweat gland
The cylinder wall (2) is heated to the surface temperature (358C) corresponding to the skin temperature. A predetermined amount of liquid water (1) is supplied to the surface (to the sweat gland – 12), where it evaporates and leaves the cylinder as water vapour. Heat is lost from the cylinder’s surface through dry heat loss, as well as through evaporative heat loss. Measurements are made of the temperatures at different points and of the heat supply, which is required to keep the surface temperature at a predetermined level. The level of heat loss is reduced when the cylinder is dressed with the test materials, as textiles act as a thermal resistance between the heated cylinder surface and the colder environment. The water vapour is only partly transferred as vapour through textile materials, and partly condensed within them. The thermal resistance of the textile materials is defined from the heat input and the temperature values, and the WVT from the water input and absorption values (Meinander, 1985; Meinander and Bro¨de, 2006). The dimension of the cylinder’s measurement area (10) of the cylinder is 300 mm high and the diameter is 300 mm. The cylinder wall (2) is made of stiff foam plastic and is approximately 30 mm thick. To avoid heat flow upwards and downwards from the test area, the upper and lower edges (11) of the cylinder are heated to the same
Figure 2. The sweating cylinder (a) and cross-section; (b) of a sweat gland
IJCST 20,2
126
temperature, using separate heating systems. The outer surface of the wall is covered by an electric heating wire (3), which is protected by an insulating film (4). A metal layer (5) spreads the heat from the wire and a plastic layer (6) gives some mechanical protection. The cylinder is covered with a laminated material (7), the inner area of which spreads the water from the tubes to the larger areas. The outer area is an expanded micro porous PTFE-membrane (9), which transmits water vapour but not liquid water. The cylinder is dressed with the test material or combination of materials, and placed in the climate chamber on a balance, which records the weight changes during the test, thus the moisture evaporation can be recorded as the difference between the supplied water and the weight increase of the cylinder dressed with materials. The amount of water condensed in the textiles is determined by weighing the samples before and immediately after the test. Cylinder surface temperature (8C), sweating level (gm2 2h), test time (min), ambient temperature (8C) and relative humidity (per cent) are set for the test. During the test a computer program is used for the control and measurement of the cylinder’s surface temperature (8C), heat supply (Wm2 2), temperatures at different points (layers of materials, 8C), total weight increase during the test (g) and weight increase of the individual material layers (g). Based on the measured values the dry thermal resistance (Rc, m2KW2 1), the evaporative heat loss (He, Wm2 2), corrected thermal resistance (Rc,corr, m2KW2 1), WVT (Me, per cent), evaporated and condensed water (me and mc g) are calculated (Meinander, 1985; Meinander and Bro¨de, 2006). 4. Results and discussion The results obtained by investigating clothing materials’ influence on heat and moisture transmission through clothing materials are presented, as: . the results of the thermal and WVT properties of clothing materials, Table III; and . the results obtained by sweating cylinder measurements (the heat loss from the cylinder in dry and sweating test, the evaporated and condensed water, and WVT). From analysis of material’s thermal resistance (Rct) we can observe that thermal resistance of textile materials increase with their thickness proportionally. By comparing thermal resistance values of male suit fabrics it can be noted, that there exist very small differences between those values and that 100 per cent wool fabric has the highest values of thermal resistance, air permeability and WVT in compare to other male suit fabrics. From results it can also be seen that liner material L1 (100 per cent CV) has higher values of air permeability (Q) and WVT than liner material L2, which is laminated with acryl/PCMs. Fabric sample
Table III. Thermal and WVT properties of clothing materials
UN1 S1 L1 L2 MS1 MS2 MS3 MS4
Thermal resistance Rct/cm2w2 1
Thermal conductivity l/Wm2 1K2 1
Water vapour transmission WVT/gm2 2 24 h2 1
0.036 0.005 0.001 0.002 0.016 0.011 0.011 0.014
0.044 0.042 0.110 0.105 0.032 0.045 0.045 0.035
5256.0 5713.0 6008.0 5368.0 5695.0 5552.0 5648.0 5306.0
The results for measured heat loss values from the cylinder in the dry (Hdry) and the sweat test (Hsw and He) through combinations of clothing materials under two environmental conditions are shown in Figures 3 and 4. As can be seen from Figure 3, the heat loss values from the cylinder in the dry test at 108C is 55 to 65 per cent higher than at 258C. This means that the cylinder at an ambient temperature 108C needs much more heat to maintain an average surface temperature (358C) compared to the test at an ambient temperature of 258C. By comparing combinations c1-c4 with c5-c8, which have different and thicker liner-layers, it can be seen that heat loss values in combinations c5-c8 are a little lower do to their thicker liner-layer. Heat loss values (Hsw) from the cylinder in the sweat test at 108C and sweating level 100 g/m2h are 19-29 per cent higher, and at 200 g/m2h 35-45 per cent higher than in the dry test (Hdry). At ambient temperature 258C there are even higher differences between heat loss values in the dry and sweat tests. The differences are about 41-56 per cent in comparison to Hdry and Hsw at sweating level 100 g/m2h, and 57-69 per cent at sweating level 200 g/m2h. It can also be seen from Figure 4 that heat loss values increase with an increasing sweating level by 17-25 per cent at 108C and 25-33 per cent at 258C. If we compare heat loss values in the sweat test we can see that, with an increasing sweating level and/or lower ambient temperature heat loss values increase by 30-45 per cent at lower temperatures. By comparing evaporative heat loss values (He) it can be noted that, with an increasing sweating level, evaporative heat loss values increase by about 50 per cent and that with changes in ambient temperature, there are no big differences between He values. In the sweat test, the feed (ed.) water partly evaporates and partly condenses. The amount of moisture that has condensed (mc) in the materials and evaporated (me) through combinations of materials at two different ambient temperatures and two
Influence of different clothing materials 127
Dry heat loss 250.0 H (W/m2)
200.0 150.0 100.0
Figure 3. Dry heat loss from the cylinder through combinations of clothing materials at 10 and 258C ambient temperatures
50.0 0.0 c1
c2
c3
c4
c5
c6
c7
c8
Combination of materials Hdry/10°C/65%
Hdry/25°C/65%
Heat loss at 25°C/65%, sweat test
H (W/m2)
H (W/m2)
Heat loss at 10°C/65%, sweat test 350.0 300.0 250.0 200.0 150.0 100.0 50.0 0.0 c1
c2
c3 c4 c5 c6 c7 Combination of materials
Hsw/100
Hsw/200
c8
He/100 He/200
(a) at 10°C
350.0 300.0 250.0 200.0 150.0 100.0 50.0 0.0 c1
c2
c3 c4 c5 c6 c7 Combination of materials
Hsw/100
Hsw/200
c8
He/100 He/200
(b) at 25°C
Figure 4. Heat losses from the cylinder in the sweat test through combinations of clothing materials under two ambient temperatures
IJCST 20,2
128
sweating levels, are shown in Figure 5. Figure 6 shows values for WVT (Me) of the tested materials combination, which are given from the amount of evaporated water (me) as percentages of the water input (ms). The evaporated water increases with an increasing sweating level by 43-48 per cent while condensed water increases by 60-72 per cent. By comparing evaporated water at different temperatures and sweating levels, we noted that in most combinations the values of evaporated water at the lowest sweating level (100 g/m2h) are lower at 108C than at 258C, while the values of evaporated water at the highest sweating level (200 g/m2h) are a little higher at 108C than at 258C. The condensation of moisture is, at the lowest temperature (108C) and lowest sweating level (100 g/m2h), a little higher than at the highest temperature (258C), whilst at the highest sweating level (200 g/m2h), the amount of condensed moisture is higher at ambient temperature 258C that at 108C. The rate of evaporated water at sweating level 100 g/m2h is about 79-86 per cent and at sweating level 200 g/m2h is about 72-82 per cent. From Figure 6, we can see that, with an increasing sweating level, the amount of evaporated water as a percentage of the feed (ed.) water decreases by 3-10 per cent at 108C and by 7-13 per cent at 258C. We also noted that the values for WVT at sweating level 200 g/m2h are higher at 108C than at 258C, while at sweating level 100 g/m2h the WVT values are a little lower at 108C than at 258C. 5. Conclusions In this study, we used a sweating cylinder for measuring heat and moisture transmission through combinations of textile materials, which simulate four-layer male business clothing system. Different combinations of clothing materials were tested for this purpose under two ambient conditions and two sweating levels, in order to evaluate its thermal comfort (heat and moisture transmission) properties. 180.0 160.0 140.0 120.0 100.0 80.0 60.0
m (g/m2h)
Evaporated water at 100 and 200 g/m2h m (g/m2h)
Figure 5. Evaporated and condensed water through/in the combination of materials at two different temperatures, and two sweating levels (a) evaporated water; (b) condensed water
c1
c2
c3 c4 c5 c6 c7 Combination of materials
me-100/25°C/65% me-200/25°/65%
c8
60.0 50.0 40.0 30.0 20.0 10.0 0.0
c3 c4 c5 c6 c7 Combination of materials
c8
me-100/10°/65% me-200/10°/65%
Water vapour transmission at 10°C/65%, 100 and 200g/m2h 90.0
85.0
85.0 Me (%)
Me (%)
c2
me-100/25°C/65% me-200/25°/65%
90.0 80.0 75.0
80.0 75.0 70.0
70.0 65.0
c1
me-100/10°/65% me-200/10°/65%
Water vapour transmission at 25°C/65%, 100 and 200g/m2h
Figure 6. WVT values of tested materials’ combinations at two ambient temperatures, and two sweating levels (a) 258C/65%, 100 and 200 g/m2h; (b) 108C/65%, 100 and 200 g/m2h
Condensed water at 100 and 200 g/m2h
65.0 c1
c2
c3 c4 c5 c6 c7 Combination of materials Me-100
Me-200
c8
c1
c2
c3 c4 c5 c6 c7 Combination of materials Me-100
Me-200
c8
The results show that dry heat loss values increase with a decreasing ambient temperature and that heat loss values in the sweat test (Hsw and He) increase with an increasing sweating level and/or, lower ambient temperature. We noted that sweating always increases the heat loss, as the supplied water partly evaporates and partly condensates in the material layers and that the corrected thermal resistance of the materials combinations decreases with an increasing sweating level and increases with a decreasing ambient temperature. The amount of condensed and evaporated moisture increases with an increasing sweating level while the amount of evaporated water as a percentage of the feed (ed.) water (Me) decreases with an increasing sweating level. The tests using the sweating cylinder and standardized test methods show that very small differences exist between combinations of materials with different outer and liner layers, due to their small differences in basic properties. We also noted that when dry testing small differences in heat loss and thermal resistance values exist between combinations c1-c4 (combinations with CV liner) and c5-c8 (combinations with CV liner with PCM particles). One explanation of this result could be that these differences exist due to the differences in thickness and thermal resistance values of liner material. Another explanation could be that combinations of materials with PCM have higher and better thermal properties due to the content of PCM particles. Further research is, therefore, needed in order to confirm this tendency. In further research, we will evaluate male business clothing prototypes (made of tested materials) with the sweating manikin Coppelius and, furthermore, on human volunteers under artificially created climatic conditions in a climate chamber, where the impact of the clothing on the physiological parameters of the tested person will be investigated. Summarized, the results show how combinations of clothing materials that simulate male business clothing system influence on the dry and evaporative heat loss between the cylinder surface and the environment. Two different climate conditions (108C/65% RH and 258C/65% RH) and two sweating levels (100 and 200 gm2 2h) were chosen to analyse the heat and moisture transmission properties of material combinations. References Frydrych, I., Dziworska, G. and Bilska, J. (2002), “Comparative analysis of the thermal insulation properties of fabrics made of natural and man-made cellulose fibres”, Fibres & Textiles in Eastern Europe, October/December, pp. 40-4. Gibson, P.W. (1993), “Factors influencing steady-state heat and water-vapor transfer measurements for clothing materials”, Textile Research Journal, Vol. 63 No. 12, pp. 749-64. Gohlke, D.J. (1980), Improved Analysis of Comfort Performance in Coated Fabrics, W.L. Gore & Associates, Newark, DE, March. Gunesoglu, S., Meric, B. and Gunesoglu, C. (2005), “Thermal contact properties of 2-yarn fleece knitted fabrics”, Fibres & Textiles in Eastern Europe, Vol. 13 No. 2(50), pp. 46-50. Hes, L. (1999), “Optimisation of shirt fabrics’ composition from the point of view of their appearance and thermal comfort”, International Journal of Clothing Science & Technology, Vol. 11 Nos 2/3, pp. 105-19. Hes, L. and Dolezal, I. (1989), “New method and equipment for measuring thermal properties of textiles”, J. Text Mach. Soc. Jpn., Vol. 42 No. 8, pp. 124-8. Hes, L., deAraujo, M. and Djulay, V.V. (1996), “Effect of mutual bonding of textile layers on thermal insulation and thermal contact properties of fabric assemblies”, Textile Research Journal, Vol. 66 No. 4, pp. 245-50.
Influence of different clothing materials 129
IJCST 20,2
130
ISO 3801 (1977), Textiles-Determination of Mass per Unit Length and Mass per Unit Area, International Organization for Standardization, Gene`ve. ISO 5084 (1996), Textiles-Determination of Thickness of Textiles and Textile Products, International Organization for Standardization, Gene`ve. ISO 5085-1 (1989), Textiles-Determination of Thermal Resistance – Part 1: Low Thermal Resistance, International Organization for Standardization, Gene`ve. ISO 9237 (1995), Textiles-Determination of the Permeability of Fabrics to Air, International Organization for Standardization, Gene`ve. ISO 11092 (1993), Textiles-Physiological Effects-Measurement of Thermal and Water-Vapour Resistance Under Steady-State Conditions (sweating guarded-hot plate test), International Organization for Standardization, Gene`ve. Le, C.V., Ly, N.G. and Postle, R. (1995), “‘Heat and moisture transfer in textile assemblies’ part I: steaming of wool, cotton, nylon and polyester fabric beds”, Textile Research Journal, Vol. 65 No. 4, pp. 203-12. Mecheels, J. (1998), Ko¨rper-Klima-Kleidung. Wie funktioniert unsere Kleidung?, Shiele & Scho¨n, Berlin. Meinander, H. (1985), Introduction of a New Test Method for Measuring Heat and Moisture Transmission Trough Clothing Materials and its Application on Winter Workwear, VTT Publication 24, Espoo. Meinander, H. (1988), “Clothing physiological measurements on winter workwear materials with a sweating cylinder”, Journal of the Textile Institute, Vol. 79 No. 4, pp. 620-33. Meinander, H. and Bro¨de, P. (2006), “Effect of long wave radiation on the heat loss through protective clothing ensembles – material, manikin and human subject evaluation”, in Fan, J. (Ed.), Proceeding of 6th International Thermal Manikin and Modelling Meeting, Hong Kong, Thermprotect Network, 17-19 October, pp. 29-41. Meinander, H. and Richards, M. (2005), “Effects of moisture on the heat transfer through clothing materials”, in Holmer, I., Kuklane, K. and Gao, C. (Eds), Proceedings of the 11th International Conference on Environmental Ergonomics, Ystad, Sweden, Thermprotect network, 22-26 May, pp. 442-4. Pac, M.J., Bueno, M.A., Renner, M. and Kasmi, S.L. (2001), “Warm-cool feeling relative to tribological properties of fabrics”, Textile Research Journal, Vol. 71 No. 9, pp. 806-12. Rossi, R.M., Gross, R. and May, H. (2004), “Water vapor transfer and condensation effects in multilayer textile combinations”, Textile Research Journal, Vol. 74 No. 1, pp. 1-6. Schneider, A.M., Hoschke, B.N. and Goldsmid, H.J. (1992), “Heat transfer through moist fabrics”, Textile Research Journal, Vol. 62 No. 2, pp. 61-6. Yoneda, M. and Kawabata, S. (1983), “Analysis of transient heat conduction and its applications. Part II. J”, Text. Machinery Society of Japan, Vol. 31, pp. 73-81. Zhang, X. (2001), “Heat-storage and thermo-regulated textiles and clothing”, in Tao, X. (Ed.), Smart Fibres, Fabrics and Clothing, Woodhead Publishing Ltd, Cambridge, pp. 34-57. Corresponding author Damjana Celcar can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Pore network modelling using image processing techniques
Pore network modelling
Application to the nonwoven material M. Dimassi, L. Koehl, X. Zeng and A. Peruwelz
137
Gemtex Laboratory, Ensait Textile Institute, Roubaix, France Abstract Purpose – The knowledge of structural parameters of nonwovens media is poorly understood. The pores size distribution (PSD) function is one of those parameters. The difficulty is not only the understanding of the distribution of pores but also the identification of pores geometry distribution (PGD) and their behaviour concerning the dynamic fluid transportation. The purpose of this paper is to present an efficient and reliable method based on image analysis which on one hand, performs the estimation of the PSD function and takes into account the geometric aspect of pores, and on the other hand, analyses liquid wicking in very thin filter media. Design/methodology/approach – The proposed methods, in this paper, are applied on thin filter media made of polyester. The samples have not sudden any treatment. The authors set up an optical test bed in order to observe the dynamic properties of the samples. Dynamic raw data about the liquid wicking are extracted directly from video sequences using the appropriate test bed. The structural parameters are extracted from the non-wetted samples. Findings – Obtained results allow a better understanding of the liquid wicking in very thin filter media. In addition to the PSD function, the PGD function adds informations about the shape of pores. The dynamic data of the liquid wicking explains that pores have different behaviour when liquid reached them. It can be deduced from this study that the fluid transport in the pore network is defined by three main parameters: geometric parameter (size, shape), capillary action and pores connection in the network. Research limitations/implications – The led back-lighting system is not sufficient to observe precisely the liquid wicking. An additional front-lighting will be added in further studies. Originality/value – The extraction of dynamic properties from video sequences, by performing image analysis is an original method to characterise the porosity in thin media filter. Keywords Modelling, Textiles, Image processing, Porous materials Paper type Research paper
Introduction A porous media such as a nonwoven is the result of the entanglement of fibres/filament. About 90 per cent of the volumes of the fabric are voids which define the pore structures and significantly influence the functionalities of the fabrics (filtration threshold, wetting, capillarity liquid strike through time, and so on). Recently, the concept of pores size distribution (PSD) is widely used in nonwoven industry to extract the influence of structural properties on functional properties. It is also widely used in theoretical modeling or specific applications to infer hydraulic properties of nonwovens behaviour and to reach expected functional features. To extract information about the porous network, appropriate devices – like porometer – have been developed to provide indirect measurements of the PSD in thin textile structure. Other devices have been developed to provide direct measurements: X-ray ( Venkatarangan et al., 2000), electronic scan microscopy, optical profiling system.
International Journal of Clothing Science and Technology Vol. 20 No. 3, 2008 pp. 137-149 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810865193
IJCST 20,3
138
Porometer uses external pressure to expel liquid out of the pores in a saturated porous sample. Both the flow rate of gases passing through the pores and corresponding external pressures are measured starting from the largest pore (bubble point) to the smallest pores. Thus, PSD is obtained based on the equivalent pore radius from Laplace equation. However, this distribution is not a real distribution. In fact, the number of smallest pores is vague because it requires higher air pressure to be measured. Moreover, the device is not suitable for providing additional information like geometry and the shape of pores are neglected. In this approach, we want to study the PSD without neglecting the geometry (Koehl et al., 1998) and the shape of pores. Moreover, all of those devices are not able to measure the dynamic aspect of nonwovens and so it is difficult to understand hydraulic properties with such kind of devices. For all those reasons, we set up a test bed that allows us to study more easily the PSD and the hydraulic properties of pores through time. Pores size and geometry distribution function (PSD and PGD) The nonwoven materials are made up of pores in which gas effluents or liquids can run out. Fatt (1956) introduced the model of porous media in the rock science. A lot of works are done also in the same field (Bakke and Øren, 1996; Blunt and King, 1991; Blunt and Sher, 1995; Delerue et al., 1999a, b; Delerue and Perrier, 2002; Denesuk et al., 1993; Hidajat et al., 2002; Lindquist and Venkatarangan, 1999; Øren et al., 1998). In the research field of nonwoven, some papers have already been written (Pourdeyhimi, 1998, 1999; Pourdeyhimi and Ramanathan, 1996; Pourdeyhimi et al., 1997, 1999; Marmur, 1992; Marmur and Cohen, 1997; McBratney and Moran, 1994; Rebenfeld and Miller, 1995; Zeng et al., 2000). A good description of this unusual structure permits to model the fluid transport in the medium. It permits also to extract relevant parameters characterizing the pore network. It allows also a classification of porous materials according to their hydraulic properties (Anderson et al., 1996; Perwelz et al., 2000, 2001). Using our home made device, we grab a grey levels image of the sample. To determine the PSD, our approach is based on the detection of the eight-connected pixels belonging to the background of the image (Serra, 1982). The eight-connected pixels are connected if their edges or corners touch. The algorithm can be split into different steps. First of all, we convert the grey image into a binary image by using Otsu threshold. This method consists of calculating the threshold to minimize the interclass variance of the black and white pixels (Otsu, 1979). In the second step of the algorithm, the resultant image is labelled by using the following general procedures: (1) scan all image pixels, assigning preliminary labels to nonzero pixels and recording label equivalences in a union-find table; (2) solve the equivalence classes using the union-find algorithm (Sedgewick, 1998); and (3) relabel the pixels based on the resolved equivalence classes. The resulting matrix contains positive integer elements that correspond to different regions. For example, the set of elements equal to 1 corresponds to region 1; the set of elements equal to 2 corresponds to region 2; and so on.
We can extract from the labelled matrix all the geometric parameters concerning every pore (equivalent radius, main orientation angle area, eccentricity,etc.). We can model a pore with an ellipse or a rectangle or a convex hull. In this paper, we choose to characterize a pore with its equivalent radius r which represent its size parameter, and with its eccentricity e which represent its shape parameter. The equivalent radius corresponds to the radius of a circle with the same area as the region: Area 0:5 r¼ ð1Þ pi where, Area is the actual number of pixels in the region, since this number is proportional to the area in accordance with the resolution size. Another approach is based on the extraction of the biggest disk contained in a region ( Delerue et al., 1999a, b), but some pixels will be ignored. The PSD is represented by a histogram describing the frequency of pore radius r : Frequencyð%Þ ¼
Number of pores having radius r : Total number of pores
Pore network modelling
139
ð2Þ
By using the same algorithm, we extract the pores geometry distribution (PGD). To be more precise, we determine the eccentricity of each pore which is the ratio of the distance between the foci of the ellipse and its major axis length. The value varies between 0 and 1. An ellipse whose eccentricity is 0 is actually a circle, while an ellipse whose eccentricity is 1 is a line segment. The PGD is a histogram describing the frequency of pore eccentricity e: Frequencyð%Þ ¼
Number of pores having eccentricity e Total number of pores
ð3Þ
Liquid wicking in a nonwoven Theory To study the wicking of a liquid in an idyllic cylindrical tube (Figure 1), we apply the movement equation to the liquid: dðMV Þ ¼ F 2 Fh 2 P dt
ð4Þ
where: . V(t) denotes the rising velocity. . M(t) denotes the liquid mass. It is given by the formula: M ðtÞ ¼ r · p · R 2 · zðtÞ:
ð5Þ
2R V(t) z (t) z 0
x
Figure 1. Capillary rise
IJCST 20,3
.
P is the weight (force) applied on the liquid. It is given by the formula: P ¼ r · g · p · R 2 · zðtÞ:
.
F is the force which causes the liquid rise in the tube. It is given by the formula: F ¼ 2 · p · R · g · cos uE :
140 .
ð7Þ
The viscous friction force. This force exists only if z is big enough. It is given by the formula: F h ¼ 8 · p · h · V · zðtÞ:
.
ð6Þ
ð8Þ
z(t) denotes the altitude along the z-axis.
Observing the filling of a pore amounts observing the saturation S(%) versus time t. The term “saturation” is commonly used to define the liquid content of a porous medium (Ghali, 1994a, b). It is defined as the fraction of pores filled with liquid: S¼
Pores filled volume Total pores volume
However, the sample thickness remains the same. It can be expressed as: SðtÞ ¼
Number of filled pores having a radius r at a time t : Total number of pores having a radius r
ð9Þ
The used liquid is the n-decane. This choice is due to the hydrophobic character of our samples. The complete wetting character of the n-decane is traduced by the small value of its contact angle uE which is almost null. The evaporation of the n-decane is also negligible because pores are filled very quickly and decane has got a high-boiling temperature (1748C). Description of the test bed One of our objectives is the identification of the fluid transport in nonwovens. We implemented a test bed (Figure 2) able to provide direct observation of fluid online. This will allow us to estimate the hydraulic properties of nonwovens. Moreover, the size of the sample will depends only on the size of the random access memory (desktop capacity), the lens and the magnification of the microscope. In a first time, we were interested in the 2D geometry of the pore. A pore is certainly a 3D structure, but the thickness of chosen samples varies between 2 and 3 fibre diameters. One filament diameter is about 37 mm. Thus, the thickness value varies between 74 and 111 mm. Thus, we can neglect this thickness and base our study on 2D images. The sample can be considered as a thin layer. This approach will simplify, at the beginning, the 3D geometry. The test bed is composed from a digital scan CCD camera and a uniform red backlighting system (based on LEDs). The choice of red light is linked to the maximum sensitivity of the camera. It also permits to have a uniform illumination. The camera is linked to a microscope in order to choose the magnification adapted to the observed zone. To read information from the camera, we link it to a PC equipped with a frame grabber.
Pore network modelling
Red LEDs backlighting
141
Camera-microscope Nonwoven\wetting fluid
Figure 2. Scheme of the device of the image acquisition
Camera -PC
Manipulation of the test bed All the settings of the CCD camera are controlled by a suitable software. The camera, which has a great sensitivity, gets 24 frames per second (fps) in live. The red backlight is adjusted by a potentiometer, which permits the regulation of the frames contrast. The sample is prepared and vertically handled (Marmur and Cohen, 1997) on the support. Those samples are made of polyester which is very hydrophobic. It implies the use of the n-decane (C10H22) (Table I) which has a very low-surface tension. The video sequence starts getting dry nonwoven frames. The liquid is put in contact with the sample. At the same time, camera continues getting frames. Then we stop the video sequence when the fluid stops filling pores. It usually takes one minute to reach the equilibrium for a sample of 6 mm2 size. Algorithm to study the fluid transport We set up an algorithm (Figure 4) which extracts informations concerning the displacement of the fluid in the sample from the video sequence. It is implemented to analyse the frames containing the fluid transport by comparing them to the frame representing the dry sample (the referenced frame). More precisely, every pixel belonging to a pore region is compared to the same one in the template frame. In fact, pores pixels have higher values of grey levels because they are the most exposed to the red backlight. When the coloured liquid goes through the pore, it can only decrease the grey value of pixels. So we can so detect the grey variation of each pixel. This step of the algorithm increases the sensitivity of the liquid tracking, even with a small grey value difference. Each filled pore is referenced with the same label as when it was empty. Pores which remain empty get the value 0. We store at each time ti, the labelled pores which are filled. i denotes the frame number (Figure 3). Density ( r (kg/dm3))
Surface tension (g (mN/m))
Dynamic viscosity (h (mN.s/m2 ))
0.73
23.9
0.88
Table I. Physical values of n-decane at 208C
IJCST 20,3
Frame number i representing the fluid transport at a time ti.
Greyscale Image of the dry sample
Thresholding
142
Labelling the binary image
PSD&PGD of the empty pores Labelled image containing fluid transport Extraction of the filled pores
Figure 3. Tracking algorithm steps
ti+1=ti+1/24; i=i+1; i=number of the frame;[t]=sec.
Results and discussion We apply this approach to many samples made of polyester. In order to show the efficiency, we chose to represent one of those samples. Images have a size of 512*512 pixels. The resolution value is 5,120 dpi. Pores size distribution function Images shown in Figure 4 represent a greyscale image of the sample. In Figure 5, coloured regions represent labelled pores of Figure 4. A visual comparison between pores of the greyscale images and those of the labelled images obviously show that the Otsu threshold is a suitable choice. However, some pores are not labelled. In fact, pores, which are well surrounded by fibres, take the same grey value as the latter which are well exposed to the light.
Figure 4. Grayscale image of the sample
500 µm
Pore network modelling
143
Figure 5. Labelled image of sample
We apply the algorithm, described in paragraph II, to the image of the sample. We obtain a PSD (Figure 6) containing a large variety of pores distributed according to their equivalent radius. All pores are taken into account because of its important role in the capillary action. We determine also the PGD (Figure 7). This distribution, expressed by the eccentricity e, informs us about the geometry of pores. We note that, some of their, are circular whereas the majority of pores are assimilated to tubes. Fluid transport algorithm We apply the algorithm shown in Figure 3 to the video sequence in order to better understand the wetting process: what is the required pore size/geometry involved in the liquid transportation, is there a linear relation between saturation S and time t (Figure 8).
14
frequency (%)
12 10 8 6 4 radius (µm)
2 0 2
10
17
24
31
38
46
53
60
67
75
82
Figure 6. PSD function of the sample
IJCST 20,3
8 7
144
Frequency (%)
6 5 4 3 2 1
Figure 7. Pores geometry distribution
Figure 8. Detection of the filled pores versus time: (a) 0 s; (b) 8 s; (c) 27 s; (d) 110 s
0 0.00
0.51
0.62
0.69
0.76
0.82
0.88
0.94
Eccentricity
(a)
(b)
(c)
(d)
Figure 9 shows the saturation S(%) versus time t. By observing the graph, we note that the relation between saturation S(%) and time t is linear for only small pores (small equivalent diameters) and only on very small interval of time (0 , t , 18 s). More the equivalent diameter of pores becomes increasingly large the relation between saturation S(%) and time t becomes less and less linear. We deduce that small pores approach the model of a tube on very small interval of time and for small pores.
Pore network modelling
S(%) = f(t) S (%) 120
Equivalent diameter 9.6 µm
100 14.4 µm
19.2 µm
80
145
24 µm 28.8 µm
60 33.7 µm 38.5 µm
40
43.3 µm 52.9 µm
20
67.3 µm
0 0
20
40
60
80
100
time t (sec)
The slopes in the graph (Figure 9) show how fast the small pores are filled. The saturation versus time t decreases as the radius of the pores increases. But the linear aspect is true only for short times (0 , t , 18 s). Beyond this interval of time, we are in a non linear case. The non linear aspect of the saturation versus the time is due to many aspects. First of all, a nonwoven is a porous media with a very low density. For our samples, the porosity is estimated up to 90 per cent. This means that there is 90 per cent of air and 10 per cent of fibers and so 10 per cent of pore wall. For medium and big pores, the shape cannot no more be considered as a proper capillary tube and thus the capillary forces in a pore are weaker than capillary forces in a tube. The second key-point concerns the connection between pores. We note from the Figure 10 that only the category of small pores reach a saturation value of 100 per cent or near to 100 per cent. In fact, some pores, which are supposed to be filled, stay empty. We notice this when we compare the PSD of empty pores and the PSD of filled pores (Figure 7). In fact, some small and medium pores are not filled because of their neighbourhood. The connection between pores is a very important parameter that appears too in our results. In our samples, pores are not distributed in a uniform way. An isolated small pore has less chance to be filled than a pore which is in the neighbourhood of other small pores (Figure 11). The probability of filling a medium or large pore, also, increases or decreases according to his vicinity of small pores. This explains the way followed by the n-decane in our sample. In fact, the wetting process always starts by filling the small pores, then the medium ones which are connected to them, and so on until the equivalent diameter reaches a threshold (for this sample 67 mm (Figure 9)).
Figure 9. Saturation S(%) versus time t
14
S (%)
IJCST 20,3
12 10
146
8 6 4 2
radius (µm)
PSD filled pores
82
75
67
60
53
46
38
31
24
17
Figure 10. PSD of filled pores sample 1
10
2
0
PSD empty pores
ISOLATED SMALL PORE
Figure 11. Connection between non uniform pores
GROUP OF SMALL PORES
Concerning the dynamic equation (4), it can be reduced to the following form: dðMV Þ ¼F ð10Þ dt We note that the force of ascension F is more important than the weight force P. In fact, if we make the assumption that the liquid fills an entire pore having a radius of 10 mm and a depth of 100 mm. According to equation (5), P ¼ 2.25 £ 102 10 N. The calculation of the force of ascension (equation (6)) gave the value of F ¼ 1.44 £ 102 6 N. We note so that F . . P. So we neglect the weigh force P in relation to the ascension force F. The viscous force Fh is negligible when the height value z is low. The height of the observed samples is also small. But the real reason to neglect the force of viscous friction is the few percentages of walls in a pore. A pore is delimited only by very few fibres (90 per cent of vacuum). Those assumptions need further experiments to get a better understanding (Figure 12).
Pore network modelling
147
Figure 12. Equilibrium state
Conclusion Nowadays image analysis is a widely used and suitable technique for extracting tool which permits to extract a lot of information from an image. In this paper, we use this tool to determine the PSD and the PGD. We showed in this paper that the equivalent radius and the eccentricity are geometric key-parameters to explain the characterisation of a pore network. The proposed model leads to promising results. However, the images used must be enhanced by modifying the camera settings in order to well identify fibres from pores even the smallest ones. It would be also nice to dye fibres before starting the wetting process. A 2D study is sufficient to analyse thin samples but to analyse thick samples, a 3D study is essential. The PSD and the PGD informs us about pores but not about their roles on the transport of a wetting fluid. The algorithm used show that the size and the shape of pores, the fibres and the distance between them, and the connection between pores are parameters which characterise the displacement of the fluid in the pore network. We note also that the capillary laws are applied only on small pores and for very short interval of time because of thickness and the very important porosity of our samples. We will make some experiences in order to adapt the capillary laws to our nonwovens. This kind of approach will help us to model the pore network in a nonwoven. It will also permit us to extract hydraulic parameters (capillary pressure, filtration level) and to model the fluid transport in any kind of nonwoven. References Anderson, A.N., McBratney, A.B. and FitzPatrick, E. (1996), “Soil mass, surface, and spectral fractal dimensions estimated from thin section photographs”, Soil Sci. Soc. Am. J, Vol. 60, pp. 962-9. Bakke, S. and Øren, P.E. (1996), “3-d pore-scale modelling of heterogeneous sandstone reservoir rocks and quantitative analysis of the architecture, geometry and spatial continuity of the
IJCST 20,3
148
pore network”, SPE 35479, European 3-D Reservoir Modelling Conference, SPE, Stavanger, pp. 35-45. Blunt, M.J. and King, P. (1991), “Relative permeabilities from two- and three-dimensional pore-scale network modeling”, Transport in Porous Media, Vol. 6, pp. 407-33. Blunt, M.J. and Sher, H. (1995), “Pore network modeling of wetting”, Physical Review E, Vol. 52, pp. 63-87. Delerue, J.F. and Perrier, E. (2002), “DXSoil, a library for 3D image analysis in soil science”, Computer & Geosciences, Vol. 28, pp. 1041-50. Delerue, J.F., Perrier, E., Yu, Z.Y. and Velde, B. (1999b), “New algorithms in 3D image analysis and their application to the measurement of a spatialized pore size distribution in soils”, Physics and Chemistry of the Earth, Part A: Solid Earth and Geodesy, Vol. 24 No. 7, pp. 639-44. Delerue, J.F., Perrier, E., Timmerman, A., Rieu, M. and Leuven, K.U. (1999a), “New computer tools to quantify 3D porous structures in relation with hydraulic properties”, in Feyen, J. and Wiyo, K. (Eds), Modelling of Transport Processes in Soils, Wageningen Pers, Wageningen, pp. 153-63. Denesuk, M., Smith, G.L., Zelinski, B.J.J., Kreidl, N.J. and Uhlmann, D.R. (1993), “Capillary penetration of liquid droplets into a porous materials”, Journal of Colloid and Interface Science, Vol. 158, pp. 114-20. Fatt, I. (1956), “The network model of porous media I. capillary pressure characteristics”, AIME Petroleum Transactions, Vol. 207, p. 144. Ghali, K., Jones, B. and Tracy, J. (1994a), “Experimental techniques for measuring parameters describing wetting and wicking in fabrics”, Textile Res. J., Vol. 64, pp. 106-11. Ghali, K., Jones, B. and Tracy, J. (1994b), “Modeling moisture transfer in fabrics”, Exp. Thermal Fluid Sci., Vol. 9, pp. 330-6. Hidajat, R.A., Singh, M. and Nohanty, K. (2002), “Transport properties of porous media reconstructed from thinsections”, SPE Journal, March, pp. 40-8. Koehl, L., Zeng, X., Ghenaim, A. and Vasseur, C. (1998), “Extracting geometrical features from a continuous filament yarn by image processing techniques”, Journal Textile Inst.89, Part1, Vol. 1, pp. 106-16. Lindquist, W.B. and Venkatarangan, A. (1999), “Investigating 3D geometry of porous media from high resolution images”, Phys. Chem. Earth (A), Vol. 25 No. 7, pp. 593-9. McBratney, A.B. and Moran, C.J. (1994), “Soil pore structure modelling using fuzzy random pseudo fractal sets”, International Working Meeting on Soil Micromorphology, pp. 495-506. Marmur, A. (1992), “Penetration and displacement in capillary systems”, Advances in Colloid and Interface Science, Vol. 39, pp. 13-33. Marmur, A. and Cohen, R.D. (1997), “Characterization of porous media by the kinetics of liquid penetration: the vertical capillaries model”, Journal and Colloid and Interface Science, Vol. 189, pp. 299-304. Øren, P.E., Bakke, S. and Arntzen, O.J. (1998), “Extending predictive capabilities to network models”, SPE Journal, Vol. 3 No. 4, pp. 324-36. Otsu, N. (1979), “A threshold selection method from grey-level histograms”, IEEE Transactions Systems, Man, and Cybernetics, Vol. 9 No. 1, pp. 62-6. Perwelz, A., Cassetta, M. and Caze, C. (2001), “Liquid organisation during capillary rise in yarns-influence of yarn torsion”, Polymer Testing, Vol. 20, pp. 553-61.
Perwelz, A., Mondon, P. and Caze, C. (2000), “Experimental study of capillary flow in yarns”, Textile Research Journal, Vol. 70 No. 4, pp. 333-9. Pourdeyhimi, B. (1998), “Reply to comments on measuring fiber orientation in nonwovens”, Textile Research Journal, Vol. 4, pp. 307-8. Pourdeyhimi, B. (1999), “Characterizing fiber diameter variability in nonwovens”, INJ, Spring, pp. 29-35. Pourdeyhimi, B. and Ramanathan, R. (1996), “Measuring fiber orientation in nonwovens part II: direct tracking”, Textile Research Journal, Vol. 12, pp. 747-53. Pourdeyhimi, B., Dent, R. and Davis, H. (1997), “Measuring fiber orientation in nonwovens, part III: fourier transform”, Textile Research Journal, Vol. 2, pp. 143-51. Pourdeyhimi, B., Dent, R., Jerbi, A., Tanaka, S. and Deshpande, A. (1999), “Measuring fiber orientation in nonwovens part V: real webs”, Textile Research Journal, Vol. 3, pp. 185-92. Rebenfeld, L. and Miller, B. (1995), “Using liquid flow to quantify the pore structure of fibrous materials”, J.Text.Inst., Vol. 2, pp. 241-51. Sedgewick, R. (1998), Algorithms in C, 3rd ed., Addison-Wesley, Reading, MA, pp. 11-20. Serra, J. (1982), Image Analysis and Mathematical Morphology, Academic Press, New York, NY. Venkatarangan, A., Lindquist, W.B., Dunsmuir, J. and Wong, T-F. (2000), “Pore and throat size distributions measured from synchrotron x-ray tomographic images of Fontainebleau sandstones”, Journal of Geophysical Research – Solid Earth, Vol. 105 No. 9, pp. 21509-27. Zeng, X., Vasseur, C. and Fayala, F. (2000), “Modeling micro geometric structures of porous media with a predominant axis for predicting diffusive flow in capillaries”, Applied Mathematical Modelling, Vol. 24, pp. 969-86. Further reading Patzek, T.W. (2001), “Verification of a complete pore network simulator of drainage and imbibition”, SPE Journal, Vol. 6 No. 2, pp. 144-56. Corresponding author M. Dimassi can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Pore network modelling
149
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,3
Modelling and simulation of filtration through woven media M.A. Nazarboland, X. Chen and J.W.S. Hearle
150
School of Materials, University of Manchester, Manchester, UK, and
R. Lydon and M. Moss Clear Edge Group, Haslingden, UK Abstract Purpose – This paper aims to discuss the development of a software tool UniverFiltere which is capable of geometrical modelling of 3D woven fabrics, interfacing with computational fluid dynamics tools to numerically determine the fluid (and more specifically liquid) flow path and simulating the filtration process by introducing particles of various shapes and sizes. Design/methodology/approach – The method employed in creating the software tool is based on geometrical modelling of the single-layer woven fabric with monofilament yarns, numerical analysis of the fluid-flow problem, and mathematical modelling of the forces exerted on particles to accurately predict the settlement of such particles on the fabric. In the case of particle motion, a Lagrangian approach is used. Findings – Creation of a software tool capable of simulation and modelling the filtration process through woven fabrics is the primary achievement. The effect of geometrical parameters of the woven fabric on fluid flow utilizing the results from fluid pressure and fluid velocity on the fabric show that the fluid flow is significantly influenced in the interstices and chamber downstream by the fabric. Fluid-flow resistance and pressure loss are obtained from the results of fluid velocity and pressure. The results from the fluid pressure on the fabric could also be employed to more accurately predict how pore shapes and sizes are transformed. Originality/value – Creation of a modelling tool for filtration through woven fabric media. This software is the foundation of establishing a standalone tool with the capability to design, test and improve fabric filter design for more efficient filtration properties. Keywords Filtration, Flow, Modelling, Simulation, Textiles Paper type Research paper
International Journal of Clothing Science and Technology Vol. 20 No. 3, 2008 pp. 150-160 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810865201
Introduction Filtration is a process widely used in numerous industrial applications, especially in chemical, medical, food and paper industries. Developing an accurate theoretical model of filtration is difficult because of the nature of the filtration process, which involves particle transport in a fluid moving through complex fabric geometry. The fluid and suspended particles flowing through the maze of yarns follow a tortuous path controlled by the fluid dynamics and equations of motion for particles. Since, the geometry and porosity of the fabric filter is determined by the weave pattern and the various parameters of the yarns constituting the fabric (Backer, 1951), it is important to optimize its structure to achieve the most efficient filtration. The past practice relied on the practical skill and experience of the fabric designers and on empirical trials. Simulation and modelling offers excellent methods in quantitative and qualitative understanding of the filtration process. Costs of experimentation can be substantially reduced if these are combined with theoretical modelling investigations. Furthermore, these investigations provide an accelerated perception of the methodology required to
breakdown the process into its fundamental physical phenomena. Once a model has attained a significant level of confidence, it can be successfully utilized as a design tool for new processes in filtration or optimization of existing ones. The work and research carried out, contributing to the construction of UniverFiltere (Nazarboland et al., 2006a), has focused on the development of a woven fabric filter model, an interface to a finite volume element modelling tool that acquires the fluid-flow paths, and introduction of particles with different size and shapes to follow such paths. The results are utilized in simulation of the initial stages of cake formation and determining the filtration properties that can lead to the most efficient filtration. Literature review Construction of a software tool to simulate the filtration process through woven fabric filters entails two major stages. The first stage focuses on the creation of the 3D woven structure and the second on modelling of the filtration process. Over the last century, various techniques have been employed to model woven fabrics, specifically, yarn paths. The simplest approach was by Peirce (1937) in which the yarn path, in the case of plain weave, is represented by two arcs tangentially connected by a straight-line segment. The Peirce technique is regarded as an idealized approximation and recent methods employ splines to formulate the yarn paths and surfaces. A spline curve can mathematically be described as a piece-wise cubic polynomial function whose first and second derivatives are continuous across the various curve sections. In computer graphics, spline refers to any composite curve formed with polynomial sections satisfying specified continuity condition at the boundary of the pieces. B-Splines are the most widely used class of approximating splines and are utilized in several texts (Lin and Newton, 1999; Lahey, 2002) for simulating woven fabrics. A general expression for the calculation of coordinate positions along a B-Spline curve can be written as follows: P ðuÞ ¼
n X
pk Bk;d ðuÞ
ð1Þ
k¼0
where umin # u # umax , 2 # d # n þ 1 and pk are an input set of n þ 1 control points and Bk,d is a B-Spline blending polynomial functions of d-1 degree, where parameter d can be chosen to be any integer value in the range from 2 up to the number of control points (n þ 1). The blending functions for B-Spline curves are defined by the Cox-de Boor recursion formulas: ( 1; if uk # u # ukþ1 Bk;1 ðuÞ ¼ 0; otherwise
Bk;d ðuÞ ¼
u 2 uk ukþd 2 u Bk;d21 ðuÞ þ Bkþ1;d21 ðuÞ ukþd21 2 uk ukþd 2 ukþ1
ð2Þ
and each blending function is defined over d subintervals of the total range of u.
Modelling and simulation of filtration 151
IJCST 20,3
152
Values for umin and umax depend on the number of control points selected, the value for parameter d and how the subintervals are set up. The second stage of the simulation software development focuses on the filtration aspect. Prior to this, the effects of woven fabrics on fluid-flow behaviour need to be analyzed and integrated to the system. Fluid flow is classified into laminar and turbulent flows and is described by a set of equations based on conservation of mass and momentum. For an incompressible fluid, these equations are, respectively: ~¼ 7·V
›u ›v ›w þ þ ¼0 ›x ›y ›z
" !# ~ Duj ›P › ›ui ›uj 2dij 7 · V ¼2 r þ m þ 2 Dt ›xj ›xi ›xj ›xi 3
ð3Þ
ð4Þ
where i and j are x, y and z coordinate directions, m is the viscosity coefficient and dij is the Kronecker d. Generally, two different approaches have been employed in previous investigations on fluid flow through woven fabrics, one being simplifying the fabric into an assembly of orifices and the other the use of fabric models. The former was proposed by Backer half a century ago suggesting that the interyarn pores in a fabric to be represented by an assembly of orifices or nozzles through which passes the major portion of the flow (Backer, 1948). Others (Lu et al., 1996; Ting et al., 2005) used the second approach for numerical analyses of the size and shape of pores using computational fluid dynamics (CFD). Over the past two decades, various group have contributed in formulation of filtration simulation software (Tiller, 1975; Shirato et al., 1986; Wakeman, 1986). Holdich (1994) devised a spreadsheet program, which used certain physical properties of the filtration process to calculate cake height and weight. Koenders and Wakeman (1997) used data from a range of experiments on dilute suspensions to determine the physics of suspension flow. However, this model did not consider the contribution of individual particles. Most other modelling tools consider depth filtration through packed beds (Shin, 2006; Latz and Wiegmann, 2003) rather than surface filtration that is the main technique utilized in filtration through woven fabrics. Thus, a filtration simulation software tool needs to be devised that considers surface filtration using woven fabric media. The level of accuracy needs to be enhanced by considering the effect of individual particles and the precise direction of fluid flow on the filtration process. Methodology Fabric geometrical modelling One of the main properties of woven media employed in filtration is their specific geometrical opening that ensures selective particle cut off. The woven fabric should filter all the contaminant particles larger than its pore size and practically none of the smaller dimensions. UniverFilter is used to create 3D geometrical models on the provision of fabric weave, yarn and fabric structural parameters. The software uses both Peirce’s woven fabric model and the more realistic B-Spline model for creating the yarn path in the fabric.
The paths in real fabrics are complicated and their shapes depend upon many factors including the yarn mechanical properties. The shape of the yarn path is finalized when the minimum energy is achieved in the yarns (Jiang and Chen, 2005). The B-Spline function represents the yarn paths by defining control points rather than coordinate points to allow a better representation of the yarn paths by taking into consideration of energy minimization of the constituent yarns in the fabric. To ensure that the curve signifying the yarn path centreline passes through the control points, the Hermite form (de Boor et al., 1987) of a cubic spline is utilized. The latter is determined by defining positions and tangent vectors at the data points (Hollig and Koch, 1995). UniverFilter uses three idealized yarn cross-sectional shapes (Figure 1) for both warp and weft yarns, which are circular, racetrack and lenticular.
Modelling and simulation of filtration 153
Boundary conditions of fluid flow Filtration entails the flow of fluid through the filter. Having already designed the filter, the next step requires establishing the fluid-flow paths. However, due to the high complexity of fluid dynamics, there is no other choice than to obtain the approximation solution numerically by the use of a CFD program. Powerful commercial CFD software, Fluent Inc. (2004) is utilized for analysis. The method used in such software is to discretize the fluid domain into small cells to form a volume mesh or grid, and then apply iterative methods to solve the equations of motion governed by the Navier-Stokes equations. The method devised in this work is to exports the fabric geometry, usually one weave repeat, into the CFD software. The fabric is placed inside a rectangular chamber (Figure 2). The specific position of Circular Racetrack
Lenticular W1<W2<W3
H1
H3
H2
W3
W2
W1
Direction of Flow
Outlet Pressure/ Velocity
Direction of Flow Y Z
Figure 1. The different yarn x-sectional shapes
Symmetry (or Wall) Boundary
Inlet Pressure
X
H1>H2>H3
Sample Fabric
Chamber (Meshed Volume)
Figure 2. Schematic of simulation set-up to find fluid-flow path
IJCST 20,3
154
the fabric inside the chamber is dependant to fabric dimensions and the physical properties of the fluid. This is a separate topic and is not discussed here. The two planes of the chamber parallel to the fabric are the inlet and outlet pressure boundaries. The cross-sectional area of the chamber is related to the size of fabric repeat. The four walls of the chamber are regarded as symmetry surfaces so that the results indicate a better depiction of real situations. The resulting volume of the structure is then meshed using tetrahedral elements. A denser grid cell distribution is applied to areas on the fabric surface to account for the large gradients that occur in these areas. All the corresponding parameters are set and the simulation is run until the data are converged. The results of fluid-flow paths, the fluid pressure and velocity in the chamber downstream and all relevant data are imported by UniverFilter for further analysis. The ultimate aim of the UniverFilter software program is that in a particular filtration dilemma, the tool should be able to predict the most efficient structural geometry of the fabric filter. By varying the parameters of the fabric, including fabric density, yarn linear density, crimp and cross-sectional shape of the yarns, a study was carried out to see how changes in structural parameters of the fabric affect fluid flow and eventually the filtration process. The setup of the experiment is as shown in Figure 2, where the four walls of the chamber are regarded as solid and therefore are not permeable to the flow. Liquid-water was used as the Newtonian fluid (with density of 998.2 kg/m3 and viscosity at 102 3 kg/ms). The inlet pressure for all cases was set at three bars and the operating pressure set at one bar. Density and viscosity of the fluid were assumed to be constant, corresponding to the isothermal approach. In every analysis, three different fabric models (all plain), with all their constituting elements except one varying are used. Among the numerous outputs from the analysis, three were chosen because of their relevance to the nature of filtration, which are the fluid pressure, fluid velocity and shear stress on the fabric. The ability to calculate the fluid pressure and fluid velocity before and after the filter fabric provides useful information for flow behaviour assessment. The differences in pressure and velocity on the two sides of the filter fabric reflect how the flow is affected by the structure of the filter fabric. The information can, for example, be used to engineer filter fabric structures with specific flow requirements. It is also useful to help establish particle capture rules and therefore to simulate the cake forming process. Assessing the fluid pressure distribution analysis on the surfaces of the fabric will provide information on yarn and fabric structures, which could be employed to more accurately predict transformations in the shape and size of the pores. This can also maximize the life cycle of the filter fabric. The shear stress can be utilized in the implementation of the particle capture rules. These values can assist in predicting the probabilities of particle attachment to the fabric. If a distribution of particle sizes was known, semi-empirical rules could be formulated to divide the particles into groups centred on some chance of the particles being stopped. Shear stress values on the surfaces of the fabric assist in defining these probabilities. Introduction of particles The next stage involves introduction of the particles, suspended in the fluid. According to the defined particles distribution, they are placed randomly at a starting position
corresponding to the starting points for elements of fluid-flow paths at the inlet pressure plane of the chamber. Particles of various size and shapes can be used in the system. The particle shapes considered are spherical, discus shaped and needle shaped. The simulation is then run with the particles following the fluid-flow paths. As the particles move towards the locality of the fabric filter, particle capture rules are applied. It is normally expected that the particles that intercept the fabric, should be stopped. However, the particle location compared to the fabric filter is not the only criteria that should be used in stoppage of such particles. For instance, the particle velocities can affect some capture mechanisms such as diffusion and the fluid pressure can influence the interception mechanism. Another factor that will contribute to particle capture is the particle size and shape. It is expected that the woven fabric should filter all the particles larger than its pore size and practically none of the smaller dimensions. However, the fluid pressure and particle abrasions on the fabric could transform the effective pore area. Such changes to the pore structure need to be considered and thus provisions be made in the program for implementing depth filtration in case particles are able to get through. As far as particle size is concerned, in case of colloidal particles, the Brownian energy governs the particle motion in the locality of the fabric. The methodology in design and implementation along with the results of particle transport and capture is extensively explained in our previous paper (Nazarboland et al., 2007). The various forces including the physicochemical interactions are modeled and the significance of variation of the particle shape and its effects in the differences in filtration properties is investigated.
Modelling and simulation of filtration 155
Results and discussions The effect of variation in the structural parameters of the fabric on fluid-flow behaviour illustrates some fascinating results (Nazarboland et al., 2006b). Variation of yarn linear density relates generally to changes in the yarn thickness if the packing factors are the same. The increase in linear density of the weft yarn decreases the effective pore area and increases the fabric resistance to the fluid flow. The filter fabric causes the fluid pressure to drop sharply. It also causes the fluid velocity to reach a peak because of the sudden reduction of cross-sectional area that the flow can go through. The former phenomenon implies that the filter fabric is subject to force because of the pressure difference it causes. The fluid pressure and velocity after the fabric shows a general response to the change in linear density of the yarn (Figure 3),
16
12
8 0 Direction of Flow
2.8 Total Pressure (bar)
Fluid Velocity (m/s)
YLD1 YLD2 YLD3
Fabric Position
20
1
2
3
Chamber Length (mm)
(a)
4
5
YLD1 YLD2 YLD3
Fabric Position
2.4 2 1.6 1.2 0 Direction of Flow
1
2
3
Chamber Length (mm)
(b)
4
5
Figure 3. Variation of (a) fluid pressure and (b) fluid velocity in the chamber downstream for increasing weft yarn linear density (from YLD1 to YLD3), keeping all other parameters constant
IJCST 20,3
156
that is the higher the yarn linear density value for the fabric, the lower the pressure and the fluid velocity in the chamber after the fabric. Increasing the fabric density, which refers to the number of warp/weft yarns in a given area, decreases the effective pore area of the fabric (Figure 4). Similar trends as those of increasing yarn linear density are observed. This is so because the effect of changing the yarn linear density and changing the fabric density is similar, which is ultimately the alteration of the effective pore size. The fundamental difference between the effect of varying the fabric density or the yarn linear density on the fluid-flow behaviour is two parameters that determine the effective pore dimensions. By varying the fabric density, only the pore x-sectional area is altered whereas by varying the yarn linear density, the pore depth changes as well. The waviness of the yarn in fabric is termed the yarn crimp. Other conditions being the same, increasing the yarn crimp will increase the length of the yarns and the depth of the pore. This leads to the increase of the yarn surface area within the fabric in the 3D space and increases the fluid-flow resistance of the filter fabric. This is shown in Figure 5. Other conditions being the same, increasing the yarn crimp will increase the length of the yarns and the depth of the pore. This leads to the increase of the yarn surface area within the fabric in the 3D space and increases the fluid-flow resistance of the filter fabric. As it can be seen in Figure 6, fluid flow before the proximities of the fabrics have similar pressure values. The fluid pressure is then dropped according to the position of the fabric. There is an exception to the fluid pressure change for model CR3 in which the fluid pressure primarily increases before being reduced. This is related to the warp yarns of the fabric to be straight due to highest crimp. This increases the fluid resistance of the fabric to its maximum since the flow cannot slide through as freely as when the warp yarns were not straight. The increasing pore depth also has some
Total Pressure (bar)
Fluid Velocity (m/s)
Figure 4. Variation of (a) fluid pressure and (b) fluid velocity in the chamber downstream for increasing fabric (weft) density (from FD1 to FD3), keeping all other parameters constant
20 15 10 5 0 Direction of Flow
1
2
3
4
5
Chamber Length (mm)
d1
Note: From left to right
FD1 FD2 FD3
Fabric Position
2.6
2.1
1.6
1.1 0 Direction of Flow
1
2
3
4
Chamber Length (mm)
(a)
l1
Figure 5. Increasing level of crimp
3.1
FD1 FD2 FD3
Fabric Position
25
(b)
d2
l2
d3
l3
c1 < c2 < c3 l1 < l2 < l3 d1 < d2 < d3
5
contributing factors. After the fabric, the fluid pressure is decreased for an increasing yarn crimp. The overall pressure drop in the chamber increases from model CR1 to CR3. Varying the x-sectional shape of the yarn is essentially changing the yarn x-sectional width and height. Some could argue that transformations of yarn x-sectional shapes could have similar influences to the performance of the fluid flow to that of a combination of changing fabric density and yarn linear density. Even though in some levels this is true, the implications are different when introducing particle-capture rules. From the latter perspective, if instead of variation in cross-sectional shape of the yarns, the fabric density for instance is changed, the ratio of the size of the particle to that of the pore and yarn width changes. This change could make the filter ineffective in stopping particles. Even in relation with the fluid-flow behaviour, altering the x-sectional shape of the yarns transforms the shape of the outer surface of the yarn. This change influences the fluid-flow performance in a different manner to those discussed previously. Changing yarn x-sectional shape from circular to racetrack to lenticular reduces the pore dimensions, to be precise the pore depth and width. As expected, fluid flow with similar pressure values reach the proximity of the fabric. Reduction in pore depth influences the fluid pressure to slightly increase, before decreasing in the chamber downstream at the back of the fabric. These changes are shown in Figure 7(a). The effect of varying the yarn cross-sectional shape on the pore is two-fold as it changes 20 16 14 12 10 8
2.5
Figure 6. Variation of (a) fluid pressure and (b) fluid velocity in the chamber downstream for increasing weft yarn crimp (from CR1 to CR3), keeping all other parameters constant
2 1.5
6 4 0 Direction of Flow
1
2
3
1 0 Direction of Flow
4
Z-plane (mm)
1
2
Circ Race Lent
2.8 Total Pressure (Bar)
Fluid Velocity (m/s)
Fabric Position 20
16
12
1
2
3
Z-Dimenson (mm)
(a)
4
(b)
24
8 0 Direction of Flow
3
Z-Dimension (mm)
(a)
4
Circ Race Lent
Fabric Position
2.3 1.8 1.3
0.8 0 Direction of Flow
1
2
3
Z-Dimenson (mm)
(b)
157
CR1 CR2 CR3
Fabric Position
3
Total Pressure (bar)
Fluid Velocity (m/s)
3.5
CR1 CR2 CR3
Fabric Position
18
Modelling and simulation of filtration
4
Figure 7. Variation of (a) fluid pressure and (b) fluid velocity in the chamber downstream for various yarn x-sectional shapes (circular, racetrack, and lenticular), while keeping all other parameters
IJCST 20,3
158
from circular to racetrack to lenticular. This is that both the pore width and the pore depth decrease. The former is directly related to the fluid velocity. For smaller pore width, the velocity before the fabric decreases and for smaller pore depth, the fluid velocity through the fabric is expected to rise. This anticipation is followed except when comparing the fluid velocity for lenticular and racetrack through the fabric. Even though the lenticular shape yarns are wider than the racetrack but the shape of the front surface of the yarn in case of racetrack reduces the fluid velocity. This is a flat surface and the plane is perpendicular to the direction of the flow. However, in the case of the lenticular yarn the front surface is more curved which decreases the fluid-flow resistance. Figure 7(b) shows how changes in yarn cross-sectional shapes affect fluid velocity in the chamber downstream. The pressure drop increases from circular to racetrack to lenticular. As shown by Figure 8, the fluid pressure is increased on the front face of the fabric due to higher flow resistance and drops dramatically at the back face. These pressure distributions could be employed to more accurately predict transformations in the shape and size of the pores. The exact levels of pore shape transformation are being investigated and will be reported in a later stage. From these results, it is deduced that ultimately changes in the effective area has a direct effect on fluid-flow behaviour. This influence of the effective pore area on the fluid-flow performance can be backed up using theoretical analysis as well. The variation in velocity is inversely proportional to the pore x-sectional area by means of the equation for conservation of mass and to the square root of pressure through the Bernoulli’s equation (Vardy, 1990). Thus, pressure is directly proportional to the x-sectional pore area. In the instance of pore depth, the effect of drag should also be considered. By enlarging the pore depth, the drag acting on the fluid (observed as shear stress) transforms portion of the fluid kinetic energy. This loss in kinetic energy will be investigated and reported in a later stage. Another important aspect in studying the fluid-flow behaviour through various filters is the consideration of the cake build-up 3.5
Fabric Front Surface
2.8
2.0 Circular
Racetrack
Lenticular
1.4
Figure 8. Pressure profile on front/back surfaces of fabrics with different x-sectional shapes constant
Fabric Back Surface
0.7 X Z 0.0 Bar
Y
and how it affects the flow performance. Similar tool as outlined here, but different simulation experiment set-up is used to investigate the effect of cake formation on fluid flow. This will also be reported in the next paper. Conclusions Evidently, the work carried out in this paper forms the foundations in establishing a software tool that can be used to test and improve fabric filter designs for more efficient filtration properties. Even at this early stage, the results generated by the computer model are encouraging. The advantage of predictive modelling can be significant in terms of system optimization and the design of new processes, although the model requires further development before this can be realized. The effect of varying fabric structural parameters, specifically yarn linear density, fabric density, crimp and yarn cross-sectional shape, on the fluid-flow performance has been studied numerically using a CFD software tool. Each one of these parameters could have a significant bearing on the effects of the fabric’s geometrical parameters on the fluid-flow behaviour. The exact influence and the links between fluid flow and filtration need to be further investigated. The fluid pressure and velocity in the proximities of the fabric and in the chamber downstream were obtained. Results of the analysis show that these parameters are significantly manipulated according to the effective pore areas. The results further illustrate that alterations in the pore width is directly and the pore depth is inversely proportional to the fluid velocity. Results attained from the shear stress, which are influenced by the yarn surface area and yarn shape, could prove useful in implementing the initial stages of cake formation. Finally, the pressure distribution on the surfaces of the fabric could facilitate calculation of deformity in the yarn shape and permeability. Ultimately, it is the particle capture mechanisms along with the fluid-flow behaviour that govern the selection of the fabric parameters. It is therefore this combination that guides the design engineer in determining the paramount solution to a filtration dilemma. References Backer, S. (1948), “The relationship between the structural geometry of textile and its physical properties, I: literature review”, Text. Res. J., Vol. 18, pp. 650-8. Backer, S. (1951), “The relationship between the structural geometry of textile and its physical properties, Part IV: interstice geometry and air permeability”, Text. Res. J., Vol. 21, pp. 703-14. de Boor, C., Hollig, K. and Sabin, M. (1987), “High accuracy geometric Hermit interpolation”, Comp. Aided Geo. Des., Vol. 4, pp. 269-78. Fluent Inc. (2004), Fluent User’s Guide, Version 6.1, Fluent Inc., Lebanon, NH, available at: www. fluent.com Holdich, R. (1994), “Simulation of compressible cake filtration”, Filt. Sep., Vol. 31 No. 8, pp. 825-9. Hollig, K. and Koch, J. (1995), “Geometric Hermite interpolation”, Comp. Aided Geo. Des., Vol. 12, pp. 567-80. Jiang, Y. and Chen, X. (2005), “Geometric and algebraic algorithms for modelling yarn in woven fabrics”, JOTI, Vol. 96 No. 4, pp. 237-45.
Modelling and simulation of filtration 159
IJCST 20,3
160
Koenders, M.A. and Wakeman, R.J. (1997), “Initial deposition of interacting particles by filtration of dilute suspensions”, AIChE J., Vol. 43 No. 4, pp. 946-58. Lahey, T.J. (2002), “Modelling hysteresis in the bending of fabrics”, MSc thesis in Systems Design Engineering, University of Waterloo, Waterloo. Latz, A. and Wiegmann, A. (2003), Simulation of Fluid Particle Separation in Realistic Three-dimensional Fibre Structure, Filtech, Dusseldorf. Lin, H.Y. and Newton, A. (1999), “Computer representation of woven fabric by using B-Splines”, J. Text. Inst., Vol. 90 No. 1, pp. 59-72. Lu, W., Tung, K. and Hwang, K. (1996), “Fluid flow through basic weaves of monofilament filter cloth”, Text. Res. J., Vol. 66, pp. 311-23. Nazarboland, M.A., Chen, X., Hearle, J.W.S., Lydon, R. and Moss, M. (2006a), “Computer simulation of filtration process through woven fabrics”, paper presented at Multi-conference in Computational Engineering in Systems Applications, Beijing. Nazarboland, M.A., Chen, X., Hearle, J.W.S., Lydon, R. and Moss, M. (2006b), “Analysing the effect of woven fabric filter on fluid using CFD”, Meeting of the Filtration Society, Loughborough University, Loughborough. Nazarboland, M.A., Chen, X., Hearle, J.W.S., Lydon, R. and Moss, M. (2007), “Effect of different particle shape on the modelling of woven fabric filtration”, J. Info. & Comp. Sci., Vol. 2, pp. 111-8. Peirce, F.T. (1937), “The geometry of cloth structure”, J. Textile Inst., Vol. 28, p. T45. Shin, C. (2006), “Finite element simulation of deep bed filtration”, Chem. Eng. Sci., Vol. 61, pp. 2324-9. Shirato, M., Murase, T., Iwata, M. and Kurita, T. (1986), “Principle of expression and design of membrane compression type filter press operation”, Encyclopaedia of Fluid Mechanics, Gulf Publishing Company, Houston, TX, pp. 905-64. Tiller, F.M. (1975), “Compressible cake filtration”, The Sci. Basis of Filt., Noordhoff, Leyden, pp. 315-97. Ting, K.C., Wakeman, R.J. and Nassehi, V. (2005), “Modelling flow in monofilament cloths – prediction of pressure loss”, paper presented at Filt. Soc. Meeting, May. Vardy, A.E. (1990), Fluid Principles, McGraw-Hill, New York, NY. Wakeman, R.J. (1986), “Theoretical approaches to thickening and filtration”, Encyclopaedia of Fluid Mechanics, Gulf Publishing Company, Houston, TX, pp. 649-83. Corresponding author M.A. Nazarboland can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
A new method of ease allowance generation for personalization of garment design Yu Chen, Xianyi Zeng, Michel Happiette and Pascal Bruniaux
Personalization of garment design 161
Ecole Nationale Supe´rieure des Arts & Industries Textiles, Roubaix, France, and
Royer Ng and Winnie Yu Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Kowloon, Hong Kong, People’ Republic of China Abstract Purpose – The purpose of this paper is to present recent work for optimizing the estimation of ease allowance of a garment using fuzzy logic and sensory evaluation. Design/methodology/approach – The current method first generates a number of fuzzy models each corresponding to one specific key body part and one specific wearer’s movement and then aggregates all the values of ease allowance generated from these fuzzy models using the ordered weighted averaging (OWA) operator. The aggregated ease allowance takes into account geometric measures on all representative human bodies, comfort sensations of wearers related to all movements or actions and different styles of trousers (tight, normal and loose). The weights of the OWA operator can be used to adjust the compromise between the style of garments and the comfort sensation of wearers. The related weights of the OWA operator are automatically determined according to designer’s linguistic criteria characterizing the relationship between wearer’s movements and the features of the garment to be designed. Findings – Based on the optimized values of ease allowance generated from fuzzy models related to different key body positions and different wearer’s movements, the authors obtain a personalized ease allowance, permitting to further improve the wearer’s fitting perception of a garment. The effectiveness of the method has been validated in the design of trousers of jean type. It can also be applied for designing other types of garment. Originality/value – Integration of wearer’s body shapes and human comfort in the design of personalized garments. Keywords Design, Fuzzy logic, Clothing, Sensory perception Paper type Research paper
Introduction Recently, mass customization has made great benefits in many manufacturing sectors (Tseng and Jiao, 2001). It can customize products quickly for individual customers or for niche markets at better than mass production efficiency and speed. In general, mass customization is realized by the use of flexible computer-aided manufacturing systems to produce custom output. In textile and garment industry, enterprises also pay great attention to mass customization and wish to quickly produce a great quantity of personalized garments meeting dynamically changing needs of consumers on garment comfort and style with low production and design cost. The garment design computer aided system presented in this paper has been developed in this background. A garment is assembled from different cut fabric elements fitting human bodies. Each of these cut fabric elements is reproduced according to a pattern made on paper
International Journal of Clothing Science and Technology Vol. 20 No. 3, 2008 pp. 161-173 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810865210
IJCST 20,3
162
or card, which constitutes a rigid 2D geometric surface. For example, a classical trouser is composed of cut fabrics corresponding to four patterns: front left pattern, behind left pattern, front right pattern and behind right pattern. A pattern contains some reference lines characterized by dominant points which can be modified. Of all the classical methods of garment design, draping method is used in the garment design of high level (Crawford, 1996). Using this method, pattern makers drape the fabric directly on the mannequin, fold and pin the fabric onto the mannequin, and trace out the fabric patterns. This method leads to the direct creation of clothing with high accuracy but needs a very long trying time and sophisticated techniques related to personalized experience of operators. Therefore, it cannot be applied in a massive garment production. Direct drafting method is faster and more systematic but often less precise (Aldrich, 1997). It is generally applied in classical garment industry. Using this method, pattern makers directly draw patterns on paper using a pattern construction procedure, implement in a garment CAD system. This construction procedure does not determine the amount of ease allowance, but instead generates flat patterns for any given value of ease allowance. In practice, it is necessary to find a compromise between these two garment construction methods so that their complementarity can be taken into account in the design of new products. To each individual corresponds a pattern whose parameters should include his body size and the amount of ease allowance of the garment. In fact, most of fabrics are extensible and cannot be well deformed. Moreover, the amount of ease allowance of a garment, defined as the difference in space between the garment and the body, can be taken into account in the pattern by increasing the area along its outline. In practice, there exist three types of ease allowance: (1) standard ease; (2) dynamic ease; and (3) fabric ease. Standard ease allowance is the difference between maximal and minimal perimeters of wear’s body. It is obtained from standard human body shape for the gesture of standing or sitting still. This amount can be easily calculated using a classical drafting method (Aldrich, 1997; Ng, 1998). Dynamic ease allowance provides sufficient spaces to wearers having non standard body shapes (fat, thin, big hip, strong leg, . . .) and for their movements (walking, jumping, running, etc.). Fabric ease allowance takes into account the influence of mechanical properties of fabrics of the garment. It is a very important concept for garment fitting. Existing automatic systems of pattern generation or garment CAD systems cannot determine suitable amounts of ease allowance because only standard ease allowance is taken into account. In this case, 2D patterns are generated according to the predefined standard values of ease allowance for any body shapes and any types of fabric. In our previous work (Chen et al., 2004), we developed a fuzzy logic-based method permitting to generate values of ease allowance at key body parts capable of taking into account two aspects: standard and dynamic ease. This fuzzy ease allowance can be more adapted to body measurements and movements of each individual because the corresponding fuzzy model was learnt from a number of selected relevant measures on wearer’s body shapes and sensory evaluations of wearers related to garment comfort.
The amount of fuzzy ease allowance has been shown more efficient than classical methods for generating suitable garment patterns. In this paper, we present our recent work on this topic permitting to further improve the quality of the model related to fuzzy ease allowance generation. Different from the previous method calculating the general ease allowance with only one fuzzy model, the current method first generates a number of fuzzy models each corresponding to one specific key body part and one specific wearer’s movement and then aggregates all the values of ease allowance generated from these fuzzy models using the ordered weighted averaging (OWA) operator. The aggregated ease allowance takes into account geometric measures on all representative human bodies, comfort sensations of wearers related to all movements or actions and different styles of trousers (tight, normal and loose). The weights of the OWA operator can be used to adjust the compromise between the style of garments and the comfort sensation of wearers. The related weights of the OWA operator are automatically determined according to designer’s linguistic criteria characterizing the relationship between wearer’s movements and the features of the garment to be designed. In this case, more flexible and more suitable garment patterns can be obtained using this new method. The corresponding scheme is shown in Figure 1. Basic notations and garment comfort degree The basic notations are formalized as follows. The whole set of body measurements are denoted as BM1, BM2, . . . , BMr. For example, BM1 ¼ waist girth, BM2 ¼ hip girth, and so on. Different parts of human body are denoted as BP1, BP2, . . . , BPm. For example, BP1 ¼ lateral abdominal region, BP2 ¼ femoral triangle, and so on. The comfort degree at each body part can be evaluated subjectively by wearers. We have produced a special sampling jean whose key body parts can be adjustable in order to generate different values of ease allowance. This sample can be used to simulate jeans of different sizes and different styles. In our project, only the normal size is studied. The corresponding ease allowance values at different body parts vary from 2 1 to 8. We select a group of n evaluators having different body shapes. These evaluators or wearers are denoted as WS1, WS2, . . . , WSn. The values of the comfort degree at different body parts are evaluated by these wearers. For each wearer WSi, the body measurements are denoted as BM1(WSi), BM2(WSi), . . . , BMr(WSi). In this case, the body measurements for all wearers constitute a (n £ r)-dimensional matrix. In order to take into account the dynamic ease allowance, we ask each wearer to do a series of movements or actions and evaluate the comfort degree at each body part for each movement. These movements include bending leg, bend waist, open legs at sitting, and so on and they are denoted as M1, M2, . . . , Mh. According to the above definitions, the comfort degree can be formalized by CD(WSi, BPj, Mk). It a function of three variables: wearer, body part and movement. It represents the sensory evaluation provided by the wearer WSi at the body part BPj ( j ¼ 1, . . . , m) when he/she does the movement Mk. In our experiments, we select 20 (n ¼ 20) wearers for evaluating comfort degrees at different body parts of the garment sample of normal size. The total number of body movements is 14 (h ¼ 14). The values of the comfort degrees given by wearers vary between 0 and 8, in which 0 represents extremely uncomfortable, 2 very uncomfortable, 4 normal, 6 very comfortable and 8 extremely comfortable.
Personalization of garment design 163
IJCST 20,3
Generation of body part A for all actions Extract relevant BM using PCA
Body measurements
164
Relevant BM
Sensory evaluation of body part A for action I
FLC of body part A for action I
Ease of body part A for action I W1
Body measurements
Extract relevant BM using PCA
Relevant BM
Sensory evaluation of body part A for action II
FLC of body part A for actionII
Ease of body part A for action II
W2
Ease of body part A OWA aggregation
Wi
Product new pattern
W14
Body measurements
Extract relevant Relevant BM BM using PCA
Sensory evaluation of body part A for action XIV
Figure 1. General scheme for generating fuzzy ease allowance
Ease of FLC of body part A body part A for action XIV for action XIV
Generation of body part B for all actions
Ease of body part B
Generation of body part X for all actions
Ease of bodypart X
Selection of relevant body measurements In garment design, a great number of geometric measurements can be taken from human bodies. However, for a specific garment, only a very small set of measurements is relevant to the corresponding comfort degree. Then, we should only take this set of body measurements as input variables of the fuzzy model. The relevant body measurements can be selected by garment designers using their professional experience. In practice, this personalized knowledge is not normalized and each designer selects his/her own relevant body measurements, different from others. Moreover, for a specific garment, it is possible that some important body measurements are neglected by designers because they have no complete knowledge on all concerned human body parts related to their movements. In our fuzzy model, the relevant body measurements are first selected using the data sensitivity criterion. Then, these selected variables are validated using the general
knowledge of garment design. The principle of the data sensitivity criterion is given as follows: . if a small variation of a body measurement corresponds to a large variation of the garment comfort, then this body measurement is considered as a sensitive variable. . if a large variation of a body measurement corresponds to a small variation of the garment comfort, then this body measurement is considered as an insensitive variable. Moreover, in practice, body measurements related to uncomfortable feeling of wearers seem to be more important than those related to comfortable feeling. According to this principle, we define, for a specific body part BPk and a specific movement Ml, an importance coefficient Pij in our sensitivity criterion. We have: P ij ¼
r ðCDðWSi ; BPk ; M l Þ þ CDðWSj ; BPk ; M l ÞÞ;
P where r is a constant so that i–j P ij ¼ 1: The value of Pij is big if the comfort degrees of the two wearers WSi and WSj are low values. In this case, both wearers have uncomfortable feeling at the body part BPk of the sample of normal size related to the movement Ml. The value of Pij is small if the comfort degrees of WSi and WSj are high values. In this case, both wearers have comfortable feeling at the body part BPk related to the movement Ml. In any cases, the value of Pij is inversely proportional to the sum of the comfort degrees of WSi and WSj. For a specific body part BPk of the sample of normal size, the sensitivity criterion for selecting the most relevant body measurements is denoted as S(BMi, BPk, Ml). It is defined by: SðBMi ; BPk ; M l Þ ¼
XP st jCDðWSs ; BPk ; M l Þ 2 CDðWSt ; BPk ; M l Þj s–t
jBMi ðWSs Þ 2 BMi ðWSt Þj
where i [ {1, . . . , r}, k [ {1, . . . , m}, l [ {1, . . . , h} and s, t [ {1, . . . , n}. From this definition, we can see that the value of the sensitivity S is big if a small variation of BMi for different wearers causes a big variation of their comfort feelings (from an uncomfortable level to a comfortable level or from a comfortable level to an uncomfortable level). For any specific body part of the studied trouser sample, all body measurements BM1, BM2, . . . , BMr can be ranked according to this criterion and the elements having the highest ranks are considered as the most relevant body measurements, which will be taken as input variables in the fuzzy model. Using this criterion, for the gluteal region and the trouser of normal size, we obtain eight body measurements with the hig figure hest ranks. These parameters can be classified into two classes as follows: (1) Vertical type. Waist to hip, out leg, curved front body rise. (2) Girth type. Thigh girth, waist girth, half back waist, hip size, half back hip. These body measurements are shown in Figure 2.
Personalization of garment design 165
IJCST 20,3
166
half back waist waist girth waist to hip hip size
half back hip
thigh girth
Figure 2. Relevant body measurements related to the gluteal region and the trouser of normal size
curved front body rise
out leg
Note: Reproduced from the only available original
In practice, the relevant body measurements obtained from this sensitivity criterion are not always consistent with human knowledge on garment design. This is because the correlation between selected body measurements and related body part is not taken into account in this criterion. In order to improve the quality of the selection criterion, we introduce a garment knowledge-based criterion and combine it with the sensitivity criterion. This new criterion can effectively decrease the influence of body measurements having weak correlation with related body parts. The garment knowledge-based criterion is generated from a mxr-dimensional matrix BPBM, defined by experts in garment design. Each element BPBM(i, j) of this matrix characterizes the relationship between the body part BPi, i [ {1, . . . , m} and the body measurement BMj, j [ {1, . . . , r}. Its value is defined according to the following principles: . If the body measurement BMj is taken on the body part BPi or passes through BPi, then we consider that the correlation between BMj and BPi is strong and BPBM(i, j) ¼ 1. For example, in Figure 3, we have BPBM(lumbar, waist girth) ¼ 1. . If the body measurement BMj is close to the body part BPi but there does not exist any intersection between them, then we consider that the related correlation is less strong and BPBM(i, j) ¼ 0.5. For example, in Figure 3, we have BPBM(gluteal, waist girth) ¼ 0.5. . If the body measurement BMj is far from the body part BPi, then we consider that the related correlation is weak and BPBM(i, j) ¼ 0.1. For example, in Figure 3, we have BPBM(neck, waist girth) ¼ 0.1. According to these principles, we obtain values of the matrix BPBM for some important body measurements and body parts in Table I. By combining the criteria of sensitivity and garment knowledge, we define a new selection criterion of body measurements, denoted by SBMBP(BMi, BPk., Ml), as follows:
Personalization of garment design
Neck grith Neck
167
chest grith pectoral
lumbar
waist grith
gluteal
hip grith
thigh girth anterior of thigh anterior of knee
knee girth
Figure 3. Some important body measurements and body parts
Note: Reproduced from the only available original
Neck Pectoral Lumbar Gluteal Anterior of thigh Anterior of knee
Neck girth
Chest girth
Waist girth
Hip girth
Thigh girth
Knee girth
1 0.5 0.1 0.1 0.1 0.1
0.5 1 0.5 0.1 0.1 0.1
0.1 0.5 1 0.5 0.1 0.1
0.1 0.1 0.5 1 0.5 0.1
0.1 0.1 0.1 0.5 1 0.5
0.1 0.1 0.1 0.1 0.5 1
S BMBP ðBM i ; BPk ; M l Þ XP st jGCDðWSs ; BPk ; M l Þ 2 GCDðWSt ; BPk ; M l ÞjBPBMðk; i Þ ¼ jBMi ðWSs Þ 2 BMi ðWSt Þj s–t In our experiments, the body measurements selected using this criterion have been proved to be more relevant and more interpretable than the former criterion.
Table I. Some values of the matrix BPBM
IJCST 20,3
168
Fuzzy model construction and ease allowance aggregation Fuzzy model construction In the last section, we select a number of the most relevant body measurements according to the criterion combining the data sensitivity and the garment knowledge. However, in the fuzzy modeling procedure, the number of these input variables is still too large related to the number of learning data obtained from 20 wearers. In this case, we apply the principal component analysis (PCA) (Fukunaga, 1990) before the fuzzy rules extraction in order to further reduce the input space. In this section, we wish to extract two first components from the relevant body measurements selected previously. From the garment design knowledge, we know that there exist a very weak correlation between the body measurements of the vertical type and those of girth type. For simplicity, the body measurements of each class (vertical type or girth type) are independently projected into 1D space using PCA. Then, we obtain two extracted variables: the vertical body measurement (x1: VBM) and the girth body measurement (x2: GBM). These two variables as well as the comfort degree (x3: CD) obtained from sensory evaluation of wearers are taken as input variables of the fuzzy model. The ease allowance for the corresponding body part and wearer’s movement, denoted as y, is taken as output variable of the model. In this way, we obtain a set of fuzzy models each generating an ease allowance related to one specific body part and one specific wearer’s movement. In each fuzzy model, in order to extract significant fuzzy rules, we have to transform measured or evaluated numerical values of the input and output variables into linguistic values. The linguistic values of x1 (VBM), x2 (GBM) and y are: {very small (VS), small (S), normal (N), big (B), very big (VB)}. The linguistic values of x3 (CD) are {very uncomfortable (VUC), uncomfortable (UC), normal (N), comfortable (C), very comfortable (VC)}. The corresponding learning input/output data, measured and evaluated on n different wearers, are denoted by {(x11, x12, x13; y1), . . . ,(xn1, xn2, xn3; yn)}. In our experiments, we have n ¼ 20. Mamdani and Assilian (1975) method is used for defuzzification. The fuzzy rules are extracted from these input/output learning data. For each input variable xi (i ¼ 1, 2, 3), the parameters of its membership functions are obtained using the fuzzy c-means clustering method (Bezdek, 1981). This method permits to classify the learning data {x1i, . . . ,xni} into five classes, corresponding to the five fuzzy values of xi. The fuzzy rules for estimation of ease allowance are extracted from the learning data using the method of antecedent validity adaptation – AVA (Chan and Rad, 2002). Ease allowance aggregation Each fuzzy model generates a value of ease allowance, denoted as y(BPi, Mj). It is related to one specific body part BPi, i [ {1, . . . , m} and one specific wearer’s movement Mj, j [ {1, . . . , h}. Then, we need to aggregate all values of ease allowance related to all wearer’s movements. In garment design, the aggregated ease allowance will be used to modify the pattern related to the body part BPi. In this procedure, an OWA operator is used to realize this aggregation (Yager, 1988). More details on this operator are presented as follows. An OWA operator of dimension h is a mapping F: Rh ! R characterized by an h dimensional vector; W, called the weighting vector, such that its components lie in the unit interval and sum one, wj [ [0.1] and w1 þ · · · þ wh ¼ 1. The aggregation of values
of ease allowance for different wearer’s movements performed by this operator is defined as: Fð yðBPi ; M 1 Þ; . . . ; yðBPi ; M h ÞÞ ¼ W T B where B is the vector of ordered values of ease allowance related to different movements and its jth component bj is the jth largest of these values. If index is a function such that index( j) is the index of the jth largest of the values of ease allowance related to different wearer’s movements, then we have bj ¼ y(BPi, Mindex( j)). The values of the components of W permit to set up a compromise between textile comfort and fashion style of garment design. If w1 ¼ 1 and wj ¼ 0 for any j [ {2, . . . , h}, the aggregation result F corresponds to the biggest value of ease allowance for all wearer’s movements (garment of loose style). The general ease allowance obtained in this case gives a very comfort feeling for any movement but the space between body and garment is probably too large and the corresponding fashion style will be far from being optimized. If wh ¼ 1 and wj ¼ 0 for any j [ {1, . . . , h 2 1}, the aggregation result F corresponds to the smallest value of ease allowance for all wearer’s movements (garment of tight style). The general ease allowance obtained in this case will make the garment too tight for some movements. In practice, the values of W should be selected between these two extreme cases by searching for a suitable compromise between fashion style and garment comfort feeling. For example, if we take w1 ¼ w2 ¼ . . . ¼ wh ¼ 1/h, we obtain a garment of normal style. In practice, the choice of these OWA weights can be determined by garment designers according to design criteria and trends of garment market. Results and discussion The proposed fuzzy models as well as the procedure for ease allowance aggregation have been validated in the design of trousers of jean type. The corresponding results and related analysis are given in this section. The learning base was built using data measured on 20 Hong Kong wearers with different body shapes (from fat to thin). These data include the relevant body measurements, the comfort degree and the value of ease allowance obtained from the adjustable trouser sample (from loose type to tight type). Based on these data, we obtain the fuzzy model using the method of AVA. The most significant fuzzy rules of this model are given as follows: . IF GBM ¼ B AND VBM ¼ N AND CD ¼ C THEN ease ¼ S . IF GBM ¼ S AND VBM ¼ VS AND CD ¼ VUC THEN ease ¼ VB . IF GBM ¼ N AND VBM ¼ N AND CD ¼ C THEN ease ¼ N . IF GBM ¼ B AND VBM ¼ N AND CD ¼ C THEN ease ¼ S . IF GBM ¼ B AND VBM ¼ N AND CD ¼ N THEN ease ¼ B In the following example, only three movements or actions (bending leg, crawling and wide open legs) are considered by wearers. In this case, we have h ¼ 3. For each of these actions, a fuzzy model related to the lumbar body part is built from the body measurements for generating new values of ease allowance. In these models, the comfort degree is set to be 5 (comfortable). The corresponding results for three wearers of different body shapes are given in Table II.
Personalization of garment design 169
IJCST 20,3
170
From Table II, we can see that ease allowance generally increases with body shape, i.e. bigger values of ease for fat body shape and smaller values of ease for thin body shape. For each wearer, the values of ease are different when he or she takes different actions. This difference is very big for fat wearers (almost 2 cm), a little big for thin wearers (around 1.5 cm) and very small for normal body shapes. The results generated from the fuzzy models can be aggregated using an OWA operator. In this procedure, we give three weighting vectors each corresponding to one style of trousers, i.e. tight style: W ¼ ð0 0 1ÞT , normal style: W ¼ ð0:33 0:33 0:33ÞT , loose style: W ¼ ð1 0 0ÞT . The aggregated values of ease allowance related to the lumbar body part are shown in Table III. The classical ease allowance is a fixed standard value (4 cm) and does not take into account the dynamic aspect and the fabric aspect. Its calculation procedure is given in Aldrich (1997) and Ng (1998). Compared to the classical ease allowance, the method proposed in this paper is more flexible and more sensitive to variations of body shapes and trouser styles. For thin body shapes, the values of ease are generally smaller than the standard value. For fat body shapes, the values of ease are generally bigger than the standard value. Moreover, for different trouser styles (loose, normal and tight),
Body shape Table II. Values of ease allowance calculated from the fuzzy model
Wide open legs
Crawling
3.10 4.72 5.17
4.54 4.71 5.05
3.45 4.43 6.91
Thin Normal Fat
Normal body shape
Fat body shape
Thin body shape
Table III. Comparison of ease allowance (EA) between different methods
Bending leg
Waist girth
Classical EA
Tight EA
Normal EA
Loose EA
72.5 78.5 68 73.7 73.9 88.5 77 87.5 85.5 88 89 93.5 76 86 72.5 71 71 68 68 68.8
4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4
4.14 4.29 3.72 4.57 4.14 6.33 4.43 4.72 4.04 5.75 6.48 5.05 3.87 5.62 4.00 3.71 3.10 3.67 3.70 3.10
4.32 4.49 3.80 4.62 4.44 6.33 4.57 5.18 5.04 6.11 6.98 5.65 4.54 5.87 4.51 3.81 3.67 3.85 3.85 3.70
4.51 4.68 3.96 4.75 4.74 6.48 4.72 5.61 5.63 6.46 7.44 6.91 4.97 6.28 4.98 4.06 4.44 4.02 4.02 4.54
the values of ease are different. A compromise between garment comfort and fashion aspect has been set up in the proposed method. A comparison between the classical patterns and the fuzzy patterns we generated is shown in Figure 4. For tight trouser style, the fuzzy pattern is tighter than the classical pattern. For loose trouser style, t fuzzy pattern is looser than the classical pattern. For normal style, these two patterns are rather similar. The concept of style is then taken into account in the fuzzy pattern. The effectiveness of the method proposed in this paper can also be validated by comparing comfort degrees between trousers made of classical patterns and fuzzy patterns for three wearers having different body shapes. The corresponding results are shown in Table IV.
Thin body shape tester (tight type)
Personalization of garment design 171
Fat body shape tester (loose type)
Normal body shape tester (normal type) 30
20
20
10
10
0
0
–10
–10
–20
–20
20 10 0 –10 –20
–30
–30
–30
–40
–40
–40
–50
–50
–50
–60
–60
–60 –70 –70
–10
0
10
–20 –10
Figure 4. Comparison of patterns between the classical method and the proposed method
–70 0
10
20
–20 –10
0
10
20
dotted line: fuzzy pattern
solid line: classical pattern
Body shape
Thin (waist 66.8 cm)
Comfort degree Bending legs Wide open legs Crawling Average
CDCP 2.33 3.67 3 3
CDFP 3.67 4.33 3.33 3.77
Normal (waist 78.5 cm) CDCP 3.33 3.67 3.67 3.55
CDFP 3.33 4 3.33 3.55
Fat (waist 93.5 cm) CDCP 2 1 3 2
CDFP 3.67 3.33 5.33 4.11
Table IV. Comparison of comfort degrees between the classical method and our method
IJCST 20,3
172
In Table IV, CDCP and CDFP denote the comfort degrees of the trousers made of classical patterns and fuzzy patterns, respectively. Their values, varying from 0 – extremely uncomfortable to 6 – very comfortable, are obtained by subjectively evaluating three important body parts, i.e. lumbar, gluteal and thigh, and then calculating their averages. For the fat and the thin wearers, the comfort degrees of the trousers made of fuzzy patterns are effectively improved related to the trousers made of classical patterns. However, for the normal wearer, the comfort degrees remain almost invariant. It means that our method is more significant in the garment design of non standard body shapes, especially for fat wearers. Conclusion This paper proposes a new method for optimizing garment design by estimating more suitable values of ease allowance using fuzzy techniques and sensory evaluation of wearers. A criterion combining the data sensitivity and garment knowledge has been proposed for selecting the most relevant body measurements. Then, these selected body measurements are separately projected into two 1D subspaces using PCA in order to generate two features: vertical body measurement and girth body measurement. These two features as well as the comfort degree evaluated using sensory evaluation of wearers constitute the input variables of the fuzzy model related to one specific body part and one specific wearer’s movement. Using the method of AVA, we extract the corresponding fuzzy rules from the learning data measured and evaluated on 20 wearers. Using an OWA aggregation operator, we obtain a general value of ease allowance related to all wearer’s movements. A compromise between textile comfort and garment style is set up in the aggregated value of ease allowance as well as the corresponding garment pattern. In this way, the proposed method can effectively improve the quality of garment pattern design. Compared with the classical design methods, the proposed method is more flexible and more adaptive to the current competitive international garment market, which requires more and more personalized products meeting human comfort and style variation. References Aldrich, W. (1997), Metric Pattern Cutting for Men’s Wear, 3rd ed., Blackwell Science Ltd., Cambridge. Bezdek, J.C. (1981), Pattern Recognition with Fuzzy Objective Function Algorithms, Plenum Press, New York, NY. Chan, P.T. and Rad, A.B. (2002), “Antecedent validity adaptation principle for fuzzy systems tuning”, Fuzzy Sets and Systems, Vol. 131, pp. 153-63. Chen, Y., Zeng, X., Happiette, M., Bruniaux, P., Ng, R. and Yu, W. (2004), “Estimation of ease allowance of a garment using fuzzy logic”, paper presented at International Conference FLINS’04, Blankenberghe, September 1-3. Crawford, C.A. (1996), The Art of Fashion Draping, 2nd ed., Fairchild Publications, New York, NY. Fukunaga, K. (1990), Introduction to Statistical Pattern Recognition, 2nd ed., Academic, San Diego, CA. Mamdani, E.H. and Assilian, S. (1975), “An experiment in linguistic synthesis with a fuzzy logic controller”, International Journal of Man-Machine Studies, Vol. 7, pp. 1-13.
Ng, R. (1998), “Computer modeling for garment pattern design”, PhD thesis, The Hong Kong Polytechnic University, Kowloon. Tseng, M.M. and Jiao, J. (2001), “Mass customization”, in Salvendy, G. (Ed.), Handbook of Industrial Engineering, Technology and Operation Management, 3rd ed., Wiley, New York, NY. Yager, R.R. (1988), “On ordered weighted averaging aggregation operators in multi-criteria decision making”, IEEE Trans. on SMC, Vol. 18, pp. 183-90. Corresponding author Xianyi Zeng can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Personalization of garment design 173
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,3
The sensory research on the style of women’s overcoats Ying Wang, Yan Chen and Zhi-ge Chen College of Material Engineering, Soochow University, Suzhou, People’s Republic of China
174 Abstract
Purpose – The research in this paper aims to investigate the perception created by clothing style and semantic space to describe this perception. The results of study could be applied to establish relation between customers’ feelings and design elements. Design/methodology/approach – Women’s overcoats were chosen as research objects. The technique of sensory engineering was applied to investigate the customers’ feelings and demands related to product images. Card system, shape analytical method, and regression analysis were applied in this research. Findings – Six word-pair were selected to establish the semantic space. The product elements space with seven items and 25 categories was established. The relation between two spaces could be quantified according to the principle of sensory engineering. Research limitations/implications – The findings can be applied to women’s overcoat and similar clothing for customer-orientated design. Originality/value – The theory of sensory engineering was applied for practical application purposes. Some useful parameters were obtained to explain the reliability of the research results and influences of design elements on perception judgment. Keywords Design, Clothing, Women, Sensory perception Paper type Research paper
International Journal of Clothing Science and Technology Vol. 20 No. 3, 2008 pp. 174-183 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810865229
Introduction Sensory engineering (Figure 1) was founded in Japan 30 years ago and defined as Kansei engineering or Kansei ergonomics. It can be applied to investigate the relationship between customer’s feeling and demands with product function and appearance (Mitsuo, 2002). This technique has been widely used in consumptive products development to meet the trend of market change towards the consumer oriented. Garments are typical consumptive products. Success or failure of the garment design depends mainly on customer’s feelings and demands. It is found that customers have difficulties in distinguish the differences of same kind of garments because of their similar functions (Chen, 2000). As a result, their decisions depend increasingly and greatly on subjective factors, such as feelings, images, impressions and demands of the products. From the point of view, garment is the product with tangible and intangible characteristics (Wang, 2002). Sensory engineering is one of the effective tools to research the intangible part of garment, especially psychological feelings and images. Since, it can translate customer’s feelings into concrete product parameters and provides support for further product design. Therefore, the aim of this paper is to find out and represent numerically the relationship between design elements and perception terms about a kind of garment. The style of women’s overcoats was chosen as the research object, for the reason that they are characteristic of simple structures and change infrequently. It is expected to
Research on the style of women’s overcoats
Choice of object
The semantic space
Update
The product elements space
Synthesis
175
Update
Test of Validity
Modeling
translate feelings and impression at these garments into the design domain on the basis of this research. Principle of research Sensory engineering makes it possible to translate customers’ impressions, feelings and demands about products or concepts into concrete design parameters. During past 30 years, this technique has been developed into six subdivisions to establish the relation between user’s impressions and design specification for consumption products. This technique can be applied to identify the design elements, and also to provide models, interferences, databases and intelligent software for design process. Owing to its capability of translating perception, sensory engineering has been widely used to various fields, such as eye glass frame, airplane interior, Japanese traditional crafting, image retrieval and so on. According to the principle and method of sensory engineering, customers’ feelings and images on women’s coat are investigated. On the basis of investigation, these subjective perceptions are transferred into concrete design elements. Figure 1 shows out the research procedures, which can be specified as follows (Simon et al., 2004a). . Selecting research objects. The objects in this work deal with not only garment products, but also service and their combination. . Establishing the semantic space. This is comprehensive work. The adjectives which are used to describe the perceptions on chosen products – women’s overcoats are collected. These adjectives are analyzed and the synonyms are eliminated to reduce the number of adjectives used. These adjectives are then specifically defined to establish semantic space used in the phase of synthesis. . Establishing the product elements space. The different styles of women’s overcoats are collected. The elements of this kind of garment are investigated and selected.
Figure 1. A framework for sensory engineering
IJCST 20,3 .
176
.
.
Only the elements which have obvious influences on customers’ feelings are retained. On the basis of these work, the typical products which contain these elements are chosen as the representatives of women’s overcoats to be used in synthesis. Synthesis. In this step, the relationship between two spaces, the semantic space and the product elements space, are established by using of statistic method. Test and update. Being influenced by culture, economy, fashion, clime and other factors, customers’ feelings and demands are changing forever. Therefore, the validity of result of research should be tested and be updated correspondingly. Modeling. The data acquired during synthesis can be presented by mathematical equation: ySensory ¼ f ðproduct elementsÞ
Semantic space and product elements space Semantic space A large number of adjectives used to describe the feelings at women’s overcoats were collected by referring to magazines, newspapers, textbooks, dictionaries and other information resources related to textiles and garments. 316 words were selected as the analyzing basis. It had been found that some words were less relative with this research. After eliminating these words, 180 words were preserved for the next steps of research. The number of adjectives, used as perception terms, should be reduced reasonably. Too many perception terms will make quizzee tired or bored and will influence the result of survey, while too little amount of perception terms will reduce the reliability of survey. Therefore, decreasing the 180 words was significant for the following survey. The card system was applied for this purpose. The words collected were showed to garment designers and other professionals. They were asked to group words according to the meanings first. After this step, 45 groups were obtained from 180 words. Then, these groups were further classified according to the intimate degrees of these groups. The number of classified groups depended on the practical condition of women’s’ overcoat as well as the research results obtained in this area. Finally, these groups were paired and one word was selected from each group as the “representative” of all the words in this group. The six word pairs were obtained in this research to define the different aspects of perception aroused by the different styles of women’s overcoats. The word pairs and their definition in this research were listed in Table I.
Table I. Word-pairs used for perception evaluation for women’s overcoat
Word pairs
Definition
Female-male Fashionable-traditional Simple-complicated Formal-informal Unique-usual Mature-young
Sex orientation Duration of prevailing Structure character The occasion of wearing Acceptant degree Age orientation
Product elements space The shape analytical method was used to establish product property space. The whole shape of product is considered as a combination of the product’s elements. Shape elements are re-arranged and new shapes are created. The principle of shape analytical method can be applied by following steps (Xu, 2004): the product shape is divided into several independent factors; the elements of first level are found out for each independent factor; and then, a new shape is created by re-arranging these elements; finally, the new shapes are evaluated. Firstly, the style of women’s overcoats was divided into seven independent factors, namely items. Then each item was analyzed and several elements were found out. For instance, the item “Silhouette” was composed of “X-line,” “H-line,” and “A-line”. By the same way, the elements, which influence Sensory judgment, were selected and design elements space is established. The space contained seven items and 25 categories, which were marked with letters and numbers, respectively, as listed in Table II. The seven items were represented by Capital “A” to “G” and categories of each item were expressed by number 1, 2,. . . For example, “X-line” was substituted by “A1”. This kind of expression was also used in the following tables and figures.
Research on the style of women’s overcoats 177
Synthesis Garment samples More than 300 women’s overcoats were collected in advance. These are all characterized of simple structure, less ornamental details such as pleats and laces, and more usages of some structures, like raising neckline, increasing coat length, tightening waist and so on. The similar designs were grouped together and only one design was selected from each group. A total of 20 styles of women’s overcoats, which were the representatives of the collection, were obtained. These 20 typical styles were combinations of design elements listed in Table II. They would be surveyed as samples during subjective evaluation. Subjective evaluation The relationship between the semantic space and the product elements space was studied by questionnaire survey and quantification analysis. The semantic differential method, developed by Osgood, is a typical procedure to translate customers’ feelings and images into proper numerical data (Simon et al., 2004b). According to the SDM, the questionnaire was composed of six word-pair and Items
Categories
Silhouette A Waist design B Coat length C Collar D
X-line A1 High B1 Short C1 Collarless D1
H-line A2 Middle B2 Middle-long C2 Stand collar D2
Sleeve E Pocket F
Set-in sleeve E1 Patch pocket F1
Raglan E2 Welt pocket F2
Facing design G
Single breast G1
Double breast G2
A-line A3 Low B3 Long C3 Turnover collar D3 Structure pocket F3 Zipper G3
Non-waist B4 Lapel D4 None F4 Hided facing G4
Collar with hat D5 Table II. The design elements space
IJCST 20,3
178
20 well prepared samples. And it used numerical scare of seven-grade serial numbers for evaluating perception relationship between samples and perception terms. The evaluation scale and answer sheet is shown in Figure 2. There were 70 volunteers participating in the questionnaire survey, included 36 males and 34 females. These volunteers included fashion designers, university students and lecturers involved in clothing design and engineering. The garment samples were displayed to the participants on the project screen by PowerPoint pictures. The time interval of the slides exchange was 10 s. As shown in Figure 3, the front and the back of sample were displayed in the picture, and the characteristics of style were listed on the same picture. The participants were asked to evaluate these garments and score the each word-pair, respectively. After questionnaire survey, the averages scores of word-pair for each sample garment were calculated and would be used in the quantification analysis of the garment perception.
Kansei word
1
2
Extreme Very
3
4
5
6
7
Kansei word
Normal Normal Extreme Central Very Questionnaire
Sample No Word-pair female-male
1
2
3
4
5
20
simple-complicated fashionable-traditional
Figure 2. Questionnaire of sensory word-pair
formal-informal unique-usual mature-young
The 1th sample
Style characteristic: X-line middle-long middle waist design lapel raglan double breast patch pocket
Figure 3. Display of one sample
The front
The back
Quantification analysis The design elements were quantified into numerical forms by using of Quantification Theory (Su and Li, 2005; Wu et al., 1992; Yang et al., 1999). Supposing there are j items and k categories, the responses of the category k of the item j for sample i is d( j, k) and can be defined: ( 1; where a sample i corresponds to item j and category k; dið j; kÞ ¼ 0; otherwise By the above function, the qualitative designing elements, such as X-line, single breast, collar with hat and so on, could be transferred into numerical forms by using “0” and “1”. The design elements are independent variables X, and the evaluation scores of word-pair are dependent variables Y. A linear relationship between design elements and word-pairs can be established by following multiple linear regression function: _ yi
¼
cj r X X _ ajk di ð j; kÞ
ð1Þ
j¼1 k¼1
Here, j ¼ 1, 2, . . . r (r is the total number of items) and k ¼ 1, 2, . . . cj (cj is the total number of category for item j) _ In the above equation, ajk is regression coefficient and is used as category scores in this paper. It depends on category k of the design item j only. 1i is the error of random sampling i. In order to relate to practical cases, a constant should be added to equation (1), which can be expressed as follow: _ yi
¼ y þ
cj r X X
a*jk di ð j; kÞ
j¼1 k¼1
Here, y is the average of criterion variable and can be calculated with: Pn i¼1 yi y ¼ : n a*jk is standard coefficient. It can be proved that there exists a transformational _ equation between ajk and a*jk . That is: _
c
a*jk ¼ ajk 2
j 1X nil a^ il n l¼1
ð2Þ
nil is the number of response of the item l within the category j among all of the samples, and there must exist the following relation: cj X l¼1
nil ¼ n
Research on the style of women’s overcoats 179
IJCST 20,3
180
By the above equations, the relationship between design elements and perception terms can be established and can be expressed as mathematical model. The standard coefficients, a*jk , are important parameters. They are called category score in the Quantification Theory, which show degree of contributions of categories. Besides category score, there are other two useful parameters used to assess model and analyze its results. R 2 is an important index to judge the precision of model. It is defined by: P_ s _y ð y 2 y Þ2 2 R ¼ ¼P i : sy ð yi 2 y Þ2 The R 2 is the coefficient of determination. It can be used to evaluate the reliability of the regression model and to decide if this model is able to explain the independent variables. Item range can present the influence of item on perception. It is obtained from the distance between the largest and the smallest category score, namely: rangeð j Þ ¼ max a^ jk 2 min a^ jk ; 1#k#r j
1#k#r j
j ¼ 1; 2; . . . ; m
ð3Þ
According to the above-mathematical principle and method, response values of each sample and the averages of evaluating score were processed by multiple linear regression tool of Statistical Analysis System. Constant, R 2 and regression coefficient are obtained by SAS. Regression coefficient is standardized by equation (2), and then category score is obtained. Item range is calculated by equation (3). Result and analysis The results of regression analysis, including R 2, category scores item ranges and constants are listed in Table III. From these data, some analyses are obtained as follows. R 2 is one of important parameters to show the reliability of statistical result (Simon et al., 2004b). Generally, the statistical result is considered to be reliable and can be accepted if R 2 .0.7. In this research, the R 2 for six word-pair were higher than 0.9. It can be concluded that the statistical results are higher reliable. And these six mathematical models could be applied to predict perception score of other women’s overcoats. The category score for each sensory word-pair can explain the correlative degree between design elements and perception terms. The plus category score means the plus perception, such as “male” “complicated” and “traditional.” Contrarily, the minus category score means the minus perception, like “female,” “simple,” and “fashionable.” If the value is close to “0,” it can be inferred that there is hardly any relationship between design elements and word-pair. Besides, the higher absolute value of category score indicates the closer relationship between the categories and the word pairs. As showed in Figure 4, among the three categories of “silhouette” the “X-line” gets the lowest score, 20.762, which shows it is the most female design element. Contrarily, the “A-line” gets the highest score, 1.811, which means that it is the most male one. For “sleeve,” the category score of “Set-in sleeve” and “Raglan” were 0.030 and 0.046, respectively. They were very close to “0”. Thus, these two kinds of sleeve have little
Constant R2
G
F
E
D
C
B
A
Items
A1 A2 A3 B1 B2 B3 B4 C1 C2 C3 D1 D2 D3 D4 D5 E1 E2 F1 F2 F3 F4 G1 G2 G3 G4
Categories 20.762 0.179 1.811 21.127 21.189 0.408 1.488 21.763 0.953 0.268 20.625 21.571 0.502 1.372 21.854 0.030 20.046 0.964 20.702 0.470 20.740 20.115 21.604 1.821 0.673 3.191 0.985 3.425
1.704
0.076
3.226
2.716
2.677
2.574
Female-male Score Range 2.001 21.885 20.977 0.934 2.087 3.636 23.881 3.922 21.824 20.855 2.155 1.169 21.104 21.986 4.774 1.176 21.764 0.879 1.514 22.047 20.811 20.908 1.769 0.296 20.675 3.645 0.988 2.677
3.561
2.941
6.760
5.746
7.518
3.887
Simple-complicated Score Range 2 1.254 1.369 0.112 2 0.436 2 0.239 2 2.502 1.526 2 2.096 0.968 0.463 2 0.659 0.656 0.599 2 0.355 2 1.381 2 0.607 0.911 2 0.972 2 0.354 0.855 0.678 0.834 2 0.168 2 1.053 2 0.367 3.858 0.997 1.887
1.827
1.518
2.037
3.064
4.028
2.622
Fashionable-traditional Score Range 0.591 21.127 1.233 20.072 20.345 3.578 21.197 3.423 21.915 20.464 1.896 21.767 20.157 0.029 2.021 1.007 21.511 3.140 0.020 23.834 0.839 22.290 21.071 2.164 2.979 4.407 0.989 5.270
6.975
2.518
3.787
5.337
4.775
2.360
Formal-informal Score Range 22.183 2.264 0.510 21.457 20.620 25.186 3.587 24.769 2.399 0.882 22.795 20.705 1.845 1.101 24.635 21.327 1.991 22.138 21.483 3.191 0.908 1.616 20.594 21.027 21.051 3.849 0.999 2.667
5.328
3.318
6.480
7.168
8.773
4.447
Unique-usual Score Range
0.645 21.585 2.291 0.284 20.073 2.972 21.373 2.379 20.998 20.614 1.336 21.196 0.356 20.707 2.108 1.026 21.540 1.610 0.060 21.743 0.076 21.631 20.643 2.945 1.160 3.622 0.996
4.576
3.352
2.566
3.304
3.377
4.345
3.876
Mature-young Score Range
Research on the style of women’s overcoats 181
Table III. The data acquired from statistical analysis
IJCST 20,3
Category Scores and Item Ranges of Word-Pair "Female-Male" 2.5
4
Figure 4. Category scores and item ranges of word-pair “female-male”
5 42
6
2.5
0
2
4
70
1.
–0.5
1.5
–1 A1 A2 A3 B1 B2 B3 B4 C1 C2 C3 D1 D2 D3 D4 D5 E1 E2 F1 F2 F3 F4 G1 G2 G3 G4
Item Range
71 2.
67 2.
3
2.
57
4
7
1
Category Score Item Range
3.5
3.
6 3.
22
1.5
0.5
Female
182
Category Score
Male
2
1
–1.5 –2 –2.5
0.5 A
B
C
D
E
6
0
7 .0
F
G 0
correlative degree with the perception of “female-male”. The other items can be analyzed in the same way. Based on these analyses, if a women’s overcoat would be designed to have highly female perception, its design elements could be arranged by “X-ling” “Middle Waist Design,” “Short,” “Collar with Hat,” “Raglan,” “With no Pocket,” and “Double Breast”. Item range for word-pair can explain the influence of this item on perception judgment. Generally, the higher value of item range indicates the closer relationship between the item and perception term. Take the word-pair “female-male” for example; the descent range order of its seven items is front, collar, coat length, waist design, silhouette, pocket and sleeve. It can be inferred that “Facing design” (3.425) has the greatest influence on perception judgment, the second is “Collar” (3.226), and “Sleeve” (0.076) has less influence. Therefore, if the perception of “female-male” wants to be changed, transformations of front would be more effective than sleeve’s. Analyses on other category scores and item ranges could be followed by the same way. The result of these analyses would be helpful and could be put into practical use in product design. Conclusions The relationship between perception and design elements of the style of women’s overcoats was studied and discussed by applying the principle of sensory engineering. Some useful parameters were also obtained. R 2, category score and range were used to indicate the reliability of the regression and influences of the every category and each item on perception judgment. The research method of this paper can be directive for other researches on the relationship between feelings and demands of customers and design elements of any kind of garment to follow. In addition, the results of this research can be widely used for both customers and clothing manufacturers. It can help customers to search clothing what they want. It can guide designers and manufactures to develop more competitive garments in the customer-oriented market. Only the style was discussed in this paper. Further research could be arranged to apply sensory technique on other garment perception areas.
References Chen, Y. (2000), “Perception created from garment”, Overseas Silk, No. 6, pp. 37-40. Mitsuo, N. (2002), “Kansei engineering as a powerful consumer-oriented technology for product development”, Applied Ergonomics, Vol. 33 No. 3, pp. 289-94. Simon, T.W., Schuttet, J.E., Axelsson, J.R.C. and Mitsuo, N. (2004a), “Concept, methods and tools in Sensory Engineering”, Theoretical Issues in Ergonomics Science, No. 5, pp. 214-31. Simon, T.W., Schutte, J.E., Axelsson, J.R.C. and Mitsuo, N. (2004b), “Concepts, methods and tools in Kansei engineering”, Theoretical Issues in Ergonomics Science, No. 5, p. 222. Su, J.N. and Li, H.Q. (2005), “Investigation of relationship of form design elements to sensory image by means of quantification-I theory”, Journal of Lanzhou University of Technology, No. 2, pp. 37-9. Wang, M. (2002), Garment Style Design, Liao Ning Technology Publishing Company, Liao Ning, pp. 153-70. Wu, G.F., An, W.F. and Liu, J.H. (1992), Practical Data Analysis Method, China Statistics Publishing Company, Beijing, No. 10, pp. 301-6. Xu, J. (2004), “The study of rules about product form based on the user’s preference image”, Light Industry Machinery, Vol. 4, pp. 4-6. Yang, S.M., Mitsuo, N. and Lee, S.Y. (1999), “Rule-based inference model for sensory engineering system”, International Journal of Industrial Ergonomics, Vol. 24, pp. 459-71. Corresponding author Yan Chen can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Research on the style of women’s overcoats 183
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,3
Analysis and evaluation of color perception Yan Chen and Zhi-ge Chen College of Material Engineering, Soochow University, Suzhou, People’s Republic of China
184 Abstract
Purpose – The research documented in this paper aims to identify the components and characters of color perception and to establish the basis for color perception description and evaluation. Design/methodology/approach – Ten pairs of antonyms obtained on the basis of investigation were suggested to evaluate the perception of color. About 15 color samples were prepared for subjective evaluation. The factorial analysis method and artificial neural network were applied to analyze the evaluation results given by professionals and to establish the objective basis for color perception evaluation. Findings – The color perceptions could be distinguished by word pairs. These word pairs were grouped according to the perception characteristics. The relation between color parameters and color perception could be established by artificial neural network technique. Research limitations/implications – The results showed in this paper were obtained on the limited number of samples. Originality/value – The principle of color perception was investigated on the subjective evaluation of color samples. The relation between color parameters and color perception was established and could be applied for reference of clothing color design. Keywords Colour, Neural nets, Sensory perception Paper type Research paper
Introduction Since, early 1970s last century, the researches on color perception have been made processes to establish the relation between color physics and physiology ( Evans, 1974). The responses and standpoints created from color observation were investigated on the basis of measurement results of light spectrum (Gu et al., 2000). The results of these researches were applied to describe and to evaluate the perception of colors (Petiot and Yannou, 2004). Based on subjective knowledge, the perception of the color was usually described by words, which provided inconsistent evaluation standards and limited the information exchanges among people. The color perception is one of the key factors in textile and garment areas. Research work was carried to discover the physiological and psychological influences of the color vision on the designers and consumers (Chen and Li, 2004). It is significant to create the analysis system for color perception evaluation for the design and improvement of the consuming products.
International Journal of Clothing Science and Technology Vol. 20 No. 3, 2008 pp. 184-191 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810865238
Description of color perception People were used to describe the color perception on their own understanding by words as bright, fresh, colorful, dark, light, cold, warm, elegant, chaste, lively, mystery, simple, and so on. Some of the words are specific while some are not. In order to investigate the understanding of the people on these words, the descriptions of the color by ten pairs of antonyms are put forward in this research. The antonyms include the words for
substantial sense as warmness (cold-warm), distance (backward-forward), volume (shrink-expand), weight (light-heavy), brightness (shade-bright), hardness (soft-hard), gender (female-male) and the words connected with spirit and emotion of the people (still-exciting, gloomy-brisk, simple-luxuriant) (Chen and Li, 2005). Subjective evaluation of color perception The differentiation of the color perception and data process can be facilitated by numbering 1 , 10 to rank the evaluation results. About five colors, grey, blue, red, yellow, and green were selected and each color had three saturations. The colors samples were printed on polyester satin fabric by ink-jet transferring printing method. The CMYK values of the samples are shown in Table I. The color perception evaluation is carried out by ten trained professionals. The color samples were showed to them together. Professionals ranked these samples from 1 to 10 grade for each of the aspects of color perception, which was described by above word pairs. The evaluation results given by professionals are shown in Figure 1. The polar coordinates 1 to 10 correspond the ten aspects of the color perceptions.
Analysis and evaluation of color perception 185
Analysis of subjective evaluation results of color perception The results of subjective evaluation show that each color has special characters which can be presented by the “fingerprint” shown in Figure 1. Following information can be obtained from the figure: Firstly, the colors with same hue show the different perception characteristics. As the increases of color saturation, the perception is varied obviously, for example the “shade-bright,” “light-heavy,” “cold-warm,” “soft-hard” of the grey colors (sample 1 to 3); the “gloomy-brisk,” “plain-luxuriant,” “stillness-exciting,” “male-female” of the blue colors (sample 4 to 6), and so on. Then, there are still a lot of similarities in the perceptions of the color with same hue, for example the “simple-luxuriant,” “gloomy-brisk,” “stillness-exciting,” “backwardforward” of the grey colors; the “backward-forward,” “cold-warm” of the blue colors, and so on. Finally, Figure 1 can be used for the references as the color design in practical uses, if the ranges of the color perception could be given on the basis of widely investigation. The ten aspects of color perception can be considered as the elements of the “styles” of color. These items are not independent completely, and the functions of them on the “styles” of color are different. The mathematic methods are used to analyze the main factors of the color perceptions, which can simplify the problems and show the compositions and the meanings of the “styles” of color.
Gray Blue Red Yellow Green
Light
C
M
Y
K
Medium
C
M
Y
K
Dark
C
M
Y
K
1 4 7 10 13
0 30 0 0 30
0 0 30 0 0
0 0 30 30 30
30 0 0 0 0
2 5 8 11 14
0 75 0 0 75
0 0 75 0 0
0 0 75 75 75
75 0 0 0 0
3 6 9 12 15
0 100 0 0 100
0 0 100 0 0
0 0 100 100 100
100 0 0 0 0
Table I. CMYK values of the color samples
IJCST 20,3
sample1
sample2
10
186 9
1
10 8 6 4 2 0
2
10
8
3
9
4
8
5
7
9
7
sample9
sample10
10
2
3
9
4
8
5
sample11 1 10 8 6 4 2 0
sample14
2
3
4
5
sample15
10 8 6 4 2 0
1 cold/warm 2 backwards/forwards
2
3 shrink/expand 4 light/heavy 3
5 still/exciting 6 gloomy/brisk
8
4
7 plain/luxuriant 8 dark/bright
7
Figure 1. Evaluation results of color perception
sample12
6
1
9
5
7
6
10
3
7
8
sample13
2
6
sample8 1 10 8 6 4 2 0
10
10 8 6 4 2 0
4
6 sample7
sample6
sample5
sample4
sample3
1
5
9 soft/hard 10 female/male
6
The integrated common factor Z is used to cover the information presented by the variables of color perception x1, x2, . . . x10 obtained in this research. The relations between them can be expressed as follows (Lu, 2000): x1 ¼ b11 Z 1 þb12 Z 2 þ· · ·þ b1n Z n þe1 x2 .. .
¼ b21 Z 1
x10
¼ b101 Z 1
þb22 Z 2 .. .
þ· · ·þ
þb102 Z 2
þ· · ·þ
b2n Z n
b10n Z n
þe2 .. .
Analysis and evaluation of color perception 187
þe3
Above equations can be shown in matrix form as X ¼ BZ þ E, where X is the color perception matrix, B is the weight coefficient matrix, Z is the common factor matrix and E is the residual matrix. If all the Zi (i ¼ 1 , n) are independent each other and the residuals are small enough, the matrix can be written as Z ¼ AX. Where, A is the conversed matrix. The weights of common factors Z1, Z2, . . . Zn decreased in turns. Only the factors ranked in the front of the equations are significant to cover the majority information contained in the original data. By rotating the factor model, the weight coefficients of the common factors are increased (1 is the biggest) or decreased (0 is the smallest). Therefore, it will be feasible to analyze, explain and name the common factors. Two factors are obtained by the factorial analyze, and 88.76 percent information in original data are included in them. The Kaiser-Meye-Olkin measure given by KMO test is 0.84 (. 0.7), which means the number of samples used is enough for this problem. The rotated factor matrix and the weight coefficient of the factors are listed in Table II. It can be found that the weight coefficients of the factors have been polarized and are in favor of the color perception explanation. According to Table II, the following explanation can be given for the color perception in the term of the “styles.” The “styles” of color is contributed by the two main factors. The first factor covers the “dark-bright,” “light-heavy,” “soft-hard,” “female-male” “shrink-expand” “cold-warm” and “backward-forward”; the second factor covers “still-exciting,” “gloomy-brisk,” “simple-luxuriant,” “cold-warm” and “backward-forward.” The “styles” of color on “cold-warm” and “backward-forward” are included in both factors. It can be found that apart from these two terms, the color perceptions in each Item Perception
Factor matrix Factor 1 Factor 2
Weight coefficient Factor 1 Factor 2
Dark-bright Light-heavy Soft-hard Female-male Shrink-expand Still-exciting Gloomy-brisk Simple-luxuriant Backward-forward Cold-warm
20.976 0.956 0.943 0.918 20.724
20.231 0.235 0.224 0.201 20.130 0.057 0.075 0.067 0.000 20.099
20.611
0.574 0.929 0.879 0.872 20.757 20.621
20.064 0.097 0.065 0.000 0.112 0.266 0.258 0.254 20.202 0.134
Table II. Circumrotated factor matrix and the weight coefficient of the factors
IJCST 20,3
188
factor are different from those in the other factor. The perceptions in the first one are concrete and substantial, while those in the second one are more immaterial and abstract. The “cold-warm” and “backward-forward” present themselves in both factors, which could be explained as the both substantial and immaterial ingredient in them. It is possible to name these two factors on above analysis. The first one can be named as “concrete” or “substantial” factor and the second one the “immaterial” or “abstract” factor. The two factors represent the different contents of the color “styles.” According the weight coefficients listed in Table II, the relative equations can be obtained to calculate the factors. The factor values for 15 colors samples listed in Table I are calculated and the results are showed in Figure 2. Following results can be deduced from the factor analysis on color perceptions. The discrepancy of the color in term of color perceptions can be distinguished by means of factor analysis. The colors with same hue have the special order in factor distribution. The factor values calculated from perception evaluation express the characters of the color. Of the 15 samples, the grey colors are obviously “substantial” one. This color is less emotional. The red, yellow and green colors have strong “immaterial” feature, which easily cause the resonance and association with ideas. The blue color lies in between, with almost same level of the “substantial” and “immaterial” features. Above results are consistent to the common sense of the people about colors. It can be also found from the figure that both “substantial” and “immaterial” features vary with the saturation of the colors. The saturation of the colors influences the discrepancy of the “color style,” and the light colors have less difference in “style” than dark colors. 10 9 8 7 factor 2
6 5 4 3 2 1
–4
–2
0
0
2
4
6
factor 1
Figure 2. Distribution of the color perception factors
gray (sample 1∼3)
blue (sample 4∼6)
red (sample 5∼9)
green (sample 13∼15)
yellow (sample 10∼12)
8
Objective evaluation of color perception by artificial neural network Research has showed that the color perception results from the physical characters of the color and influenced by psychological and physiological factors of the people. There are very complicated relations between the perception and the physical, psychological and physiological of the colors. The artificial neural networks (ANNs) can imitate the processes of human neural networks and have great advantages in establishing the relation between color and color perception, therefore the color perception can be objectively evaluated according to physical characters. Back propagation (BP) network is a multi-layer net and carries out the no-linear mapping process from inputs to outputs. The network can be composed of three or more layers neural nodes, one input layer, one output layer and one or more hidden layers. The number of input nodes and output nodes correspond to the number of input- and output-variables, respectively. The number of hidden nodes is varied and usually determined by tests and trials. To investigate the relations between the results given by objective measurement and subjective evaluation by means of ANNs, 30 polyester satin color samples were used for experiments. The color samples were obtained by inkjet transferring printing. Of these samples, 15 are grey, blue, red, yellow, and green with three different saturation levels, and others are taken from chromatogram cards randomly. L * a * b * values of the samples are measured under CIE standard illumination D65 and “1964CIE 108 supplement standard observer.” About ten subjective sensory values of color perception are obtained by ranking method given by ten professional persons. The objectively measured data were used as the inputs and the subjective evaluation results as the outputs of the network. The network with three input nodes and ten outputs nodes was constructed as Figure 3. The error threshold of training was set at 5 percent. BP net worked in two steps: first, the network worked in “training mode.” The measured colors were used as inputs, and the subjective evaluation results were used as expected outputs. The training process cycled until the errors between network outputs and expected outputs reached the preset threshold. Trained network could be used in “enquiry mode,” and could predicate the color perception on the given L * a * b * values, which were inputted to the network as enquiries. Figure 4 shows the number of training cycles of the network with different hidden nodes. It can be found that, the number of hidden nodes affects the training process of the network and properly selected hidden nodes reduce the number of training cycles. Training Input
Data Scale
Training Output
Data Scale
Analysis and evaluation of color perception 189
Error training
Enquiry Input
Data Scale
Artificial Neural Network
enquiry Data Unscale
Enquiry Output
Figure 3. Working chart of BP network
IJCST 20,3
50,000
training cycle
40,000
190
30,000 20,000 10,000
Figure 4. Training cycle of different hidden nodes
0
10
11
12
13
14
15
hidden node
It can also be found that, the number of hidden nodes beyond the certain ranges results in the difficult for error convergence. Additional six color samples were used to test the network. Figure 5 shows the comparison of the subjective evaluation results and objective evaluation results given by BP neural network with 12 and 14 hidden nodes. The comparison made in Figure 5 shows that the trained neural networks with different hidden nodes give out the similar results for color perception, prediction. Table III is the relativity analysis for the predicting results given by different structure networks. It is found that all the networks constructed in this work give out quiet precise prediction on color perception. The higher relativity coefficients exist between subjective evaluation results and network prediction. Therefore, it can be deduced that the ANNs can be used to predict the perception of people on colors according to the objective measurement results. 3-12-10 net
3-14-10 net 12
Table III. Correlation analysis of the results given by subject evaluation and network prediction
10 31 32 33 34 35 36
8 6 4 2 0
0
2
4 6 8 net prediction
10
Coefficient of correlation Sig.a
10 31 32 33 34 35 36
8 6 4 2 0
12
(a) net with 12 hidden nodes
Hidden node
subjective evaluation
Figure 5. Comparison of objective evaluation results of BP net
subjective evaluation
12
0
2
4 6 8 net prediction
10
12
(b) net with 14 hidden nodes
10
11
12
13
14
15
0.812 0.000
0.780 0.000
0.794 0.000
0.792 0.000
0.788 0.000
0.808 0.000
Note: aRelated to the root mean square error
Conclusions A lot of color phenomena in practice can be proved and explained by the results of this research. Every color possesses specific “styles,” which can be shown by the evaluation results of the color perception in this work. To investigate and understand the characters of the perceptions are helpful for textile and garment color design and uses. By means of factor analysis, two common factors of the color perceptions are obtained, which can be related to the “substantial” and “immaterial” elements of the color perception, respectively. This is consistent with the experience of the people and is helpful for us to study color phenomena and substances. The research of using BP network to predict the perception of color samples showed great success. The predicting results given by BP networks according to the objective measurement are highly related with the results given by subjective evaluation. References Chen, Y. and Li, D.G. (2004), “Physiological research on color perception”, Journal of Textile Research, Vol. 25 No. 3, pp. 68-9. Chen, Y. and Li, D.G. (2005), “Evaluation method of color perception”, Journal of Textile Research, Vol. 26 No. 2, pp. 118-20. Evans, R.M. (1974), The Perception of Color, Wiley, New York, NY. Gu, D.Z., Fu, S.S. and Yang, R.M. (2000), Principle of Color and Figure Vision, Scientific Publishing Company, Singapore. Lu, W.D. (2000), SPSS for Windows – Statistics, Publishing House of Electronics Industry, Beijing. Petiot, J.F. and Yannou, B. (2004), “Measuring consumer perceptions for a better comprehension, specification and assessment of product semantics”, International Journal of Industrial Ergonomics, Vol. 33, pp. 507-25. Corresponding author Yan Chen can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Analysis and evaluation of color perception 191
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Automated side-seam placement from 3D body scan data
Automated side-seam placement
Susan P. Ashdown Department of Fiber Science & Apparel Design, Cornell University, Ithaca, New York, USA
Mee Sung Choi Department of Costume and Design, Dongshin University, Naju, Republic of Korea, and
199 Received 2 May 2007 Revised 12 September 2007 Accepted 12 September 2007
Eric Milke Factset Research Systems, Norwalk, Connecticut, USA Abstract Purpose – The three-dimensional (3D) body scanner is an important new technology that will impact the design and production of apparel, but use of this tool is at an early stage of development. Appropriate measurement extractions from the complex 3D scans that will address the needs of apparel patternmakers are an essential part of the development process for this new tool. The paper aims to address these developments. Design/methodology/approach – In this study, a method of automatically locating the side seam for torso fitting garments from 3D body scans for a variety of body types was developed and tested. The method is based on the location of center points of body depth measurements, and five different body landmarks or combinations of body landmarks were tested to determine the best choice for implementation. Findings – Based on rankings and ratings of the results by apparel experts, a method using the average value of body depth measurements taken at about 100 locations equally spaced from the axilla to the crotch was chosen as the best solution. Research limitations/implications – Additional testing of this method and development of a method for locating the side seam for lower body garments is the next step in this research. Originality/value – Identifying appropriate landmarks and body measurement extraction processes for apparel or style-based measurements is as important as the more commonly derived anthropometric measures based on body landmarks. Landmarks such as side-seam placement pose unique challenges that must be solved with analysis and reconstruction of style-based data. The paper provides information on these factors. Keywords Clothing, Image scanners, Textile technology, Design and development Paper type Research paper
Background Three-dimensional (3D) body scan data provides a wealth of information about the body that can be used in anthropometric studies to quantify the range and variation of the population, in custom fit patternmaking to create personalized patterns for individuals, and in mass customization and size selection operations to identify appropriate sizes for the individual from a range of choices (DesMarteau, 2000; Fallon, 1999; Haisley, 2002; Hye, 1999; Incremona, 1996; Robinette, 2000; Size USA, 2004; Size UK, 2004; Staples et al., 1994). Researchers are experimenting with many different ways to generate custom-fitted patterns (Hinds et al., 1992; Vollinger, 1998; Winakor et al., 1990; Yavatkar, 1993).
International Journal of Clothing Science and Technology Vol. 20 No. 4, 2008 pp. 199-213 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810878829
IJCST 20,4
200
For each of these uses, it is necessary to identify landmarks on the body that correspond to traditional patternmaking locations for apparel. One of the more difficult landmarks to identify is the side-seam location. The side seam divides the torso (and the garment) on the transverse axis into front and back components. Appropriate placement of this landmark is important for the creation of well-balanced garments. 3D body scan data There are a variety of 3D body scanners on the market, but the essential process of data collection and the resulting data are similar for all scanners. The scanner consists of a light source (laser or white light) and a number of cameras (2-16) to capture coordinate data to model the surface of the body. Some scanners collect information about the color and surface texture of the body in addition to the measurement data, but the basic data consists of an x, y, z-coordinate dataset that can be used to reproduce the surface of the body on a computer screen. These data can be manipulated, measured manually on the computer screen, and automatically measured. Most developers of 3D body scanners provide some level of automated measurement extraction with the scanner (Daanen and van de Water, 1998; Human Solutions, 2002; TC2, 2004; Cyberware, 1999; Hamamatsu, 2003). Writing programs for automated measurement extraction that will provide useful measurements for apparel end uses is difficult as traditional measurements depend on information not available directly from a 3D body scan. Some traditional apparel landmarks are identified by palpitating for skeletal protrusions that cannot be located from surface data of the body. For example, the point of the shoulder is an important landmark for shoulder and arm measurements. In order to fit apparel correctly based on these measurements, it is necessary to locate the point of articulation of the arm between the bones in the shoulder girdle and the top of the humerus. This landmark is generally located by palpitating for the medial superior point of the acromial process of the scapula, as this bone is located directly above the arm joint. However, this point is not visible on the surface of the skin, so software developers of automated measurements must generate a method of finding an equivalent landmark based on the surface geometry of the body (Brunsman et al., 1997; Pargas et al., 1997). This can be difficult given the great variation in body shapes that exist in the population. A geometric solution that is appropriate for one type of shoulder shape may not work well for a different shoulder shape. Other body landmarks used in apparel are based on apparel conventions and styles, and are located by the apparel expert based on individual concepts of the appropriate fit of clothing or on changing styles and proportions of apparel. The location of the waist is an example of a landmark that can be either based on an identifiable body landmark or a style or fashion-based location. The omphalion, the narrowest point of the body in a silhouette view, or the point of greatest curvature of the small of the back are all physical landmarks that can be used to identify the waist. However, pant and skirt styles for women for current styles require a waist measurement lower than any of these body landmarks. In this case, the waist is set according to the judgment of the apparel expert. The placement of a side seam is a conceptual process necessary for the creation of most shaped and tailored clothing styles for which there is no clear body landmark. The location of the side seam is a concept based in current clothing styles of many different cultures. In current western clothing styles, which tend to be shaped to the body and therefore greatly differentiated between the front and back of a garment, appropriate placement of the side seam is important to divide the garment. There is little guidance for
the appropriate location of the side seam in common practice, and methods that do exist are unreliable given the variation in body profiles in the population. Various experts in patternmaking and in fitting apparel locate the side seam in different ways. Farmer and Gotwals (1982) state that the side seam should begin at the center of the armscye and should be vertical. According to Rasband (1994), the side seam should extend from the center of the underarm, it should be perpendicular to the floor, and it should appear to intersect the waistline at a 90 angle. A classic fitting book, Vogue Fitting, instructs that the side seam should be “centered on the body” from the side view (Lenker, 1987). According to Liechty et al. (1992), the side seam should “divide the body in becoming proportions.” They also indicate that it should be placed so that it creates an equal visual distance from the front, side, and back, i.e. the side seams should not be visible from the front or back view of the body and should be centered on the side view of the body. Hazen (1998) describes a body measurement process that makes use of a chalk landmark line drawn on a leotard and tights donned for this purpose. She identifies the center of the underarm as the appropriate landmark and instructs the reader to mark a line down the side of the body to the outside of the ankle. The challenge of creating an automated expert There are no obvious physical landmarks for placement of the side seam on different individuals, and based on the search of the literature there are few rules for placement and those that do exist are in conflict. The most common landmark, centered under the arm, works well for individuals with balanced profiles, but when applied on a figure with protruding buttocks, abdomen, bust, or shoulder blades this landmark does not result in the balanced, centered, or proportional placement that is the other element cited to achieve appropriate clothing fit. Side-seam placement is therefore a skilled task generally performed by apparel professionals during pattern development and fitting on fit model bodies for ready-to-wear and on the individual for custom-fitted clothing. Appropriate placement depends on the configuration of the body and different professionals use different clues to assist with placement. Our experiments with automated methods of placing the side seam using body scan data are therefore based on different criteria that expert tailors have used. Several different landmark locations were used and then visually judged for appropriateness by a panel of experts to identify the most valid and reliable method for a variety of body types. Assumptions For this analysis, we assume that the side seam is vertical. For some unbalanced profiles or extreme postures, a tilted side seam may be appropriate, but at this stage we tested only vertical side seams on a range of body types, proportions, and postures. Therefore, to solve our problem, we only need one coordinate point on the body. A frontal plane (the x-z plane in the orientation of the scan shown in Figure 1) that contains this point will divide the front and the back of the body scan at the chosen side-seam location. A vertical line drawn on the profile view of the body at this plane location will provide a visualization of the side seam on the body. For this project, we concentrated on finding an appropriate side-seam location for upper torso garments only. This may not provide the best location for lower body garments. For this analysis, we translate the data from the body scan so that the origin of the coordinate axis is located in the center of the body on the y- and z-planes, and at the
Automated side-seam placement 201
IJCST 20,4
202
Figure 1. Subject sliced with a frontal plane (x-z plane) Note: Reproduced from the only available original
most inferior point on the x-plane, placing the (0, 0, 0) point on the floor centered between the feet. We assume that the scan has been taken with the body held square to the cameras with no twisting of the torso so that a frontal plane through the body scan will derive the same solution on the left and the right sides of the body. Methods tested Two basic methods with variations were tested. One solution was to create a frontal plane centered on the body depth using data at one selected body landmark. The following landmarks were chosen as possible candidates for side-seam location using this method (Gordon et al., 1989a): . The axilla, which is located at the fold where the armscye joins the body. This landmark will provide a mark close to the center of the underarm location favored by many experts in apparel.
.
.
The tenth rib, which is located at the most inferior point of the curve of the lowest rib. This landmark provides a torso landmark that is just above the waist on the side of the body. The iliocristale, which is located at the anterior superior iliac spinepoint of the pelvis (high-hip bone) on the pelvic girdle. This landmark provides a lower torso location that is just below the waist.
A second solution uses available computational power to use more of the 3D data by averaging the central body depth data from a series of positions on the torso. An average value that includes more of the 3D data may give a more robust solution as it takes into account many features of the body. This solution was initially calculated using slices from the axilla to the iliocristale, and then a second solution was derived using slices from the axilla to the crotch. Method of calculation The calculations were done with the software Matlab. The data acquired by the body scanner for each scan were a set of about 300,000 points, stored as x, y, z-coordinates. As a first step this dataset was reduced to a set of about 60,000 points by deleting redundant data. The method of analysis chosen was to consider this data as a 60,000-line matrix, and the “Matrix laboratory” in Matlab allows us to do this. The calculation is based on a simple principle. We extract each desired point by taking a transverse slice of data using a y-z plane at each landmark location. Once we acquire the slice the next step is to find the center of the body depth measure from that slice and project this center point to the side of the slice. We do this by finding the mid-point between the front and the back on a centerline drawn on the sagittal axis of the slice. The mid-points between front and the back are in fact defined by locating the two points on y where z ¼ 0 (Figure 2) because the data are centered on 0. This body depth measurement allows us to concentrate on the core of the body, as it does not measure the full depth of the body which would include the most anterior or posterior points of the breast or buttocks in our calculation. These more extreme points would tend to shift the resulting side-seam placement away from the core of the torso. The first step is to locate the desired landmarks for the axilla, tenth rib, iliocristale, and crotch point from each individual body scan, so that the y-z slices can be taken at these points. However, these landmarks are difficult to identify from body scan data as they are based on skeletal features that are not visible in the body scan (the tenth rib and iliocristale) or on body features that are not clearly captured by the scan (axilla and crotch point). An alternate method of locating landmarks is to derive them from a measurement that can be easily identified on the scan. Population data can then be used to calculate any measurements that are highly correlated with this identifiable measurement. Measurements for this study were calculated using population ratios derived from the 1988 ANSUR study (Gordon et al., 1989b). Stature was chosen as a measurement that could be easily taken from the scan that is highly correlated with all of the desired landmarks (Table I). Although it can be more accurately measured if subjects are scanned in a bathing cap to compress their hair, a process not used in this study, reasonably accurate values for stature could be visually derived from the scans. Ratios based on height were calculated separately for the males and the females in the study for each landmark. This method will only be completely accurate for those
Automated side-seam placement 203
IJCST 20,4
x
y Axilla
y
204
front 10th rib lliocristale Crotch
o z
back
Figure 2. The landmarks and the orientation of the coordinate system
Table I. Regression values for stature vs the listed measurements
Notes: A horizontal slice of the subject at the level of the tenth rib shows the method of dividing each slice. This diagram is reproduced from the only available original
Axilla 10th rib Iliocristale Crotch height
Men
Women
0.935 0.864 0.847 0.725
0.941 0.869 0.837 0.706
individuals whose proportions match the population ratio, but it was judged to be sufficiently close to the true landmark for our purposes. For each of the landmark levels on each scan, we extracted a transverse (horizontal) slice. The side-seam point for this slice is identified on its diameter based on the division of the body calculated from the mid-point taken from the y-z plane. The side seam is then visualized as a vertical line from this point on a profile view of the scan (Figure 3). For the second method, the body is decomposed into about a 100 slices, depending on the size of the subject. The position of the side seam can now be calculated as the average of the mid-points of this complete set of slices. Two different sampling ranges were tested. Both ranges start at the axilla, but one goes only as far as the iliocristale while the other goes all the way to the crotch. This average value then is used to locate the side seam, and the visualization is generated as a vertical line as in the previous method. Assessment of resulting side seams In order to compare the different methods for locating the side seam, we produced a test instrument consisting of profile views of 20 subjects. On each page, we presented all the different proposed solutions for one subject presented in a random order (Figure 3). Five additional cases were made from duplicate scans. This set of 25 pages was given to
Axila
10th Rib
lliocristale
Mean value
Full Mean Value
1,000
1,000
1,000
1,000
1,000
800
800
800
800
800
600
600
600
600
600
400
400
400
400
400
200
200
200
200
200
0
0
0
0
0
–200
–200
–200
–200
–200
–400
–400
–400
–400
–400
–600
–600
–600
–600
–600
–800
–800
–800
–800
–800
–1,000
–1,000
0 –200
0
200 –200
–1,000
0
200 –200
–1,000
0
200 –200
205
–1,000
0
200 –200
Automated side-seam placement
0
200
Note: Figures were labeled using a code and randomly arranged on the page in each case
seven expert judges of side-seam location. The judges included four apparel professors, three industry experts (patternmaker/fitters), and an expert in anthropometry. All experts had at least ten years of experience. Judges were first asked to rank the five choices on each page from best to worst. They were then asked to go back through the images and look at the top ranked choice on each page and to rate it from 1 to 7, with 1 being bad placement for a side seam and 7 being the optimal placement of the side seam for that individual (six of the seven judges completed this section). Before they began this process, judges were asked to describe what they looked for when placing a side seam. They were also asked to respond to this question again at the end of the rating and ranking process. Results Methods of placing side seams As was anticipated, each judge uses slightly different criteria to locate the ideal side seam. These criteria fell in three general categories: (1) centering on the profile at a body landmark (under the arm, at the waist, at the hip);
Figure 3. One of the 25 images from the test instrument
IJCST 20,4
206
(2) generating a vertical line from a landmark (the ear lobe, the shoulder point, the position of fingers in relaxed posture at side, the center of the armscye, and the hip joint); or (3) looking for balance among front and back points of greatest protrusion (shoulder blade to bust, bust to hip, shoulder to hip, and leg position). Judges stated that they look for a balanced position, and for both aesthetically and structurally appropriate placement. Six of the seven judges listed two or three location methods. When ranking the five figures judges had the most difficulty with figures that had unusual postures; five of the seven judges commented on this factor. Three judges suggested that the optimal side seam would not be perpendicular to the floor in these cases. Other difficult figures mentioned were those with large bust, abdomen, or hips, or those with spinal lordosis. The ratings for the top ranked selections for all side-seam locations were rated acceptable to optimal for 22 of the 25 cases. Of the three profiles for which no acceptable side seam was created judging by the ratings, two had an unusual posture and one had lordosis of the spine. Rankings of side-seam placement methods Table II shows the results of the expert rankings of the figures by numbers of figures placed in each ranking and the percent of the whole that this rank represents. The method that was ranked first most often was the full torso mean (mean of 100 slices from the axilla to the crotch) with 71 of the figures using this method placed in this position. The second most frequent choice for first rank was the side seam placed at the axilla, with 53 cases chosen. However, there was some variation among the judges for this choice, as three of the judges (Experts 3, 4, and 7) were more likely to rank this choice at the 3rd through 5th position than the top two ranks. The third most frequent choice for first rank (chosen 33 times) was the partial torso mean (mean of 100 slices from the axilla to the iliocristale). This was also a popular choice for the second ranked choice, with 60 of the figures placed in this position. The side seam located at the iliocristale and the one located at the tenth rib were the least popular choices; they were ranked either fourth or fifth 130 times and 94 times, respectively. Figures 4 and 5 show the top two and the bottom two rankings. The figure that was ranked first in each case was also judged (rated) to assess how successful this side-seam placement was generally. The best ranked side-seam placement in the set of five is still not acceptable if the placement is not judged to be correct. Six of the seven judges rated the seam placement on a scale from 1 to 7, with one being bad placement, and 7 being optimal placement. The average rating for the first placed side seams averaged 5.5, indicating that the methods for locating the side seam were generally very acceptable. Table III shows the same data with the five figures with the most extreme postures removed, in order to judge whether this would change the results. One of these five figures had severe kyphosis (rearward curvature of the spine along with hunchback) and the other four exhibited lordosis (extreme curvature of the spine in the lumbar region and hip tilt). The results from this analysis show the same relative ranking, indication that the inclusion of these extreme figures did not affect the results.
Total
5th rank
4th rank
3rd rank
2nd rank
1st rank
B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch)
Seam location method 12 (6.9) 1 (0.6) 1 (0.6) 4 (2.3) 7 (4.0) 5 (2.9) 3 (1.7) 0 8 (4.6) 9 (5.1) 6 (3.4) 3 (1.7) 0 6 (3.4) 9 (5.2) 2 (1.4) 10 (5.7) 9 (5.1) 5 (2.8) 0 0 8 (4.6) 15 (8.7) 2 (1.1) 0 125
Expert 1 (percent) 8 (4.6) 4 (2.3) 0 5 (2.9) 8 (4.6) 1 (0.6) 4 (2.3) 3 (1.7) 7 (4.0) 10 (5.7) 3 (1.7) 7 (4.0) 6 (3.4) 4 (2.3) 5 (2.9) 5 (3.4) 5 (2.9) 8 (4.5) 5 (2.9) 2 (1.1) 8 (4.6) 5 (2.9) 8 (4.6) 4 (2.3) 0 125
Expert 2 (percent) 4 (2.3) 1 (0.6) 0 7 (4.0) 13 (17.4) 2 (1.1) 4 (2.3) 1 (0.6) 12 (6.9) 6 (3.4) 5 (2.9) 4 (2.3) 7 (4.0) 4 (2.3) 5 (2.9) 4 (2.7) 10 (5.7) 9 (5.1) 1 (0.6) 1 (0.6) 10 (5.7) 6 (3.4) 8 (4.6) 1 (0.6) 0 125
Expert 3 (percent) 4 (2.3) 3 (1.7) 1 (0.6) 4 (2.3) 13 (17.4) 4 (2.3) 5 (2.9) 2 (1.1) 9 (5.1) 5 (2.9) 1 (0.6) 6 (3.4) 4 (2.3) 8 (4.6) 6 (3.4) 3 (2.0) 10 (5.7) 8 (4.5) 3 (1.7) 1 (0.6) 13 (7.4) 1 (0.6) 10 (5.7) 1 (0.6) 0 125
Expert 4 (percent) 11 (6.3) 1 (0.6) 0 3 (1.7) 10 (5.7) 5 (2.9) 2 (1.1) 1 (0.6) 6 (3.4) 11 (6.3) 4 (2.3) 6 (3.4) 4 (2.3) 9 (5.2) 2 (1.1) 3 (2.0) 9 (5.1) 8 (4.5) 4 (2.3) 1 (0.6) 2 (1.1) 7 (4.0) 12 (6.9) 3 (1.7) 1 (0.6) 125
Expert 5 (percent) 12 (6.9) 0 1 (0.6) 5 (2.9) 7 (4.0) 4 (2.3) 2 (1.1) 0 8 (4.6) 11 (6.3) 3 (1.7) 9 (5.2) 3 (1.7) 4 (2.3) 6 (3.4) 1 (0.7) 11 (6.3) 8 (4.5) 4 (2.3) 1 (0.6) 5 (2.9) 3 (1.7) 13 (7.4) 4 (2.3) 0 125
Expert 6 (percent) Total
55 (31.4) 12 (6.9) 4 (2.3) 33 (18.9) 71 (40.6) 22 (12.6) 24 (13.7) 12 (6.9) 60 (34.3) 57 (32.6) 27 (15.5) 45 (25.9) 29 (16.7) 36 (20.7) 37 (21.3) 21 (14.2) 63 (35.8) 57 (32.4) 27 (15.3) 8 (4.5) 50 (28.6) 31 (17.7) 73 (401.7) 19 (10.9) 2 (1.1) 875
Automated side-seam placement
Expert 7 (percent) 4 (2.3) 2 (1.1) 1 (0.6) 5 (2.9) 13 (7.4) 1 (0.6) 4 (2.3) 5 (2.9) 10 (5.7) 5 (2.9) 5 (2.9) 10 (5.7) 5 (2.9) 1 (0.6) 4 (2.3) 3 (2.0) 8 (4.6) 7 (4.0) 5 (2.9) 2 (1.1) 12 (6.9) 1 (0.6) 7 (4.0) 4 (2.3) 1 (0.6) 125
207
Table II. Results of the rankings of the five side-seam locations by seven experts (25 subjects)
IJCST 20,4 Number of times chosen
208
1st Rank (best choice) 75 65 55 45 35 25 15 5 –5
B(axilla)
D(10th rib)
F(iliocristale)
H(axilla to iliocr. )
K(axilla to crotch)
H(axilla to iliocr.)
K(axilla to crotch)
Method 2nd Rank
Figure 4. Number of times each method was selected for the first and second ranked choices (best solutions)
Number of times chosen
75 65 55 45 35 25 15 5 –5 B(axilla)
D(10th rib)
F(iliocristale) Method
Of the 25 figures judged, one was repeated twice and another was repeated three times. The purpose of introducing the same figure was to give an indication of the reliability of the different judge’s scores. With the repeated figures, judges had 24 opportunities to match rankings or ratings exactly. This measure of reliability showed that Judge 2 was the least reliable as this judge only matched ratings or rankings of identical figures twice out of 24 opportunities. Judge 5 was the most reliable, with 12 out of 24 possible scores matched (Table IV). Discussion and conclusions This work addresses the development of a method for automatically locating the side seam for a variety of body types. Our experiments with automated methods using body scan data are based on different criteria that expert tailors use. Several different methods were developed and then visually judged for appropriateness by a panel of experts to identify the most valid and reliable method. Comparison of the judge’s assessments of our five different methods for locating the side seam shows that the methods utilizing an average of a set of depth values is most frequently the preferred method. However, the placement of the side seam centering on a body slice taken at the level of the axilla was also a popular choice. Examination of
Automated side-seam placement
4th Rank Number of times chosen
75 65 55 45 35
209
25 15 5 –5
B(axilla)
D(10th rib)
F(iliocristale)
H(axilla to iliocr.)
K(axilla to crotch)
Method 5th Rank (worst choice) Number of times chosen
75 65 55 45 35 25 15 5 –5
B(axilla)
D(10th rib)
F(iliocristale)
H(axilla to iliocr.)
K(axilla to crotch)
Method
the comparative rankings of the judges shows that there are two different emerging patterns. Judges 1, 5, and 6 tended to identify the location from the axilla as the highest ranking most frequently, with the full torso mean (average slices from axilla to crotch) as their second most popular choice and also the most frequent second choice. This group of judges most frequently identified the location at the iliocristale as the lowest ranked solution. On the other hand, Judges 3, 4, and 7 most frequently ranked the full torso mean first, with the partial torso mean (average slices from axilla to iliocristale) most frequently ranked second. This group of judges identified the location from the axilla as the lowest ranked solution. Judge Number 2 demonstrates the ambivalence about the location of the side seam from the axilla, as this judge chose this solution equally as the highest ranking and the lowest ranking choice. Judge 2 exhibited the most variation in the rankings, and was also the least reliable of the judges (Tables II and IV). A comparison of the methods of locating the side seam cited by each of these sets of judges shows slight differences which may account for the differences. Judges 5 and 6 described finding a balanced location at both shoulder/armscye and waist/hip locations; Judge 1 described the shoulder point, bust point, and shoulder blade protrusions as the primary balance points. Judges 3 and 7 concentrated on finding a balance and aesthetically pleasing location but did not mention defined landmarks
Figure 5. Number of times each method was selected for the forth and fifth ranked choices (worst solutions)
Table III. Results of the rankings of the five seam location by the experts for the 20 normal figures
B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch) B (axilla) D (10th rib) F (iliocristale) H (axilla to iliocr.) K (axilla to crotch)
9 (6.4) 1 (0.7) 1 (0.7) 4 (2.9) 5 (3.6) 4 (2.9) 2 (1.4) 0 6 (4.3) 8 (5.7) 5 (3.6) 2 (1.4) 0 5 (3.6) 7 (5.0) 2 (1.4) 9 (6.4) 7 (5.0) 3 (2.1) 0 0 6 (4.3) 12 (8.6) 2 (1.4) 0 100
7 (5.0) 2 (1.4) 0 5 (3.6) 6 (4.3) 1 (0.7) 3 (2.1) 1 (0.7) 7 (5.0) 8 (5.7) 2 (1.4) 6 (4.3) 6 (4.3) 2 (1.4) 4 (2.9) 4 (2.8) 4 (2.8) 8 (5.7) 2 (1.4) 2 (1.4) 6 (4.3) 5 (3.6) 5 (3.6) 4 (2.9) 0 100
Expert 2 no. (percent)
Note: Five sets of figures with extreme posture removed
Total
5th rank
4th rank
3rd rank
2nd rank
1st rank
Expert 1 no. (percent) 4 (2.9) 0 0 5 (3.6) 11 (7.9) 2 (1.4) 4 (2.9) 0 10 (7.1) 4 (2.9) 4 (2.9) 3 (2.2) 6 (4.3) 3 (2.2) 4 (2.9) 3 (2.1) 7 (5.0) 8 (5.7) 1 (0.7) 1 (0.7) 7 (5.0) 6 (4.3) 6 (4.3) 1 (0.7) 0 100
Expert 3 no. (percent) 4 (2.9) 3 (2.1) 1 (0.7) 3 (2.1) 9 (6.4) 4 (2.9) 3 (2.1) 2 (1.4) 6 (4.3) 5 (3.6) 0 5 (3.6) 2 (1.4) 8 (5.8) 5 (3.6) 2 (1.4) 8 (5.7) 7 (5.0) 2 (1.4) 6 (4.3) 10 (7.1) 1 (0.7) 8 (5.7) 1 (0.7) 0 100
Expert 4 no. (percent) 9 (6.4) 0 0 2 (1.4) 9 (6.4) 5 (3.6) 2 (1.4) 0 6 (4.3) 7 (5.0) 4 (2.9) 5 (3.6) 3 (2.2) 6 (4.3) 2 (1.4) 1 (0.7) 7 (5.0) 8 (5.7) 3 (2.1) 1 (0.7) 1 (0.7) 6 (4.3) 9 (6.4) 3 (2.1) 1 (0.7) 100
Expert 5 no. (percent) 10 (7.1) 0 0 4 (2.9) 6 (4.3) 3 (2.1) 2 (1.4) 0 6 (4.3) 9 (6.4) 3 (2.2) 7 (5.0) 2 (1.4) 4 (2.9) 4 (2.9) 1 (0.7) 8 (5.7) 8 (5.7) 2 (1.4) 1 (0.7) 3 (2.1) 3 (2.1) 10 (7.1) 4 (2.9) 0 100
Expert 6 no. (percent)
3 (2.1) 2 (1.4) 1 (0.7) 3 (2.1) 11 (7.9) 1 (0.7) 3 (2.1) 5 (3.6) 9 (6.4) 2 (1.4) 3 (2.2) 9 (6.5) 3 (2.2) 1 (0.7) 4 (2.9) 3 (2.1) 5 (3.5) 7 (5.0) 3 (2.1) 2 (1.4) 10 (7.1) 1 (0.7) 4 (2.9) 4 (2.9) 1 (0.7) 100
Expert 7 no. (percent)
210
Seam location method
46 8 3 26 57 20 19 8 50 43 21 37 22 29 30 16 48 53 16 8 37 28 54 19 2
(32.9) (5.7) (2.1) (18.6) (40.7) (14.3) (13.6) (5.7) (35.7) (30.7) (15.1) (26.6) (15.8) (20.9) (21.6) (11.3) (34.0) (37.6) (11.3) (5.7) (26.4) (20.0) (38.6) (13.6) (1.4) 700
Total
IJCST 20,4
whereas Judge 4 only cited landmarks (center of armscye and fingers in relaxed position at the side). Judge 2 had the least defined landmarks and method. Judges 2 and 7 indicated the most difficulty with the process generally, which is reflected in the lower reliability of these judges. From this, one might conclude that the judges that had a more defined set of landmarks in mind preferred the axilla generally and those who used a more intuitive method preferred the method from the full torso mean. However, the sample is too small for this conclusion to be fully supported. Generally, this study reinforces the contention that there is variation in the methods of placing the side seam and in the preferred location of the ideal side seam. However, most of the judges indicated that the differences in the side-seam placement in this study were generally very small, giving support to the possibility an automated process that gives an acceptable result in most cases. Although no one best solution was identified for all figure types, the full torso mean selection may be the optimal choice for the full range of figures in the population. A comparison of the figures revealed that in every case at least one expert chose this as the best option. Also, in all cases except one, when the full torso mean was not selected it was identified as the next best option. As the differences were sometimes very small between the full torso mean, the partial torso mean, and the axilla this solution may be acceptable to optimal for all bodies. The high scores given for the axilla method has implications for traditional anthropometry as a caliper measurement taken at this location may be the best proxy for locating the side seam when a 3D scanner is not available. All that would be required would be to design a tool that could locate the side-seam point on the body from this measurement. Although there were differences in the reliability of the judge’s assessments, the small number of exact repeats of images in the instrument does not give a fair assessment of this factor. The fact that the differences were very small between several of the methods also contributes to the difficulty of testing the reliability of the results. Consideration of the differences between male and female figures may reveal that one method is better for males and a different method for females. The differences in the profile views would seem to support this assumption, as the visual balance of the body would be different given the curve of the breasts of the female figures. As the data for placing the side seams were taken from the central axis of the body, the depth of the breasts was not included in these numbers. This choice was made because the central core of the body, the ribcage and shoulders, were considered more important in pacing the side seam. On a large-busted woman, this would have the effect of placing the side seam visually further toward the back. However, examination of the data showed no particular trends that would identify one method to be more successful for men or women. As less than one quarter of the scans used in this study were males, there is not Expert 1 4 20 percent (n ¼ 20)
Expert 2
Expert 3
Expert 4
Expert 5
Expert 6
Expert 7
2 8.3 percent (n ¼ 24)
9 38 percent (n ¼ 24)
6 25 percent (n ¼ 24)
12 50 percent (n ¼ 24)
6 25 percent (n ¼ 24)
4 17 percent (n ¼ 24)
Note: Number of times judges gave identical figures the same ranking or rating, and percent of total number of ratings for identical figures
Automated side-seam placement 211
Table IV.
IJCST 20,4
212
enough data for a conclusive comparison to determine if one method would be preferable for males and a separate one for females. The general method for placing side seams from the body scan data, of locating landmarks using population data, and then placing the side seam using an average of the values from the whole range of data on the torso, is very promising. This method takes into consideration all of the individual variations of profile that are possible in the population. Although the results for figures with extremes of posture were not successful, this method should be applicable to the majority of the population. A similar method taking data from the lower body could be developed for pant and skirt side seams. This could then contribute to automated measurement systems for effectively creating custom-fitted patterns, size selection systems, or mass customization systems for any style of garment for individuals. References Brunsman, M.A., Daanen, H.A.M. and Robinette, K.M. (1997), “Optimal postures and positioning for human body scanning”, paper presented at the International Conference on Recent Advances in 3-D Digital Imaging and Modeling. Cyberware (1999), available at: www.cyberware.com (accessed June 18, 2004). Daanen, H.A.M. and van de Water, G.J. (1998), “Whole body scanners”, Displays, Vol. 19 No. 3, pp. 111-20. DesMarteau, K. (2000), “Pre-production and CAD: let the fit revolution begin”, Bobbin, Vol. 42 No. 2, pp. 42-56. Fallon, J. (1999), “Britain’s scanning bodies for virtual fit”, Women’s Wear Daily, February 10. Farmer, B.M. and Gotwals, L.M. (1982), Concepts of Fit: An Individualized Approach to Pattern Design, Macmillan, New York, NY. Gordon, C., Churchill, T., Clauser, C.E., Bradmiller, B., McConville, J.T., Tebbetts, I. and Walker, R.A. (1989a), 1988 Anthropometric Survey of US Army Personnel: The Measurer’s Handbook, (No. Natick-TR-89-044), US Army Natick Research & Design Center, Natick, MA. Gordon, C., Churchill, T., Clauser, C.E., Bradtmiller, B., McConville, J.T., Tebbetts, I. and Walker, R.A. (1989b), 1988 Anthropometric Survey of US Army Personnel: Methods and Summary Statistics, (No. Natick-TR-89-044), US Army Natick Research & Design Center, Natick, MA. Haisley, T. (2002), “Brooks Brothers digital tailors measure up”, Bobbin, February, pp. 26-30. Hamamatsu (2003), available at: http://usa.hammamatsu.com/en/products/system-division/ 3d-scanning.php (accessed June 18, 2004). Hazen, G.G. (1998), Fantastic Fit for Every Body, Rodale Press, Emmanaus, PA. Hinds, B.K., McCartney, J., Hadden, C. and Diamond, J. (1992), “3D CAD for garment design”, International Journal of Clothing Science & Technology, Vol. 4 No. 4, pp. 6-14. Human Solutions (2002), available at: www.human-solutions.de/main_produckte_scan_e.php (accessed June 18, 2004). Hye, J. (1999), “Body scanning has immediate applications”, Women’s Wear Daily, August 25, p. 15. Incremona, A. (1996), “If the shoe fits – you’ve been scanned: imaging for customized retail apparel”, Advanced Imaging, Vol. 10 No. 11. Lenker, S. (1987), Vogue Fitting: The Book of Fitting Techniques, Adjustments, and Alterations, Harper & Row, New York, NY.
Liechty, E.G., Pottberg, D.N. and Rasband, J.A. (1992), Fitting & Pattern Alteration: A Multi-method Approach, Fairchild Fashion and Merchandising Group, New York, NY. Pargas, R.P., Staples, N.J. and Davis, J.S. (1997), “Automatic measurement extraction for apparel from a three-dimensional body scan”, Optics and Lasers in Engineering, Vol. 28 No. 2, pp. 157-72. Rasband, J.A. (1994), Fabulous Fit, Fairchild Publications, New York, NY. Robinette, K.M. (2000), “CAESAR measures up”, Ergonomics in Design, Vol. 8 No. 3, pp. 17-23. Size UK (2004), available at: www.size.org (accessed June 18, 2004). Size USA (2004), available at: www.tc2.com/waht/sizeusa/index.html (accessed June 18, 2004). Staples, N.J., Pargas, R.P. and Davis, J.S. (1994), “Body scanning in the future”, Apparel Industry Magazine, Vol. 55 No. 10, pp. 48-52. TC2 (2004), available at: www.tct.com (accessed June 18, 2004). Vollinger, J.C. (1998), “Incorporating three-dimensional body scan data into apparel pattern development: exploring a new methodology”, unpublished MS thesis, Cornell University, Ithaca, NY. Winakor, G., Beck, M.S. and Park, S.H. (1990), “Using geometric models to develop a pattern for the lower bodice”, Clothing and Textiles Research Journal, Vol. 8 No. 2, pp. 49-55. Yavatkar, A. (1993), “Anthropometric shape analysis strategy for design of personal wear”, paper presented at the Human Factors and Ergonomics Society 37th Annual Meeting, October 11-15. Corresponding author Susan P. Ashdown can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Automated side-seam placement 213
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,4
Automatic slub detection using Gabor filters Xiuping Liu, Zhijie Wen, Zhixun Su and Shaogeng Yi
214
Department of Applied Mathematics, Dalian University of Technology, Dalian, People’s Republic of China
Received 8 April 2007 Revised 28 September 2007 Accepted 28 September Abstract 2007 Purpose – Automatic slub detection is vital in the classification and identification of fabric images.
This paper seeks to present a rapid and accurate approach for automatic detection of slub in fabric images using Gabor filters. Design/methodology/approach – Slub can be regarded as defects along weft or warp. Gabor filters as bandpass filters consider the directional characteristics of slub and its frequency spectrum after Fourier transform. Choosing appropriate parameters for Gabor filters, slub can be detected accurately. Findings – The proposed method achieves automatic detection of slub. The experimental results suggest that the authors approach is effective. Originality/value – This paper considers appropriate parameters to design a Gabor filter for automatic detection of slub. And it is helpful to classify and identify fabric images. Keywords Textile industry, Image processing, Error handling Paper type Research paper
International Journal of Clothing Science and Technology Vol. 20 No. 4, 2008 pp. 214-221 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810878838
Introduction Slub-yarn is a sophisticated yarn, whose slub appearance is gained via the variation of the yarn linear density during the spinning process and because of its special appearance, has been widely used in a variety of garments. The mechanical properties of slub-yarns, including the twist distribution, were analyzed in the past (Lu et al., 2006). Most of the studies on the slub-yarn concentrate on its parameters and the principle of formation (Wang and Huang, 2002), but slub detection often belongs to defect detection in textile materials stated by some researchers (Kumar and Pang, 2002; Zhang and Breese, 1995; Chan and Pang, 2000), and it is hard to achieve automation. In the fabric industry, conventional means of detecting slub usually adopt indirect testing and contrasting to the real fabric sample. These manual operations are usually tedious, time-consuming and easily tire an operator’s eyes. In our method, automatic slub detection is proposed. Moreover, we can get the parameters of slub according to the detecting results, such as slub length and intervals, etc. which are very important to the classification and identification of fabric images. Since a woven fabric can be regarded as a typical periodic image, Fourier transform is useful to analyse it (Xu, 1996; Su et al., 2006; Ravandi and Torimi, 1995). The peaks in the frequency domain after Fourier transform characterize periodicity of fabric image. Slub destroys the periodicity of the fabric and can be regarded as defects along the weft or warp. However, 2D Gabor filters have been shown to be particularly useful for This work is supported by National Natural Science Foundation of China (60673006), New Century Excellent Talents in University of China (No. NCET-05-0275). The authors would like to thank the anonymous reviewers for their useful comments.
analyzing texture images containing highly specific frequency or orientation characteristics (Dunn and Higgins, 1995; Bovik et al., 1990; Jain and Farrokhnian, 1991). They have been applied by Kumar and Pang (2002) to detect defects in textured materials and by Dunn and Higgins (1995) to segment texture. The 2D Gabor filters are multi-channel filters and are appropriate for textural analysis in several senses: they have tunable orientation and radial frequency bandwidths, and can achieve optimal joint localization in spatial and frequency domain. In the next section, a review of Gabor filters is presented, including the characteristic in the frequency domain of a typical periodic image after Fourier transform. Followed by the section that describes the slub detection method. The next section gives the experimental results, which are followed by the conclusion in the penultimate section. Gabor filters Gabor channel functions Inspired by the multichannel operation of the human visual system (HVS) for interpreting texture, research has been focused on using a multichannel approach based on Gabor filtering to simulate the operation of HVS for texture regions. Simple cells in the visual cortex were found to be sensitive to different channels of combinations of various spatial frequencies and orientations (Tan, 1995). Moreover, Gabor filters achieve optimal joint localization in spatial and frequency domain (Daugman, 1985). As mentioned above, Gabor filters can decompose the image into component images corresponding to different scales and orientations. Therefore, they have been used extensively for texture analysis (Dunn and Higgins, 1995; Bovik et al., 1990; Jain and Farrokhnian, 1991) and defect detection (Kumar and Pang, 2002). Gabor transform was first defined by Gabor (1946) and later extended to 2D by Daugman (1985). In the spatial domain, the Gabor function is a Gaussian modulated by a complex sinusoid: hðx; yÞ ¼ gðx0 ; y0 Þ · cosð2pjðUx þ VyÞÞ; 0
ð1Þ
0
where ðx ; y Þ ¼ ðx cos f þ y sin f; 2x sin f þ y cos fÞ. U and V are frequencies along the x- and y- axes, and: 1 ðx=lÞ2 þ y 2 exp 2 gðx; yÞ ¼ ; 2pls 2 2s 2 and syffi where lðl ¼ sx =sy Þ is the aspect ratio of Gaussian function, while s pxffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi characterize the spatial extent and bandwidth of the filter, respectively. F ¼ U 2 þ V 2 (cycles/image width) is the radial center frequency and the orientation is u ¼ tan21 ðV =U Þ (degrees or radians measured from u-axis). It is usually convenient to consider filters whose modulating Gaussians have the same orientation as the complex sine grating (f ¼ u). In this case, the Gabor function (1) reduce to: hðx; yÞ ¼ gðx0 ; y0 Þ · cosð2pjFx0 Þ: In the frequency domain, Fourier transform of the Gabor functions is: 1 H ðu; vÞ ¼ exp{22p 2 s 2 ½ðu0 2 FÞ2 l 2 þ ðv0 Þ2 } 2 1 þ exp{22p 2 s 2 ½ðu0 þ FÞ2 l 2 þ ðv0 Þ2 }; 2
ð2Þ
ð3Þ
Automatic slub detection using Gabor filters 215
IJCST 20,4
216
where ðu0 ; v0 Þ ¼ ðu cos f þ v sin f; 2u sin f þ v cos fÞ. Thus, the Gabor frequency response has the shape of two Gaussians. One of the Gaussian’s major and minor axis widths are determined by sx and sy, it is rotated at an angle u from the positive u-axis and centered about the frequency (U, V ). Thus, Gabor filters act as bandpass filters. Figure 1 shows intensity plots of several 2D Gabor functions having this “daisy petal” configuration. There are a total of 40 different Gabor filters with l ¼ 0:5; B ¼ 0:6; F ¼ ½4; 8; 16; 32; 64; u ¼ kp=8; k ¼ 0; 1; . . . ; 7. When l ¼ 1, Gabor filters are symmetric filter. Because most of real fabric images contain textels (Kumar and Pang, 2002) not arranged in a square lattice, asymmetric Gabor filters are more useful. For a given input image I(x, y), the magnitude of filtered image f(x, y) is: f ðx; yÞ ¼ I ðx; yÞ £ hðx; yÞ;
ð4Þ
where h(x, y) is the Gabor filter defined in equation (2). Properties of Gabor filters The crucial step in the application of Gabor filters is the choice of the filter parameters. There are large numbers of papers stating approaches of parameters choosing (Kumar and Pang, 2002; Dunn and Higgins, 1995). The parameters of Gabor filters are usually chose after consideration of spatial frequencies and orientations of detected features. The radial frequency bandwidth B is defined as: ðpF ls þ aÞ ðoctavesÞ; ð5Þ B ¼ log2 ðpF ls 2 aÞ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi where a ¼ ðln 2Þ=2. By varying the free parameters of B, F, u, l, Gabor filters with arbitrary orientation and bandwidth characteristics can be generated by spanning any oval region of the frequency domain.
Figure 1. The filter set in the frequency domain (256 £ 256)
Notes: There are a total of 40 Gabor filters. The origin is at (row, col) = (128,128)
The most important property of Gabor filters is that they are sensitive to the orientation of detected features, which is shown by the model image in Figure 2. The directional analysis algorithm developed using Gabor filters consists of the following steps: first, generate a set of Gabor filters (in the spatial domain) tuned to a set of angle bands. Then, convolve the Gabor filters with the given image, which results in a set of component images, one for each angle band. For example, we programme a Gabor filter to 908. The filtered image is shown in Figure 2. From this example, we can find that Gabor filter is useful to analyze the directional features in the image.
Automatic slub detection using Gabor filters 217
Slub detection algorithm Choice of filter parameters There are two kinds of implementations of Gabor filters, one in the spatial domain, and the other in the frequency domain. According to equation (2), we first decide four parameters l, s, f, F to make the filter fit to the specific texture image, and then convolve the image with the chosen filter. But it is hard to find proper s, F in the spatial domain. In order to choose parameters easily, we implement filtering process in frequency domain. Owing to the convolution theory of Fourier transform, it follows that: I ðx; yÞ £ hðx; yÞ ¼ ifftðFðu; vÞH ðu; vÞÞ
ð6Þ
where ifft is the inverse Fast Fourier transform. Then we just need to obtain the parameters s, F, u, l. As the textels (Kumar and Pang, 2002) in real fabric image are not arranged in a square lattice, let l ¼ 0.5. F is obtained by the following operations. We must first inspect the characteristic of the frequency spectrum of slub-yarn. For a fabric image without slub, its frequency spectrum has six centrosymmetric peaks during one period after eliminating the central peak. The values and location of peaks (high-magnitude areas) reveal the periodicity and orientation of the periodic structure in the fabric image, respectively, (Su et al., 2006). For the slub-yarn, its frequency
(a)
(b)
Figure 2. Model image and the segmentation of filtered image with the Gabor filter at 908
IJCST 20,4
218
spectrum has more peaks around the center corresponding to slubs (enclosed by ellipse in Figure 3(b)) except for the six peaks mentioned above. We only need to choose the location of the maximum among these peaks as the central frequency and the orientation (warp or weft) as the orientation of the chosen filter (Figure 3(c)). Moreover, the location of central frequency is similar for the different slub-yarn. Finally, set B to the empirical data 2. s is obtained by formula s ¼ ðað2B þ 1ÞÞ=ðpF lð2B 2 1ÞÞ deducted from equation (5). And slubs in the real fabric samples are distributed along warp (u ¼ 0) and weft orientation (u ¼ p/2). They are substantially thicker than individual yarns in the fabric, so they can be regarded as directional defects in the fabric. Taking advantage of the bandpass technique of Gabor filters, this approach can restrain the frequency component of the normal texture in the fabric, and strengthen the frequency component of slubs. The parameters are similar for different slub-yarn, so our method achieves the detection of slub automatically. Figure 4 shows the detection result of Figure 3. The left image shows successful slub detection. For exhibiting the accuracy of our method, we also show the overlay result (obtained by covering original image with slub image) on the right. More experimental results are shown in the next section. The implementation of algorithm The slub detection algorithm developed using Gabor filters consists of the following steps:
Figure 3. (a) Original image; (b) frequency spectrum after Fourier transform; (c) filter shape with parameters chosen above; (d) the filtered image
(a)
(b)
(c)
(d)
Automatic slub detection using Gabor filters 219
(a)
(b)
(1) Get the frequency spectrum F(x, y) of a real fabric image sample I(x, y) which is transformed by FFT. (2) Choose the parameters s, F, u, l of Gabor filter H(u, v) for equation (3). (3) Generate the filtered image by the following operation: f ðx; yÞ ¼ ifftðFðu; vÞH ðu; vÞÞ where ifft is the inverse Fast Fourier transform. (4) Get the threshold according to the histogram of the filtered image. (5) Remove the unwanted spectral components from the filtered image obtained in Step (3). (6) Using morphological operations to eliminate the noise in the image obtained in Step (5), we finally get the slub distribution image finally. Experimental results The performances of slub detection described in the above section are evaluated on fabric samples in Figure 5. The left images in Figure 5 are the images containing slub, the middle images are the original images. The overlay results are shown on the right in Figure 5. From overlay results, we can see the efficiency of our approach. Because we just choose one filter to deal with one image, this method is rapid too. In this experiment, 32 real fabric samples are used to examine our method. The efficiency of this approach is illustrated with 99 percent accuracy in slub detection. According to the detection results, we can get the parameters of slubs, for example, slub length, slub intervals, etc. These parameters can be trained with a self-organizing feature map. The networks have different output maps for different input states, which can be used for classifying slub-yarns. Conclusion In our paper, an approach for automatic identification of slub defects is studied. The approach is based on Gabor filters, which are bandpass filters and sensitive to the
Figure 4. (a) Slub image; (b) overlay results
IJCST 20,4
220
Figure 5. Figure 5 (b), (e), (h), (k): fabric samples with slub; (a), (d), (g), (j): corresponding segmented images with slub using a Gabor filter; (c), (f), (i), (l): overlay results
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
(j)
(k)
(l)
directional features in the image. Regarding slubs as directional defects, we choose the Gabor filter to detect them in the slub-yarn. At last, this paper confirms the performance of the proposed method by testing 32 real fabric samples. References Bovik, A.C., Clark, M. and Geisler, W.S. (1990), “Multichannel texture analysis using localized spatial filters”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 12 No. 1, pp. 55-73. Chan, C.H. and Pang, K.H. (2000), “Fabric defect detection by Fourier analysis”, IEEE Transactions on Industry Applications, Vol. 36 No. 5, pp. 1267-76. Daugman, J.G. (1985), “Uncertainty relations for resolution in space spatial frequency and orientation optimized by two-dimension visual cortical filters”, Journal of the Optical Society of America, Vol. 16 No. 2, pp. 1160-9. Dunn, D. and Higgins, W. (1995), “Optimal gabor filters for texture segmentation”, IEEE Transactions on Image Processing, Vol. 4 No. 7, pp. 947-64. Gabor, D. (1946), “Theory of communication”, Journal of the Institute of Electrical Engineers, Vol. 93 No. 26, pp. 429-57. Jain, A.K. and Farrokhnia, F. (1991), “Unsupervised texture segmentation using Gabor filters”, Pattern Recognition, Vol. 24 No. 12, pp. 1167-86. Kumar, A. and Pang, G. (2002), “Defect detection in textured materials using Gabor filters”, IEEE Transactions on Industry Application, Vol. 38 No. 2, pp. 425-40. Lu, Y.Z., Gao, W.D. and Wang, H.B. (2006), “A model for the twist distribution in the slub yarn”, International Journal of Clothing Science & Technology, Vol. 19 No. 1, pp. 36-42. Ravandi, S.A.H. and Toriumi, K. (1995), “Fourier transform analysis of plain weave fabric appearance”, Textile Research Journal, Vol. 65 No. 11, pp. 676-83. Su, Z.X., Wen, Z.J., Qiao, W.S., Yi, S.G. and Shi, X.Q. (2006), “Automatic identification of the fabric structure based on Fourier transform”, Journal of Information and Computer Science, Vol. 3 No. 3, pp. 527-34. Tan, T.N. (1995), “Texture edge detection by modeling visual cortical channels”, Pattern Recognition, Vol. 28 No. 9, pp. 1283-98. Wang, J. and Huang, X.B. (2002), “Parameters of rotor spun slub yarn”, Textile Research Journal, Vol. 72 No. 1, pp. 12-16. Xu, B.G. (1996), “Identifying fabric structure with Fourier transform technique”, Textile Research Journal, Vol. 66 No. 8, pp. 496-506. Zhang, Y.X. and Breese, R.R. (1995), “Fabric detection and classification using image analysis”, Textile Research Journal, Vol. 65 No. 1, pp. 1-9. Corresponding author Zhijie Wen can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Automatic slub detection using Gabor filters 221
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,4
Shrinkage prediction of plain-knitted fabric based on deformable curve
222
Zhixun Su, Xiaojie Zhou, Guohui Zhao and Xiuping Liu
Received 8 April 2007 Revised 7 June 2007 Accepted 7 June 2007
Department of Applied Mathematics, Dalian University of Technology, Dalian, People’s Republic of China, and
Ka-Fai Choi Institute of Textiles & Clothing, The Hong Kong Polytechnic University, Kowloon, Hong Kong Abstract Purpose – The aim of this paper is to develop a new method to predict the potential shrinkage of plain-knitted fabric. Design/methodology/approach – The presented method is based on deformable curve. The delivered plain-knitted fabric is represented as a deformable parametric curve, and the relaxed fabric can be reached by minimizing the energy of the curve. Compared to the delivered-knitted fabric, the length and width shrinkage percentages can be calculated accordingly. Findings – The new method is more convenient than the traditional trial and error method, and need less-input parameters than the STARFISH technique. Experimental results show that this method is feasible. Originality/value – This paper presents a new method for shrinkage prediction of plain-knitted fabric based on deformable curve and energy minimum. The work can be linked with shrinkage control in textile industry. Keywords Textiles, Predictive process, Fabric testing Paper type Research paper
1. Introduction Knitted fabric is a very popular product in textile industry, and it is faced with ever-rising demands for better quality and reliability. One of the key demands is dimensional stability, i.e. low levels of potential shrinkage. Potential shrinkage is expressed as a percentage of the dimensions of the reference fabric: SL ¼
International Journal of Clothing Science and Technology Vol. 20 No. 4, 2008 pp. 222-230 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810878847
ðcpcr 2 cpcd Þ ðwpcr 2 wpcd Þ · 100% SW ¼ · 100% cpcr wpcr
ð1Þ
where SL, SW are length and width shrinkages, respectively, cpcr (courses per cm) and wpcr (wales per cm) are the course and wale densities in the reference fabric (after relaxation), cpcd and wpcd are corresponding values of the delivered fabric (before relaxation). The reference sample is the fabric after the standard relaxation procedure as the STARFISH project, i.e. five cycles of washing and tumble drying under closely prescribed conditions (Heap et al., 1983). Project supported by New Century Excellent Talents in University in China (No. NCET-05-0275).
Since traditional trial and error methods are too costly and uncertain for shrinkage control, many researchers studied this problem from different point of views (Postle, 1968; Postle and Munden, 1967; Shanahan and Postle, 1970; Knapton et al., 1975; Lo, 1981; Doyle, 1953; Munden, 1959). Postle et al. (Postle, 1968; Postle and Munden, 1967) and Shanahan and Postle (1970) treated the yarn as an elastic object and proposed the theoretical models of knitted fabrics. Experimental observation made by Doyle (1953) confirmed the dependence of the fabric area on the loop length of plain-knit fabrics. The dimensions of a knitted fabric are believed to be predominately determined by the loop length and independent of other parameters such as yarn and knitting parameters. k-values are designed to describe the relationship between the dimensions and loop length. However, the k-values are affected quite significantly by several factors including especially certain aspects of the yarn specification and wet processing. As a result, the k-values have limitations for the prediction of fabric dimensions due to a poor precision. Instead of k-values, The STARFISH project is founded on database, in which the effects of many factors, such as yarn type, yarn count, and fibre type, are taken into account. It is based on statistical analysis of fabrics, and yarn and knitting parameters have to be known as input. What is more, material properties of yarn, which have significant effects on fabric dimensions, are not considered in their project. In this paper, a new method for shrinkage prediction of plain-knitted fabric is proposed based on deformable curve. The plain-knitted fabric is represented as a deformable parametric curve. The reference state of the fabric is obtained by energy minimization, and the shrinkage percentages can be achieved by the analysis of the delivered and the reference fabrics. We need less-input parameters to make our method work. 2. Parametric model of plain-knitted fabric Many researchers discussed the geometric models of knitted fabric (Postle, 1968; Postle and Munden, 1967; Peirce, 1947; Leaf and Glaskin, 1955; Choi and Lo, 2003; Demiroz and Dias, 2000). These models lay emphasis on the visualization of fabric. We need a parametric model suitable for shrinkage prediction, which can describe the plain-knitted fabric accurately with several variables. Choi and Lo (2003) presented a parametric model. In their model, the fabric loop is treated as symmetrical parts, and the central axis of the left part is represented as: xðtÞ ¼ a · t 3 2 1:5 · t 2 þ 0:5 · ða þ wÞ · t
ð2:1Þ
yðtÞ ¼ 0:5 · ðc þ 2eÞ · ð1 2 cos p tÞ
ð2:2Þ
zðtÞ ¼ 0:5 · ðth 2 2rÞ · ð1 2 cos 2p tÞ
ð2:3Þ
where w is the loop width, (c þ 2e) is the loop height, e is the adjacent loop overlapping distance, tk is the fabric thickness, r is the radius of the yarn, a is a free parameter, t [ [0, 1] is the parameter of the space curve (shown as Figure 1(a)). Suppose A is the extremum point of x(t) in the interval t [ [0.5, 1], and B is the extremum point of x(t) in the interval t [ [0, 0.5] of the adjacent loop (Figure 1). The above-parametric model assumes that y(t) component of A is the same as B. However, we observe that vertical shift may occur due to relaxation procedure. So, a new variable s is introduced to describe the vertical shift, that is: yðtÞ ¼ 0:5 · ðc þ 2e þ sÞ · ð1 2 cos ptÞ
ð2:20 Þ
Shrinkage of plain-knitted fabric 223
IJCST 20,4
e e
c
224
c
A
B
s
e w
e w
Figure 1. Parametric model of fabric loop
(a) Geometric model
(b) The modified geometric model
(c) Knitted fabric
Source: Choi (2003)
equations (2.1), (2.20 ) and (2.3) give the modified parametric model (Figure 1). Similar to the original parametric model, e can be denoted by a, w, c, s, i.e. ðc þ sÞ 1 21 ; e¼ 2 cos pb where:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 1 a 2 2w b¼ 2 : 2 2 3a In the modified geometric model, the five independent variables a, w, c, s, tk determine the plain-knitted fabric. 3. Energy model Kass et al. (1987) introduced deformable curve model, named active contour models or “snakes,” for solving various problems in computer vision and image analysis. In the active contour models, curves are used to deform or define edges or contours or to track motion in a moving image driven by energy. The energy of a curve is composed of internal energy, including stretching and bending energy ensuring its smoothness, and the external energy pulling it to the edges. In our context, fabric loop is represented as a spatial curve, besides the stretching and bending energy, twisting energy should be considered. The internal energy of the loop is: ð3Þ U int ¼ U t þ U t þ U k ; where Ut, Uk, and Ut are stretching, bending and twisting energy, respectively. We call the yarn unraveled from a finished fabric the natural state of a loop, or natural loop. And we observed that it is very rare that the loop of natural state is straight, it looks wavy. So, the energy of the loop should be relative to natural state. Stretching energy due to elongation can be expressed as: Z E Ly 2 e ds; Ut ¼ 2 0 y
where E is the tensile modulus, L, is the length of loop, ey is the yarn strain. By assuming that the yarn extension is constant along the yarn and the loop length does not change significantly, the stretching energy can be rewritten as: Ut ¼
Ee2y L2y EðLy 2 Ly0 Þ2 < ; 2 2
ð4Þ
where Ly0 is the length of natural loop. Bending energy illustrates the bending degree of the yarn deviating from the natural state: Uk ¼
B 2
Z
Ly
ðk 2 k0 Þ2 ds;
ð5Þ
0
where B is the bending modulus, k and k0 are the curvatures of the loop and the natural loop, respectively. Twisting energy describes its twisting degree deviating from the natural state: C Ut ¼ 2
Z
Ly
ðt 2 t0 Þ2 ds;
ð6Þ
0
where C is torsional modulus of the loop, t and t0 are the torsions of the loop and the natural loop. For the shrinkage prediction of fabric, the reference state is the fabric after relaxation, and no additional load on the fabric. The external energy is due to yarn jamming in the fabric. Larger jamming volume brings larger energy, so the external energy can be expressed as: U ext ¼ P
N X
Vj
ð7Þ
j¼1
where N is the number of loops which contact with the current loop, Vj is the jamming volume with loop j. Although the jamming volume can be accurately calculated by intersecting volume of two polyhedrons, it is too costly and not necessary. By assuming that the yarn near the contact point can be treated as a cylinder, the jamming volume can be approximately calculated by the intersecting volume of cylinders. First, we have to find the contact point, i.e. the point where the distance of the central axis of two cylinders attains minimum. We define the corresponding distance function of two loops as: ds : ½0; Ly £ ½0; Ly ! R
d s ðs1 ; s2 Þ ¼ dðX 1 ðs1 Þ; X 2 ðs2 ÞÞ
Shrinkage of plain-knitted fabric
ð8Þ
where d (· , · ) compute the distance of two points in the space, si, X i(si), i ¼ 1, 2 are the arc length and the spatial coordinates at si respectively. The contact point is where ds ðs1 ; s2 Þ , 2r and ds ðs1 ; s2 Þ attain extremum. In practice, possible contact region can be estimated beforehand, and local distance function is calculated to determine the contact point (Figure 2). According to calculus, the intersecting volume of two cylinders with radius r is:
225
IJCST 20,4
Z
Y
p
X
226
2 1.5 1 0 3
1
Figure 2. Sketch map of local distance function
s2
2
5
4 s1
3 (a) Possible contact region
V ¼4
Z
(b) The distance of possible contact region r
Z
d2r 0
4 ¼ sin u
Z
r
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi r 2 2 x 2 dy dx pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiqffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi r 2 2 x 2 r 2 2 ðx 2 d Þ2 dx
ð1=sin uÞ
r 2 2ðx2d Þ
ð9Þ
d2r
where d is the distance of the central axis of the two cylinders at contact point, u is the angle between the two cylinders. The algorithm of computing the jamming volume of loops is as follows: (1) Initialize N, and the numbers Ml of possible contact regions for loop lðl ¼ 1; L N Þ, and jamming volume V ¼ 0, i ¼ 0, and j ¼ 0. (2) If i $ N, stop; else goto (3). (3) For one possible contact region Aij, compute the distance function, and get the contact point by searching for the extremum of the distance function. (4) Compute the angle between two loops at the contact point. (5) Compute intersecting volume Vij from equation (9), and V ¼ V þ Vij. If j $ Mi, i ¼ i þ 1, goto (2), else j ¼ j þ 1, goto (3). The jamming volume can be calculated by the above method, however, the integral may cause computation complexity, an alternative method is to take d 3 as the approximation of the jamming volume. It is not as accurate as the above method. According to the principle of minimum potential energy, the fabric is in stable equilibrium, i.e. the reference state, while the total energy is at a minimum. So, the reference fabric can be achieved by minimizing the total energy. Taking the geometric meanings of the variables in the parametric model of the loop, we have w . 0, c . 0, and tk $ 2r. Considering the trend of x component along t, we have: dx 1 ,0 ð10Þ dt 2
From equations (2.1)-(10), we get: 3a · 0:52 2 3a · 0:5 þ 0:5ða þ wÞ , 0
ð11Þ
that is a . 2w. So, a constrained optimization: ( min U ð12Þ
w; c . 0; th $ 2r; a . 2w
Shrinkage of plain-knitted fabric 227
is utilized to compute the reference fabric, where U ¼ Uint þ Uext is the total energy. The constrained optimization can be converted to an optimization problem without constraints through punish function. And gradient descent method or quasi-Newton method can be used to solve it. After the variables a, w, c, s, tk of the reference fabric are obtained, the shrinkage percentages can be calculated by equation (1). 4. Determination of k0 and t0 The determination of k0 and t0 is important to the energy computation. In practice, it is difficult to get the shape of natural loop. Choi and Lo (2003) defined the degree of set c to describe the shape of natural loop relative to the loop of the delivered fabric. c ¼ 0 illustrates the natural loop is a straight line, and c ¼ 1 shows that the natural loop is the same as the loop of the delivered fabric. By assuming that k0 =ks ¼ t0 =ts ¼ c, k0, t0 can be obtained by:
k0 ¼ ks · c; t0 ¼ ts · c
ð13Þ
where ks and ts are the curvature and torsion of a standard loop. According to the curve theory in differential geometry, the shape of a spatial curve, the natural loop here, is determined by the curvature and torsion. While the shape obtained through the above k0, t0 is not the natural loop obviously (shown as Figure 3). So, we need a new method to compute k0 and t0. Instead of determine them by degree of set, we compute them by estimating the natural loop directly. Suppose that the natural loop can also be represented by the parametric model in Section 2, and denote h ¼ c þ 2e þ s. The natural loops are put between two glasses (shown as Figure 4), then we can measure the width w 0 and height h 0 at this state, assume that the projection of the natural loop onto the yOz plane is an arc of a circle (shown as Figure 5), then we have: Z
Z
X
(a) The natural loop of Choi's method
Y
X
(b) The natural loop of the new method
Y
Figure 3. The natural loop
IJCST 20,4
Ru ¼
h 0 2
R cos u þ t h0 2 d ¼ R
h0 ¼ 2R sin u
ð14Þ
Since the width of the loop almost maintain constant before and after put into glasses, we assume w0 ¼ w 0 . And the thickness of natural loop is between the yarn diameter and the thickness of original loop, and can be approximated by:
228 th0 ¼
2r w 0 þ 2r: Ly
ð15Þ
From equations (14) and (15), we can get w0 ; th0 ; h0 , and, a0 can be computed by solving the non-linear equation: Z 1 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðx0 ðtÞÞ2 þ ð y 0 ðtÞÞ2 þ ðz 0 ðtÞÞ2 dt: Ly0 ¼ 0
5. Data acquisition and experiments We need only measure some properties of the delivered fabric and the yarn unraveled from it to make our method work. First, it is easy to measure the yarn count, wpc, pffiffiffiffifficpc, ffi and loop length, then the yarn diameter is approximated by D ¼ 2:54=23 Ne, and the width and the course of a loop are w ¼ 1/wpc and c ¼ 1/cpc. w0, h0 can be measured as discussed in Section 4. For simplicity, the corresponding data are divided by yarn diameter. The real length and width percentages can be obtained by measuring the fabric before and after relaxation. The prediction results are compared to them to illustrate the validity of the present method. We arbitrarily choose 14 plain-knitted fabrics, the tensile, bending and torsional modules are set to be constants. The prediction results are listed in Table I, the subscript d, r, p represent the values corresponding to the
Figure 4. The natural loop between glasses
h0 t h0 - d
Figure 5. The sketch map of the projection of natural loop on yOz-plane
h0 q
2 R
2
No
Ly
Ne
wpcd
cpcd
wpcr
cpcr
wpcp
cpcp
SW
SL
SWp
SLp
1 2 3 4 5 6 7 8 9 10 11 12 13 14
0.307 0.300 0.302 0.308 0.258 0.256 0.259 0.264 0.266 0.267 0.263 0.271 0.265 0.304
19.9 19.3 19.9 20.1 26.5 26.2 27.6 26.0 31.8 31.5 30.3 31.7 31.5 20.0
11.70 11.76 10.99 11.46 14.44 14.55 14.39 14.04 14.34 14.13 14.65 14.49 13.84 11.05
16.95 17.47 17.47 17.02 20.30 20.20 20.20 20.10 18.69 19.42 19.05 18.52 18.52 17.70
12.46 12.62 12.62 12.50 15.33 15.27 15.63 14.98 15.15 15.15 15.27 15.38 15.15 12.31
18.02 18.69 18.43 18.60 21.51 21.51 21.28 22.10 19.70 20.41 19.80 19.61 19.60 18.35
13.35 13.53 14.46 14.05 16.34 16.48 16.80 15.93 16.13 16.48 16.22 16.47 16.73 13.34
18.52 19.44 18.91 19.83 22.16 22.38 22.65 23.94 20.43 20.68 19.82 19.94 20.03 18.58
6.14 6.76 12.91 8.31 5.78 4.73 7.91 6.32 5.38 6.71 4.03 5.80 8.65 10.22
5.93 6.55 5.24 8.51 5.58 6.06 5.05 9.05 5.14 4.85 3.81 5.56 5.52 3.54
7.10 7.20 14.62 12.41 6.65 7.95 7.55 6.36 6.46 8.75 6.24 7.08 10.40 8.36
2.80 4.00 2.58 6.60 3.05 4.09 6.47 8.33 3.67 1.34 0.11 1.67 2.18 1.28
delivered, reference and predicted fabric. There are three samples with error within 2 percent, 7 within 3 percent, and almost all of them within 4 percent. 6. Conclusion In this paper, we present a shrinkage prediction method for plain-knitted fabric. The reference fabric is obtained by minimizing the total energy to predict the shrinkage. Experimental results show that this method is feasible, but the precision need to be promoted. Further research focus on the measurement of the actual tensile, bending and torsional modules of the yarn, and we believe that accurate measurement of the material properties help to improve the prediction precision. The optimization in Section 4 is a nonlinear constrained optimization, finding efficient algorithm to solve it is also an important issue. References Choi, K.F. and Lo, T.Y. (2003), “An energy model of plain knitted fabric”, Textile Research Journal, Vol. 73 No. 8, pp. 739-48. Demiroz, A. and Dias, T. (2000), “A study of the graphical representation of plain-knitted structure, Part I: stitch model for the graphical representation of plain-knitted structures”, Journal of the Textile Institute, Vol. 91, pp. 463-80. Doyle, P.J. (1953), “Fundamental aspects of the design of knitted fabrics”, Journal of the Textile Institute, Vol. 44 No. 8, pp. 561-78. Heap, S.A., Greenwood, P.F., Leah, R.D. and Eaton, J.T. (1983), “Prediction of finished weight and shrinkage of cotton knits: the Starfish Project Part I: an introduction and general overview”, Textile Research Journal, Vol. 53, pp. 109-19. Kass, M., Witkin, A. and Terzopoulos, D. (1987), “Snakes – active contour models”, International Journal of Computer Vision, Vol. 1 No. 4, pp. 321-31. Knapton, J.J.F., Truter, E.C. and Aziz, M.A. (1975), “The geometry, dimensional properties and stabilization of the cotton plain jersey structure”, Journal of the Textile Institute, Vol. 66, pp. 413-9. Leaf, G.A.V. and Glaskin, A. (1955), “The geometry of the plain knitted loop”, Journal of the Textile Institute, Vol. 46, pp. 587-605.
Shrinkage of plain-knitted fabric 229
Table I. Fabric parameters and the prediction results
IJCST 20,4
230
Lo, T.Y. (1981), “Dimensional properties of weft knitted fabrics from blended yarn”, PhD dissertation, Huddersfield Polytechnic, Huddersfield. Munden, D.L. (1959), “The geometry and dimensional properties of plain-knit fabric”, Journal of the Textile Institute, Vol. 50, pp. 448-71. Peirce, F.T. (1947), “Geometrical principles applicable to the design of functional fabrics”, Textile Research Journal, Vol. 17, pp. 123-47. Postle, R. (1968), “Dimensional stability of plain knitted fabrics”, Journal of the Textile Institute, Vol. 59, pp. 65-77. Postle, R. and Munden, D.L. (1967), “Analysis of the dry relaxed knitted loop configuration”, Journal of the Textile Institute, Vol. 58, pp. 329-65. Shanahan, W.J. and Postle, R. (1970), “A theoretical analysis of plain knitted structure”, Textile Research Journal, Vol. 40, pp. 656-65. Corresponding author Zhixun Su can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Seam pucker indicators and their dependence upon the parameters of a sewing machine Milda Jucie˙ne˙ and Vaida Dobilaite˙ Kaunas University of Technology, Kaunas, Lithuania
Seam pucker indicators
231 Received 20 January 2008 Revised 24 February 2008 Accepted 24 February 2008
Abstract Purpose – The paper aims to establish and introduce seam pucker assessment indicators and their dependence upon the fabric characteristics and parameters of a sewing machine. Design/methodology/approach – Seams pucker defect is specific to garment from lightweight fabrics, therefore the investigation was performed with this kind of fabrics. The structure and friction properties of fabrics were determined. The influence of rotational frequency of the main shaft (varied from 200 till 2,300 min2 1) and pressing force (from 25 till 85 N) on seam pucker was analysed. The seam pucker was evaluated by characteristic of sharpness calculated as ratio of pucker height and length. Findings – Analysis of the obtained results allows stating that pucker sharpness increases with growing rotational frequency of the main shaft and decreases, increasing pressing force. On the basis of correlation analysis, the relationship between proposed new indicator of seam pucker evaluation and parameters of sewing machine was sought. The obtained result demonstrates that in most cases the relative pucker coefficient is dependent upon rotational frequency of the main shaft: increasing rotational frequency also leads to higher pucker coefficient. The linear relationship between the pucker coefficient and pressing force was not observed. Practical implications – This investigation has practical implications in the clothing and other nearly related industries. In the paper, results involved with evaluation of seam pucker are presented. Originality/value – The research showed that in order to define seam pucker more comprehensive, it is necessary to evaluate not only ratio of pucker height and length, but also the quantity of puckers as well as their propagation. Considering to this, the new criterion for seam pucker evaluation was proposed. Keywords Textiles, Textile technology, Force, Friction, Textile making-up processes Paper type Research paper
Introduction One of the parameters for assessing, the quality of aesthetical thread joins is puckering, i.e. formation of a creasy surface (formation of disordered, undesirable folds) of the sewing garment along the seam. Pucker is often observed when sewing thin, dense fabrics; it impairs the final appearance of an article. In the course of sewing, textiles are exposed to the complicated impact of outer forces: fabric is pierced with a needle, then an upper sewing thread (of a needle) is inserted, it is interlaced with a bottom thread, a thread loop is retracted inside the sewing fabrics. Then fabrics, pressed between a toothed plate and pressing foot, are shifted by stitch length. In the tightened stitch, sewing threads press fabric layers, whereas due to tensioning and later due to deformation relaxation processes, compress them in the longitudinal direction between adjacent pricks. There are several studies of the investigation of the mechanism of seam pucker occurrence (Stylios and Lloyd, 1989a, b; Kawabata and Niwa, 1998). Whether a seam under the impact of the aforementioned
International Journal of Clothing Science and Technology Vol. 20 No. 4, 2008 pp. 231-239 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810878856
IJCST 20,4
232
processes and forces acting during them will crease or not, depends upon the fabric being sewn, properties of sewing threads, and performances of a needle, sewing machine equipment and technological modes (Stylios and Lloyd, 1990; Fan and Leeuwner, 1998; Park and Kang, 1999a, b; Dobilaite˙ and Petrauskas, 2002; Dobilaite˙ and Jucie˙ne˙, 2006). If it is possible to change these factors, value of puckering may be decreased; technological sewing modes often are the main parameters variation whereof is rational in the manufacture process of sewing garments. Seam pucker is a relevant problem subject to a range of factors, therefore for its evaluation a variety of methods is offered. In a subjective way, this defect is assessed referring to the photographs of seam evenness: under standard lighting conditions, respondents compare samples of seams with the seams captured in photos and featuring a different level of pucker (ISO 7770, 1985). For objective assessment of seam pucker, various methods and instruments have been established (Stylios and Lloyd, 1989a, b; Stylios and Sotomi, 1993a, b, c, Stylios, 1998). Up-to-date instruments are required that have been continuously improving with the development of computer technologies. For instance, currently images are read by laser technologies (a scanner, amplifier, analog-to-digital converter), data are processed by computer software (Park et al., 1997; Park and Kang, 1999a, b; Fan et al., 1999a, b). Measures are taken to assess seam pucker not only on a flat surface, when an investigated test sample is placed on a plane, but also in a 3D article (Fan and Liu, 2000). In the context of objective assessment of seam pucker, an issue of quantitative evaluation criteria remains relevant. The aim of this paper is to establish and introduce seam pucker assessment indicators and their dependence upon the fabric characteristics and parameters of a sewing machine. Materials and methods Seams pucker defect is specific to an article from lightweight synthetic fabrics, when creasy surface emerge in the area of part joint and downgrade a final look of a garment. To this the lightweight fabrics having different structure, mechanical properties were chosen for the investigation. Basic characteristics of the used fabrics are listed in Table I. The fabric settings Pwa, Pwe were determined in accordance with EN 1049-2, surface density W in accordance with ISO 3801, as well as fabric thickness T2 (at pressure of 196 Pa) was measured according to the FAST system. The final result was calculated as arithmetic average of five specimen results, confidence limit does not exceed 5 per cent. In the case of fabric thickness, measurements confidence limit does not exceed 3 per cent. The fabric structure characteristics were supplemented by Peirce cover factors, which were computed considering fibre density of the fabric’s threads and overall density. Setting (dm2 1) Fabric symbol
Table I. Basic characteristics of the investigated fabrics
M1 M2 M3
Composition 100 per cent PES 100 per cent PES 45 per cent PES, 55 per cent CV
Weave structure
Surface density W (g/m2)
Pwa
Pwe
Thickness T2 (mm)
Combined Plain Combined
149 130 111
650 490 560
380 260 260
0.35 0.32 0.46
Also, friction has an influence on the changes of fabric structure after sewing process. During investigation, the sliding friction force was measured using a tensile-tester equipped with a special device. The stainless metal sled was dragged at a constant speed along the fabric mounted over a horizontal platform. An experiment was performed having the sled contact between fabric and metal (variation of the result seek 3.3 per cent), then covering the sled with investigated fabric (variation was 4.6 per cent). The fabric friction properties was characterized by coefficient of friction m, which was computed on the well-known relationship F ¼ mN, where F is frictional force, N is normal force. The fabrics cover factors and coefficients of friction of the used fabrics are presented in Table II. The specimens of 30 £ 3 cm dimensions were cut in warp (Wa) and weft (We) directions for seam pucker investigations. Two specimens (Wa þ Wa, We þ We) were sewn together across the centre line in longitudinal direction using lockstitch (301 type) Unicorn sewing-machine. The specimens were sewn with Gu˝termann Nm 120 sewing threads and needle no. 90. Stitch length was 2.5 mm, thread tension was chosen to form well-balanced stitch. The experiment was performed when rotational frequency of a sewing machine main shaft w was 200, 900, 1,600 and 2,300 min2 1, and pressing force P was 25, 45, 65 and 85 N. In order to estimate magnitude of seam pucker at different sewing regimes, a crease generated in a seam was analysed as a 3D wave, obtained during intersection of wavy surface and plane, perpendicular to the fabric and lying along the stitch. According to this, the sewn specimens were photo-captured from both sides by digital camera recording the view of the shape of generated wave and the geometric characteristics (wavelength l, and wave height h) of this shape were measured. Mentioned geometric characteristics of seam pucker were measured from both sides in the length of 20 cm of the specimen. The wave height h was measured as the distance between the highest wave’s contour point to the lowest contour point of this wave; and the wavelength was measured as the distance between point at which the wave starts to rise up (h $ 0) to its lowermost point (h ¼ 0). The final result of geometric characteristics of one fabric seam pucker was taken as the all heights hn and wavelengths ln average of a fabric specimens, where n is the wave amount. When human eyes see a garment with puckered seams, this defect is perceived according to areas of bright and dark parts of the wrinkled surface of the fabric along the seam line. The apprehension of pucker gauge depends on an angle between the humane eyes and the light source, the length and the amplitude of the wrinkle propagated in the space. However, the same length puckers may have a different amplitudes, and conversely (Figure 1). Stylios and Sotomi (1993a, b, c) with reference to cognitive psychology in seam pucker assessment noticed that visual perception of this defect is influenced by the sharpness of a pucker, i.e. when amplitude of pucker is even, the surface with narrower puckers will look more creasy. In the case of the same length, higher pucker is visually evaluated as being more severe than lower pucker.
Fabric symbol M1 M2 M3
Warp ewa
Cover factors Weft ewe
Fabric ef
0.623 0.534 0.493
0.537 0.380 0.467
0.826 0.711 0.729
Seam pucker indicators
233
Coefficient of friction m Fabric/fabric Fabric/metal 0.25 0.48 0.49
0.15 0.20 0.18
Table II. Fabrics cover factors and coefficients of friction
IJCST 20,4
On the basis of the above, fabric puckering near the seam were evaluated by ratio of wave height h and length l, describing pucker sharpness SH: h SH ¼ : l
234
ð1Þ
In the process of the investigation, the influence of different sewing parameters on seam pucker sharpness was performed. Considering to different character of pucker formation in the specimen, the other criterions of pucker evaluation were sought. Results and discussion As it was mentioned already, seam pucker depends upon the properties of the fabrics being sewn together and upon the sewing modes. Results of the dependence of pucker sharpness SH upon rotational frequency w of the main shaft and pressing force P are shown in Figure 2. Analysis of the obtained results at different sewing modes allows stating that sharpness SH grows with increase in rotational frequency w of the main shaft and decrease in pressing force P of a pressing foot. Such results could be influenced by the fact that growing rotational frequency of the main shaft leads to higher inertial force, whereas decrease in pressing force leads to lower friction in the system of the fabrics being sewn together, thus the defect under investigation manifests itself stronger. The aforementioned tendency is observed in all cases; however, in different fabrics it displays itself differently. The most prominent pucker is typical of the specimens from fabric M1 (Figure 2(a) and (b)), which features the highest slipperiness among all the fabrics investigated. At the fabric/fabric friction surface, a friction coefficient is almost two times lower compared to other investigated fabrics, therefore this fact could result in low traction between the sewing garment layers. The lowest pucker sharpness is observed in the direction of warp in the specimens from fabric M3, which features the highest thickness and lowest slipperiness among all the fabrics investigated (Figure 2(e)): increasing force of a pressing foot leads to the fact that sharpness falls within the limits of lower values compared to other fabrics even at higher speed. The nature of creasing in the specimens from fabric M2 differs significantly in the directions of warp and weft. Another feature characteristic of the sharpness value distribution is related to the fact that a range of this performance values is the widest in the case of rotational frequency of the main shaft. In the direction of warp, sharpness in fabric M2 is approximate to one in fabric M1, whereas in the
hA= hB, lA > lB
hA > hB, lA = lB
hA > hB, lA > lB A
A A
B
hA lA
Figure 1.
hA
hB lB
B
B
lA
hB lB
lA
lB hA
hB
(b) (c) (a) Notes: The influence of wave height h and wavelength l on pucker surface visual perception: (a) the height h of waves A and B are the same; (b) the length l of waves A and B are the same; (c) the height h and length l of waves A and B are the different
Seam pucker indicators
235 SH
SH 2,300
0.24 0.21 0.18 0.15 0.12 0.09 0.06
1,600 900
j, min–1
200 25
45
65
2,300
0.24 0.21 0.18 0.15 0.12 0.09 0.06
1,600 900 200 25
85
j, min–1
P, N
45
65
85
P, N
(a)
(b)
SH
SH 2,300
0.24 0.21 0.18 0.15 0.12 0.09 0.06
1,600 900
j, min–1
200 25
45
65
2,300
0.24 0.21 0.18 0.15 0.12 0.09 0.06
1,600 900 200 25
85
j, min–1
P, N
45
65
85
P, N
(c)
(d)
SH
SH 2,300
0.24 0.21 0.18 0.15 0.12 0.09 0.06
1,600 900 200 25
45
65
85
P, N
j, min–1
2,300
0.24 0.21 0.18 0.15 0.12 0.09 0.06
1,600 900 200 25
45
65
P, N
(e)
(f)
85
j, min–1
Figure 2. Dependence of pucker sharpness SH on main shaft rotational frequency w (min2 1) and pressing force P (N) from the M1 fabric cut in the warp (a) and weft; (b) direction, the M2 fabric cut in the warp; (c) and weft; (d) direction, the M3 fabric cut in the warp; (e) and weft; (f) direction
IJCST 20,4
236
direction of weft this value is approximate to sharpness in fabric M3 (Figure 2(c) and (d)). Different puckering of the specimens cut in the directions of warp and weft is determined by anisotropy of textiles and this fact confirms the influence of fabric properties on the puckering of seams. In the directions of warp and weft, cover factors of fabric M2 are the most different among all the fabrics investigated. It may be seen that seam sharpness SH is influenced both by the properties of the fabrics being sewn together and by the sewing parameters investigated, i.e. pressing force and sewing speed with the higher influence of the latter. Visual perception of an object is known to be dependent upon the distance between the adjacent points. Thus, when in the section of given length a higher number of creases develops, a seam will look more wavy (Safranova, 1983). Therefore, in order to assess puckering in a more objective way, it is not enough to evaluate only sharpness of creases; nature of their distribution and their quantity in specimens shall be considered as well. Comparing for this purpose, the characteristics connected to the ratio of creases and quantity of creases within the section of given length, relative pucker coefficient k was established that allows complex evaluation of the nature of crease distribution in a specimen and is calculated as follows (equation (2)): SH £ n £ l p £ 100; ð2Þ k¼ lw where n – average number of puckers, lp – average length of a creasy section in a specimen, mm, lw – length of the working section in a specimen, mm. An attempt to relate relative pucker coefficient k with rotational frequency w of the main shaft (Figure 3) on the base of correlation analysis demonstrates that in most cases this coefficient is dependent upon rotational frequency of the main shaft. In all cases, even at a moderate correlation coefficient, increasing rotational frequency w of the main shaft also leads to higher coefficient k. Hence, it is possible to maintain that a relative pucker coefficient depends upon rotational frequency of the main shaft. After considering individual cases, when correlation dependence between the aforementioned parameters was weak, it was established that such conditions were influenced by decrease in the quantity of puckers: observing in these cases the maximum rotational frequency of the main shaft, the quantity of puckers declined, although the general tendency illustrates that increasing sewing speed results in a greater number of puckers. The provided data demonstrate that higher (from 30 to 80 per cent) relative pucker coefficients k were obtained in fabrics M1 and M2 in the direction of warp. These fabrics feature a characteristic tendency to have a higher coefficient in the direction of warp than in the direction of weft, i.e. when sewing under the same conditions, sharpness and/or quantity of pucker in the direction of warp is higher than in the direction of weft. In fabric M3, another tendency is observed: coefficient k in the direction of warp is very negligible (some 8-25 per cent), in the direction of weft it is higher (some 10-40 per cent), whereas in particular cases, at higher rotational frequency w of the main and at the lowest pressing force P, coefficient k is up to 80 per cent. An attempt to correlate pucker coefficient k with pressing force P illustrated no significant linear dependence between these characteristics. Only in the case of fabric M3, when greater rotational frequency of the main shaft is observed, increase in pressing force P leads to lower pucker coefficient k (correlation coefficient is from 0.7 to 0.9); in fabrics M1 and M2, a correlation coefficient is lower. The carried out
78 68
68 R = 0.51
58 k
R = 0.04
48
k R = 0.78
38
R = 0.67
28 18
58
R = 0.95
48
R = 0.62
38
R = 0.68
28
R = 0.99
237
18 0
k
Seam pucker indicators
78
500
1,000 1,500 2,000 2,500
0
j, min–1
(a)
(b)
90 80 70 60 50 40 30 20 10
R = 0.97 R = 0.33 R = 0.63 k R = 0.82
0
90 80 70 60 50 40 30 20 10
500 1,000 1,500 2,000 2,500
R = 0.95 R = 0.87 R = 0.63 R = 0.95 0
500 1,000 1,500 2,000 2,500
j, min–1
j, min–1
(c)
(d)
80
80
65
65
50 R = 0.76
35
R = 0.24
20
R = 0.82 R = 0.52
5 500 1,000 1,500 2,000 2,500 j, min–1
(e)
Figure 3. Dependence of relative R = 0.80 pucker coefficient on main shaft rotational frequency w (min2 1) from the M1 R = 0.97 fabric cut in the warp R = 0.19 (a) and weft; (b) direction, the M2 fabric cut in the warp; (c) and weft; (d) direction, the M3 fabric 500 1,000 1,500 2,000 2,500 cut in the warp; (e) and j, min–1 weft; (f) direction R = 1.00
50
k
0
500 1,000 1,500 2,000 2,500
j, min–1
k 35 20 5 0
(f)
investigations demonstrated that seam sharpness SH in all the cases had a tendency to decrease with increase in pressing force P. Taking into account pressing force P, variation of sharpness values SH in fabric M3 was observed within considerably wider limits compared to fabrics M1 and M2, therefore dependence of relative pucker coefficient k in the case of fabric M3 is stronger. However, coefficient k depends not only upon pucker sharpness SH, but also upon the quantity of puckers and length of a creasy section in a specimen. Results of investigations of fabrics M1 and M2 show that a greater influence on relative pucker coefficient k is made by the quantity of puckers, whereas a change tendency in the number of creases with increase in pressing force P is not significant; it was just observed that in a general case the quantity of the puckers formed is higher. Without distinct influence on the change in the quantity of puckers,
IJCST 20,4
no strong linear dependence between pressing force P and relative pucker coefficient k was obtained. Hence, the provided results demonstrate that in order to assess seam puckering and influence of different factors made on it, application of data on pucker height and length is not sufficient. It is also necessary to consider the quantity of puckers as well as their layout and nature of distribution along the seam.
238 Conclusions The results of the investigation showed that pucker sharpness increases with growing rotational frequency of the main shaft and reduction of pressing force. It was established that sewing speed has higher influence on the changes of the seam pucker sharpness then pressing force. The results of quantitative evaluation of seam pucker can be more objective considering the nature of their distribution and their quantity in specimens, not only sharpness. Analysing the influence of sewing machine parameters on the relative pucker coefficient it was obtained that this coefficient increases with increasing of rotational frequency of the main shaft. Taking into consideration the results of the correlation analysis, it may be concluded that linear relationship between pucker coefficient and pressing force was not observed. Also, the investigation showed that the results of seam puckers in specimens cut in the directions of warp and weft are significantly different regarding with anisotropy of textiles. This result confirms the influence of fabric properties on the puckering of seams. References Dobilaite˙, V. and Jucie˙ne˙, M. (2006), “The influence of mechanical properties of sewing threads on seam pucker”, International Journal of Clothing Science & Technology, Vol. 18 No. 5, pp. 335-45. Dobilaite˙, V. and Petrauskas, A. (2002), “The effect of fabric structure and mechanical properties on seam pucker”, Material Science (Medzˇiagotyra), Vol. 8 No. 4, pp. 495-9. Fan, J. and Leeuwner, W. (1998), “The performance of sewing threads with respect to seam appearance. Part 1”, Journal of Textile Institute, Vol. 89 No. 1, pp. 142-54. Fan, J. and Liu, F. (2000), “Objective evaluation of garment seams using 3D laser scanning technology”, Textile Research Journal, Vol. 70 No. 11, pp. 1025-30. Fan, J. et al., (1999a), “The use of a-2D digital filter in the objective evaluation of seam pucker on 3D surface. Part 1”, Journal of Textile Institute, Vol. 90 No. 3, pp. 445-55. Fan, J. et al., (1999b), “Towards the objective evaluation of garment appearance”, International Journal of Clothing Science & Technology, Vol. 11 Nos 2/3, pp. 151-9. ISO 7770 (1985), “Textiles. Method for assessing the appearance of seams in durable press products after domestic washing and drying”, available at: www.iso.ch Kawabata, S. and Niwa, M. (1998), “Clothing engineering based on objective measurement technology”, International Journal of Clothing Science & Technology, Vol. 10 Nos 3/4, pp. 263-72. Park, C.K. and Kang, T.J. (1999a), “Objective evaluation of seam pucker using artificial intelligence. Part I: geometric modeling of seam pucker”, Textile Research Journal, Vol. 69 No. 10, pp. 735-42. Park, C.K. and Kang, T.J. (1999b), “Objective evaluation of seam pucker using artificial intelligence. Part III: using the objective evaluation method to analyze the effects of sewing parameters on seam pucker”, Textile Research Journal, Vol. 69 No. 12, pp. 919-24.
Park, C.K. et al., (1997), “A new evaluation of seam pucker and its applications”, International Journal of Clothing Science & Technology, Vol. 9 Nos 1/3, pp. 252-5. Safranova, I.V. (1983), The Measurement Methods and Devices Applied in Sewing Industry, Light and Food Industry, Moscow, p. 232 (in Russian). Stylios, G.K. (1998), “Method and apparatus for assessment of seam pucker and other surface irregularities”, British Patent No 9819614.0. Stylios, G. and Lloyd, D.W. (1989a), “The mechanism of seam pucker in structurally jammed woven fabrics”, International Journal of Clothing Science & Technology, Vol. 1 No. 1, pp. 5-11. Stylios, G. and Lloyd, D.W. (1989b), “A technique for identification of seam pucker due to structural jamming in woven textiles”, International Journal of Clothing Science & Technology, Vol. 1 No. 2, pp. 25-7. Stylios, G. and Lloyd, D.W. (1990), “Prediction of seam pucker in garments by measuring fabric and thread mechanical properties and geometrical relationships”, International Journal of Clothing Science & Technology, Vol. 2 No. 1, pp. 6-15. Stylios, G. and Sotomi, J.O. (1993a), “A new instrument for routine objective assessment of seam deformations in limp materials”, in Blackshaw, D.M.S., Hope, A.D. and Smith, G.T. (Eds), Laser Metrology and Machine Performance, Lamdamap 93, Computational Mechanics Publications, Southampton, pp. 233-8. Stylios, G. and Sotomi, J.O. (1993b), “Investigation of seam pucker in lightweight synthetic fabrics as an aesthetic property. Part I: a cognitive model for the measurement of seam pucker”, Journal of Textile Institute, No. 4, pp. 593-600. Stylios, G. and Sotomi, J.O. (1993c), “Investigation of seam pucker in lightweight synthetic fabrics as an aesthetic property. Part II: model implementation using computer vision”, Journal of Textile Institute, No. 4, pp. 601-10. Corresponding author Milda Jucie˙ne˙ can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
Seam pucker indicators
239
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,4
240 Received 2 August 2007 Revised 12 January 2008 Accepted 12 January 2008
Heat and moisture transmission properties of clothing systems evaluated by using a sweating thermal manikin under different environmental conditions Damjana Celcar Mura European Fashion Design, Murska Sobota, Slovenia
Harriet Meinander SmartWearLab, Tampere University of Technology, Tampere, Finland, and
Jelka Gersˇak The Faculty of Mechanical Engineering, Institute of Engineering Materials and Design, University of Maribor, Maribor, Slovenia Abstract Purpose – The paper aims to investigate thermal comfort properties, such as heat and moisture transmission through male business clothing systems, by using a sweating thermal manikin Coppelius that simulates heat and moisture production in a similar way to the human body and measures the influence of clothing on heat exchange in different environmental and sweating conditions. Design/methodology/approach – Ten different combination of male business clothing systems were measured using the sweating manikin, under three environmental conditions (108C/50 per cent RH, 258C/50 per cent RH and 2 58C), and at 0 and 50 gm2 2 h2 1 sweating levels, in order to evaluate the influence of environmental and sweating conditions on thermal comfort properties of clothing systems. Findings – The results show how business clothing systems influence on the dry and evaporative heat loss between the manikin surface and environment in different environmental and sweating conditions. Practical implications – When using sweating thermal manikin Coppelius, water vapour transmission (WVT) through and water condensation on the clothing can be determined simultaneously with the thermal insulation (It) of clothing system. Measured thermal comfort properties of clothing systems evaluated with a sweating thermal manikin can provide valuable information for the clothing industry by manufacturing/designing new clothing systems. Originality/value – In this investigation, the heat and moisture transmission properties of male business clothing systems were measured in different environmental and sweating conditions. In the past few years, clothing materials containing microencapsulated phase-change materials (PCMs) have appeared in outdoor garments, particularly sportswear; therefore, we decided to investigate the thermal comfort properties of different standard male business apparel, as well as male business clothing that contain PCMs used as liner and outerwear material. Keywords Thermal properties of materials, Moisture, Clothing, Heat transfer International Journal of Clothing Science and Technology Vol. 20 No. 4, 2008 pp. 240-252 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810878865
Paper type Research paper
This research was performed in the SmartWearLab of the Tampere University of Technology. The authors would like to thank the university for offering their facilities in order to perform this research instead of the Leonardo da Vinci II programme.
1. Introduction Clothing wear comfort is a state of mind influenced by a range of factors and is the result of a balanced process of heat exchange between the human body, the clothing system and the environment. It depends on many factors such as the temperature of the environment, the relative humidity, the wind velocity, the metabolism of the wearer and, of course, the characteristics of the clothing materials, e.g. materials’ thermal comfort properties, which display their abilities to transport heat and moisture from the human body’s surface into the environment. The measuring values that reflect this ability are clothing’s thermal resistance or thermal insulation, and water vapour resistance. Many other factors such as colour, fashion, a person’s physical and psychological state also influence the feeling of comfort (Mecheels, 1998). Wear comfort of business clothing is one of more important factors when selecting clothing, and a decisive factor in the evaluation of the clothing’s quality. In view of the fact that people wear business clothing throughout the whole day, and because today’s requirements regarding clothing comfort are much higher than in the past, we decided to investigate the thermal comfort properties of different male business apparel, as well as male business clothing that contain phase-change materials (PCMs) used as liner and outerwear material. PCMs, also called latent heat storage materials, are materials that are able to interact with the human body and produce thermoregulatory control by affecting the microclimate between the clothing and the human skin. These materials can store, release or absorb thermal energy as latent heat whilst they oscillate between solid and liquid form, giving off heat as they change to a solid state and absorbing it as they return to a liquid state (Zhang, 2001). The thermal comfort properties of clothing systems can be defined through physical measurements using thermal and sweating manikins or through wear trials using human test subjects. In this investigation a sweating thermal manikin Coppelius (Meinander, 1994) was used for measuring the male business clothing’s thermal comfort properties. By using a sweating thermal manikin, which simultaneously measures thermal insulation and water vapour transfer in clothing systems, it is possible to test different clothing systems in a climate chamber under different climate and sweating conditions. The amount of condensation and its distribution in clothing was also analysed and discussed in this research. In the past few years, the research work of the European Subzero project has been reported where relationship between physically measured thermal insulation values of cold protective clothing and the corresponding physiological reactions on human test subjects were defined (Meinander et al., 2004). The project results showed that the thermal insulation values defined with thermal manikins (using different types of thermal manikins) correspond well with the wear trial values. It was also concluded that for a direct comparison between manikin test and wear trials with human subjects, the test conditions should match as well as possible. 2. Measuring the effect of clothing on heat exchange Clothing’s thermal comfort properties can be evaluated using several physical and physiological testing methods. Physical tests use some kind of device to simulate the skin’s heat and/or water-vapour production. Tests can be carried out either on textile materials or on completed clothing systems. Most of the physical test methods are concerned with only one property, either resistance to dry heat loss or to water vapour transmission (WVT). Some test methods attempt to simultaneously measure
Heat and moisture transmission 241
IJCST 20,4
242
dry and evaporative heat loss through textile materials or completed clothing systems. In the physiological tests the human subjects are dressed in the test clothing systems and perform some kind of given metabolic workload in a controlled environment. The temperature and humidity at different points, the oxygen consumption, the heart rate, the weight loss, and other physical and physiological values of interest regarding the subject and in the environment can all be determined. In physiological testing a long series of tests has to be performed to obtain a reliable picture of the overall clothing’s properties, which makes these tests time consuming, and thus expensive (Mecheels, 1998, Meinander, 1985). The commonly used procedure for estimating the thermal comfort properties of clothing systems is the use of thermal and sweating manikins. Many thermal manikins (reviews of manikin history and applications, Holme´r, 1999, 2004) have been developed around the world since the first one-segmented copper manikin from the US Army in the 1940s. A complete understanding of human heat exchange requires not only the convective, conductive and radiative heat losses to be measured but also the important mechanism for heat loss, namely sweat evaporation. Several manikins in operation can simulate human sweating and provide valuable information about heat exchange by evaporation (review of manikins, Holme´r, 2004). So-called sweating thermal manikins simulate heat and moisture production in a similar way to the human body and measure the influence of clothing in different environmental and sweating conditions. However, they are relatively rare, and their design and test method vary considerable from lab to lab. The Swiss sweating agile thermal manikin (SAM) was designed to study the static dry thermal insulation of clothing as well as the dynamic effects of workload, sweating and realistic body movements (Richards and Mattle, 2001). SAM’s anatomically formed body is divided into 30 separately heated sectors (26 shell parts and four heated joints). Shell parts are moulded from a plastic mixed with aluminium powder. Over its surface 125 sweat outlets are distributed thus the sweat rate can be varied from 20 ml/h up to at least 4 l/h to simulate all possible activities and conditions. During an active test phase the repetitive body movements such as walking and climbing can be performed. Fan and his co-workers developed another fabric sweating manikin Walter (Chen et al., 2003; Fan and Qian, 2004). Walter simulates perspiration using a waterproof, but moisture-permeable, fabric “skin”, which holds the water inside the body, but allows moisture to pass through the skin. The manikin is heated by heaters within the trunk, and its core temperature is controlled at 378C. Water within the manikin is circulated by a pump and a piping system, which distributes heat to the heat, arms, and legs by pumping the warm water to the extremities. The improved sweating manikin Walter can simulate walking since his arms and legs can be motorised. Another manikin is the American sweating manikin ADvanced Automotiv Manikin (ADAM), which is completely self-contained (power and water) and designed to respond to transient, non-uniform thermal environments like a human sweating and breathing. ADAM has 120 separate heated and sweating segments that provide uniform heating and sweating across the skin surface. Energy and water storage are contained in the manikin, which has wireless communication and control (Burke and McGuffin, 2001). In this study, we simultaneously evaluated the total (dry and evaporative) heat loss from a clothed body, and water vapour transfer through the clothing, with the thermal
insulation of clothing systems, using a sweating thermal manikin Coppelius developed by Meinander (1994, 1999) (Figure 1). 3. Experimental part 3.1 Test method – sweating thermal manikin Coppelius A sweating thermal manikin Coppelius was used in this study, which was constructed for objective measuring the simultaneous transmission of heat and water vapour through clothing systems. This manikin was developed in the 1980s within a Nordic project and its construction is based on the Swedish dry manikin Tore, to which an additional sweating mechanism has been added. Using the sweating manikin, which is constructed to be similar in shape to the human body, it is possible to evaluate the
Heat and moisture transmission 243
Figure 1. Sweating thermal manikin Coppelius dressed in business clothing
IJCST 20,4
244
thermal properties of clothing systems. The construction of the sweating manikin consists of 18 individually controlled body sections, which are electrically heated. Continuous sweating from the body’s surface (except head, hands and feet) is possible through 187 individually controlled sweat glands. These sweat glands on the surface supply a controlled amount of water to the manikin surface. Water in liquid form is supplied through a thin tube to the laminated skin material, which consist of two layers. The inner layer is highly wicking and spreads the water over a large surface. The outer layer is a micro porous membrane, which transmits water in vapour but not in liquid form. The heating underneath the skin evaporates the water and keep the surface temperature at a constant, set level. Heat and water vapour is thus produced from the manikin surface. The shape of the manikin with anatomic body dimensions (size 52) and prosthetic joints in shoulder, elbows, hips and knees resemble the actual shape of the human body. In the climate chamber the manikin is suspended from a balance, which records the weight changes during the test, thus the moisture evaporation can be recorded as the difference between the supplied water and the weight increase of the manikin with clothing. Water is supplied from a reservoir, which is placed on another balance near the ceiling in the control room. Each item of the clothing is also weighed before and immediately after the test in order to determine in which clothing layers the moisture is condensing. A computer program is written for the control and measurement of the following parameters: water supply, condensation and evaporation (weight changes), the surface temperatures of all the manikin’s body sections, temperatures at selected points, heat supply to all body sections, thermal insulation of all body sections, and the climatic conditions (temperature and relative humidity). Figure 2 shows the test configuration with the manikin in the climate chamber (Meinander, 1994, 1999). From the recorded measurement values the following values could be calculated (Meinander, 1994):
Water supply Manikin balance
Computer control
Figure 2. Test configuration with the manikin in the climate chamber
Source: Meinander (1999)
.
.
The thermal insulation (dry, It) from the manikin’s surface to the environment is determined from the dry heat supplied to the manikin and the temperature values, by: ts 2 ta · A ð8C m2 W21 Þ ð1Þ It ¼ H dry
Heat and moisture transmission
where Hdry is the mean dry heat supplied to the manikin, ts is the manikin’s mean surface temperature, which correspond to the skin’s temperature (348C), ta is the mean ambient temperature (8C), and A is the surface area of the manikin (1,376 m2). Taking into account that heat supply (total heat supply in the sweat test Hsw) is partly used to evaporate the water, the corrected thermal insulation (sweating, It,corr) value is calculated by:
245
I t;corr ¼
ts 2 ta · A ð8C m2 W21 Þ H sw 2 H e
ð2Þ
where He is the evaporative heat loss, which is calculated by: H e ¼ ðms 2 mc Þ · w ¼ me · w
.
ðW m22 Þ
ð3Þ
where ms is the water fed into the manikin (g), mc is the condensed water in the clothing (g), me is the evaporated water (g), and w is the specific heat of evaporation for water (0.674 Wg2 1 at 258C). The amount of evaporated water Me as a percentage of the water input, giving a value for the WVT of the tested clothing systems. Me is calculated by: Me ¼
me · 100 per cent ms
ð4Þ
3.2 Clothing ensembles, conditions and procedure The first part of the experimental work covered the selection and testing of different textile materials and material combinations used for male business clothing (Celcar et al., 2008). The determination of material properties in steady-state conditions was carried out according to standardized test methods, as follows: thermal resistance (Rct): ISO 5085-1 (1989), WVT: Gore cup method modified by Gore-Tex (Gohlke, 1980), air permeability (Q): ISO 9237 (1995), thickness (h): ISO 5084 (1996) and mass per unit area (W): ISO 3801 (1977). Table I shows a review of the selected materials, and the results for the standard measurements of the materials’ properties. In the first part of the experimental work (Celcar et al., 2008) different combinations of materials were tested on the sweating cylinder under different environmental and sweating conditions in order to find out the best combination of textile materials, which simulate business clothing system. Different prototypes of male business clothing systems were manufactured for testing on the sweating manikin Coppelius on the basis of the thermal properties of material combinations evaluated with the sweating cylinder and standard test methods. In this part of investigation ten combinations (cs1-cs10) of male business clothing systems were chosen for testing on the sweating manikin Coppelius under different environmental and sweating conditions. Table II shows a review of the selected
IJCST 20,4
Clothing system layer
Fabric sample code Fabric content
Underwear UN2 Shirt S1
246
Liner
Male suit
L1 L2 L3 MS1 MS2 MS3
Table I. Chosen materials and their properties determined according to the standard test methods
MS4
Male coat
C1
100 per cent CO 78 per cent CO, 22 per cent PES 100 per cent CV 1.layer: 100 per cent CV, 2.layer: outlast-acryl with PCM 100 per cent CV 100 per cent WO 88 per cent WO, 12 per cent PA 98 per cent WO, 2 per cent EL 68 per cent Outlast-Acryl with PCM, 28 per cent WO, 4 per cent EL 100 per cent WS
Clothing system layer Underwear Shirt Liner for jacket
Table II. Combinations of clothing systems and intended environmental conditions
Weight Thickness Rct Q WVT (g/m2) (mm) (8Cm2/W) (l/m2 s) (g/m2 · 24 h)
Fabric code
UN2 S1 L1 L2 Male suit MS1 MS2 MS3 MS4 Liner for coat L3 Male coat C1 Environmental conditions
cs1
cs2
* * *
* * *
179.0 85.0
1.29 0.21
0.028 0.005
1,180.0 322.0
5,455.0 5,713.0
76.0
0.11
0.001
596.0
6,008.0
93.0 101.0 179.0
0.21 0.14 0.51
0.002 0.002 0.016
151.0 125.2 323.5
5,368.0 5,804.0 5,695.0
206.0
0.49
0.011
75.2
5,552.0
189.0
0.49
0.011
223.0
5,648.0
168.0 300.0
0.49 1.71
0.014 0.053
277.0 196.5
5,306.0 4,520.0
Combination of clothing system cs3 cs4 cs5 cs6 cs7 cs8 * * *
* * *
* * * *
*
* * *
* * *
* * *
cs9
cs10
* * *
* * * *
*
*
* *
* * * *
108C/50 per cent RH 258C/50 per cent RH
* *
* *
* * *
* *
108C/50 per cent RH 258C
clothing systems’ layers and materials, combinations of clothing systems, and the intended environmental conditions. In the first series of tests, we tested five clothing ensembles (cs1-cs5) of three-layer clothing system (underwear, male shirt, male suit with liner) at 108C/50 per cent RH and 258C/50 per cent RH. In the second series of tests, which simulated a cold environment, we tested five clothing ensembles (cs6-cs10) of four-layer clothing system (underwear, male shirt, male suit with liner and male coat with liner) at 108C/50 per cent RH and 2 58C. The same underwear, male shirt and male coat with liner (in a cold environment) were used for testing. In this research, we varied the textile materials of male suits and liners for the male suit. All tests were done at 0 and 50 g/m2 h sweating levels and two
parallel tests were carried out for all combinations. If the difference in results was . 5 per cent a third test was done and the outer layer ignored. The test procedure was as follows: (1) Weighing of the individual garments and manikin (dry weight). (2) Dressing the manikin with garments and the fastening of 6 PT 100 surface temperature sensors onto the each surface of the tested garment. The surface sensors measure the temperatures of the tested garment’s materials. One sensor measures air temperature in the climate chamber. (3) Switching on heating. (4) Dry testing. Duration of the dry test was 1.5 h. (5) Sweating test. Duration of wet test was 3 h. (6) Switching off heating and sweating. (7) Undressing the manikin and immediate weighing of the individual garments (wet weight).
Heat and moisture transmission 247
4. Results and discussion Heat loss from the skin consists of two parts: the dry heat loss Hdry (convection, conduction and radiation) and the evaporative heat loss He. The results for dry heat losses (proportional to the temperature difference between skin and ambient) through the clothing systems under different environmental conditions, are shown in Figure 3. Evaporative heat losses that calculated from the amount of WVT under different environmental conditions are shown in Figure 4. Dry heat loss
H [W/m2]
H [W/m2]
Dry heat loss 120.0 100.0 80.0 60.0 40.0 20.0 0.0 cs1
cs2
cs3 cs4 Clothing system
Hdry/10°C/50%
120.0 100.0 80.0 60.0 40.0 20.0 0.0
cs5
cs6
cs5
Clothing system He/10°C/50%
He [W/m2]
Hdry/–5°C
Figure 3. Dry heat loss through clothing systems under different ambient temperatures
Evaporative heat loss
He [W/m2] cs4
cs10
(b) 10°C/50 per cent and –5°C
Evaporative heat loss
cs3
cs9
Clothing system
35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 cs2
cs8
Hdry/10°C/50%
Hdry/25°C/50%
(a) 10°C/50 per cent and 25°C/50 per cent
cs1
cs7
He/25°C/50%
(a) 10°C/50 per cent and 25°C/50 per cent
35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 cs6
cs7
cs8
cs9
Clothing system He/10°C/50%
He/–5°C
(b) 10°C/50 per cent and –5°C
cs10
Figure 4. Evaporative heat loss through clothing systems under different ambient temperatures
IJCST 20,4
248
It can be seen from Figure 3 that the heat loss (Hdry) from the manikin through male business clothing systems at dry test is 60 per cent higher at 108C than at 258C. This means that the manikin at an ambient temperature of 108C needs much more heat to maintain an average skin temperature (348C), compared to the test at an ambient temperature of 258C. In cold conditions, where the temperature difference is even higher, we noted that dry heat losses at an ambient temperature 2 58C are 48 per cent higher than at an ambient temperature 108C. By comparing dry heat losses at 108C in all combinations of male business clothing systems it can be seen, that the manikin needs much more heat to maintain the average skin temperature with clothing systems cs1-cs5, e.g. combinations without a male coat. This means that those combinations have lower thermal resistance values when compared to combinations where we tested four-layer male business systems with a coat (cs6-cs10). By comparing evaporative heat losses (He) it can be noted, that there exists very small differences between the values at different ambient temperatures. As can be seen from Figure 4, the evaporative heat loss values at 108C/50 per cent RH in all combinations of male business clothing systems are about 4.5 per cent higher than at 258C and about 10 per cent higher than at 2 58C. Figure 5 shows the influence of ambient temperature on the thermal insulation (It) values of clothing systems, which are determined from the heat loss and temperature values. The thermal insulation values of male business clothing systems are at 108C about 10 per cent higher than at 258C and at 2 58C about 20 per cent higher than at 108C. By comparing all clothing systems at 108C it can be seen that thermal insulation values for clothing systems with a male coat (cs6-cs10) are higher, therefore they provide better protection against dry heat lost. The results of dry heat loss and thermal insulation show that small differences in heat loss and thermal insulation values exist between clothing systems with and without PCM particles in/on fabric, e.g. combination cs1, male suit with CV liner has a little lower thermal insulation than same combination cs5, male suit with CV liner and PCM particles because of higher dry heat loss values. An explanation for the difference between these thermal insulation values is probably the difference in thickness and weight of liner materials with/without PCM particles. In previous research, where we tested combinations of materials on the sweating cylinder, this tendency was even more significant. It can be seen from the results that combination cs4 with the lowest percentage of wool (28 per cent) and 68 per cent of Thermal insulation of clothing system
0.600
It [m2°C/W]
It [m2°C/W]
0.500
0.400 0.300 0.200 0.100
Figure 5. Thermal insulation values of clothing systems under different ambient temperatures
Thermal insulation of clothing system
0.600
0.500
0.400 0.300 0.200 0.100
0.000 cs1
cs2
cs3
cs4
cs5
Clothing system It/10°C/50%
It/25°C/50%
(a) 10°C/50 per cent and 25°C/50 per cent
0.000 cs6
cs7
cs8
cs9
cs10
Clothing system It/10°C/50%
It/–5°C
(b) 10°C/50 per cent and –5°C
acryl with PCM microcapsules in the fabric of a male suit has a little lower thermal insulation than other clothing systems. Using the sweating manikin, WVT (Me) of the clothing systems was determined, which is given by the amount of evaporated water (me) as a percentage of the water input (ms). Figure 6 shows the values for WVT of male business clothing systems under different ambient conditions. During the sweat test the feeded water (water input) partly evaporates and partly condenses. The amounts of moisture that have condensed (mc) in the clothing systems and evaporated (me) through the clothing during the sweat test at sweating level 50 g/m2 h, are shown in Figure 7. It can be seen from Figure 6 that the values for WVT are higher at 108C than at 258C and 2 58C. We also noted that the rates of evaporating water at sweating level 50 g/m2 h are about 80 per cent (cs1-cs5) and 65 per cent (cs6-cs10) at 108C, 75 per cent at 258C and 53 per cent at 2 58C. This means that at 108C evaporated water was much
Me [%]
Me [%] cs2
cs3 cs4 Clothing system
Me/10°C/50%
90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 cs6
cs5
Me/25°C/50%
cs10
Me/–5°C
Evaporated and condensed water at 10°C/50 per cent, 50g/m2h 50.0
Evaporated and condensed water at 25°C/50 per cent, 50g/m2h 50.0
40.0
40.0 m [g/m h]
30.0
2
2
m [g/m h]
cs8 cs9 Clothing system
(b) 10°C/50 per cent and –5°C
20.0 10.0
Figure 6. WVT values for clothing systems under different ambient conditions
30.0 20.0 10.0
0.0
0.0 cs1
cs2
cs3 cs4 Clothing system me
cs5
cs1
cs2
mc
cs3 cs4 Clothing system me
(a) 10°C/50 per cent
cs5
mc
(b) 25°C/50 per cent
Evaporated and condensed water at 10°C/50 per cent, 50 g/m2h 50.0
50.0
40.0
40.0
Evaporated and condensed water at –5°C, 50 g/m2h
m [g/m h]
30.0
2
2
cs7
Me/10°C/50%
(a) 10°C/50 per cent and 25°C/50 per cent
m [g/m h]
249
Water vapour transmission in per cent
Water vapour transmission in per cent 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 cs1
Heat and moisture transmission
20.0 10.0
30.0 20.0 10.0 0.0
0.0 cs6
cs7
cs8
cs9
Clothing system me
mc
(c) 10°C/50 per cent
cs10
cs6
cs7
cs8
cs9
Clothing system me
mc
(d) –5°C
cs10
Figure 7. Evaporated and condensed water through/in clothing systems under different ambient temperatures
IJCST 20,4
250
higher compared to 25 and 2 58C, in spite of the fact that WVT values decrease with a decreasing ambient temperature. An explanation for the difference between 10 and 258C is probably the difference in partial water pressure. With the higher amount of water vapour in the air at 258C, the gradient between the manikin’s surface and the environment is lower than at 108C, and, therefore the WVT is also lower, even if the temperature is higher. By comparing the evaporated water at different ambient temperatures, we noted that the values for evaporated water at 108C are about 7 per cent higher than at 258C and about 23 per cent than at 2 58C. The condensation of moisture is 18 per cent higher at 258C than at 108C and 25 per cent higher at 2 58C than at 108C. If we compare the results for condensed water regarding all the clothing systems tested at 108C, it can be seen that 44 per cent more condensed water appeared on clothing systems cs6-cs10 (with male coat) than at combinations without a coat (cs1-cs5). Weighing clothes before and immediately after the test determined the amount of condensed water in the individual clothing. As an example, the results for male business clothing systems tested at 258C/50 per cent RH and 2 58C with 50 g/m2 h sweating level, are shown in Figure 8. The accumulation of water in the clothing, shown in Figure 8, was noted at ambient temperatures 258C/50 per cent RH and 2 58C. The largest amount of water at ambient temperature 258C/50 per cent RH was noted to be in the underwear and also in the jackets of the male suits. By comparing clothing system cs1 (combined with WO jacket and CV liner) with cs5 (combined with WO jacket and CV liner with PCM particles), it can be noted that, for clothing system cs5, more water was condensed in the jacket than for clothing system cs1. We used the same cotton underwear for testing, which is hygroscopic and has excellent properties for everyday clothing worn in normal-wear situations with only a limited amount of sweating. It can be seen from Figure 8(b) that at ambient temperature 2 58C much more condensed water was in the male coats than in the underwear or male suits, where all the fed water, and the water accumulated in the clothing because of water content in the air, evaporated throughout the clothing and condensed in the male coat.
Condensed water in clothing at 25°C/50 per cent, 50 g/m2h 10.00 8.00 6.00 4.00 2.00 0.00 –2.00 cs1 cs2 cs3 cs4 cs5
Condensed water in clothing at –5°C, 50g/m2h 35.00 25.00 mc [g]
Figure 8. Water condensation in individual clothing of the clothing system, at 258C/50 per cent RH and 258C, and 50 g/m2 h sweating level
mc [g]
5. Conclusions A sweating thermal manikin Coppelius was used for measuring heat and moisture transmission properties through male business clothing systems. Different combinations of clothing systems were tested under three ambient conditions and two sweating levels, in order to evaluate its thermal insulation and moisture evaporation values.
15.00 5.00 –5.00 –15.00 cs6
undshirt
undpants
shirt
cs7
cs8
cs9
cs10
Clothing system
Clothing system jacket
(a) 25°C/50 per cent
trousers
undshirt
undpants
shirt
jacket
(b) –5°C
trousers
coat
By comparing the results of dry and evaporative heat loss values, it can be noted that dry heat loss values increase with a decreasing ambient temperature and that there exist very small differences in evaporative heat loss values at different ambient temperatures, the highest being at 108C. The results of thermal insulation show that, with a decreasing ambient temperature, the thermal insulation of male business clothing systems increases. We also noted, that when dry testing small differences in heat loss and thermal insulation values exist between clothing systems with and without PCM particles in/on the fabric, e.g. combination cs1, combined with WO male suit (100 per cent WO jacket with CV liner) has lower thermal insulation than the same combination cs5, combined with WO male suit (100 per cent WO jacket with CV liner with PCM particles) because of higher dry-heat loss values. An explanation for the difference between thermal insulation values is probably the differences in thickness of the used liner materials with/without PCM particles. This tendency was also noted in previous research where combinations of materials on the sweating cylinder were tested (Celcar et al., 2008), therefore, we cannot confirm that these small differences exist because of the content of PCM particles in/on materials. The condensation of moisture at the lower ambient temperature (2 58C) is higher than at 108C and 258C, while the amount of evaporated water as a percentage of water input is higher at 108C compared to 25 and 2 58C. Many earlier studies showed that the WVT values decrease with decreasing temperature, due to the higher condensation in the clothing. An explanation for the difference between WVT values at 10 and 258C is probably the difference in partial water pressure. With the higher amount of water vapour in the air at 258C, the gradient between the manikin’s surface and the environment is lower than at 108C and, therefore, the WVT is also lower, even if the temperature is higher. From those thermal comfort properties of male business clothing systems evaluated using a sweating manikin, it can be seen that very small differences exist between clothing systems made of materials with PCMs and standard wool materials and wool mixtures, but we cannot confirm that these small differences exist because of the content of PCM particles in/on the materials. Further research is, therefore, needed in order to confirm this tendency, because the standard testing methods used for determining the insulation value of traditional materials do not measure the effects of PCM materials in a dynamic environment. In further research, we will evaluate male business clothing prototypes on human volunteers under artificially created conditions in a climate chamber, where the impact of the clothing on the physiological parameters of the tested person will be investigated. A questionnaire and an assessment scale will also be completed in order to determine the impact of the garments’ thermal comfort properties on the wearer’s subjective feeling of comfort. References Burke, R. and McGuffin, R. (2001), “Development of an advanced thermal manikin for vehicle climate evaluation”, Proceedings of the 4th International Meeting on Thermal Manikins, EMPA Switzerland, 27-28 September, pp. 14-18. Celcar, D., Meinander, H. and Gersˇak, J. (2008), “A study of the influence of different clothing materials on heat and moisture transmission through clothing materials, evaluated using a sweating cylinder”, International Journal of Clothing Science & Technology (in press).
Heat and moisture transmission 251
IJCST 20,4
252
Chen, Y.S., Fan, J. and Zhang, W. (2003), “Clothing thermal insulation during sweating”, Textile Research Journal, Vol. 73 No. 2, pp. 152-7. Fan, J. and Qian, X. (2004), “New functions and applications of Walter, the sweating fabric manikin”, European Journal of Applied Physiology, Vol. 92, pp. 641-4. Gohlke, D.J. (1980), Improved Analysis of Comfort Performance in Coated Fabrics, W.L. Gore &Associates, Inc., Newark, DE. Holme´r, I. (1999), “Thermal manikins in research and standards”, in Nilsson, H.O. and Holmer, I. (Eds), Proceedings of the 3rd International Meeting on Thermal Manikin Testing 31MM at the National Institute for Working Life, 12-13 October, pp. 1-7. Holme´r, I. (2004), “Thermal manikin history and applications”, European Journal of Applied Physiology, Vol. 92, pp. 614-8. ISO 3801 (1977), Textiles-determination of Mass Per Unit Length and Mass Per Unit Area, ISO, Gene`ve. ISO 5084 (1996), Textiles-determination of Thickness of Textiles and Textile Products, ISO, Gene`ve. ISO 5085-1 (1989), Textiles-determination of Thermal Resistance – Part 1: Low Thermal Resistance, ISO, Gene`ve. ISO 9237 (1995), Textiles-determination of the Permeability of Fabrics to Air, ISO, Gene`ve. Mecheels, J. (1998), Ko¨rper-Klima-Kleidung. Wie funktioniert unsere Kleidung?, Shiele & Scho¨n, Berlin. Meinander, H. (1985), Introduction of a New Test Method for Measuring Heat and Moisture Transmission Trough Clothing Materials and its Application on Winter Work Wear, VTT Publication, Espoo. Meinander, H. (1994), “Thermal properties of clothing systems studied with a sweating thermal manikin”, 3rd International Symposium on Clothing Comfort Studies, Mt. Fuji, Japanese Research Association of Textile End Users, 23-25 October, pp. 51-70. Meinander, H. (1999), “Extraction of data from sweating manikin tests”, in Nilsson, H.O. and Holmer, I. (Eds), Proceedings of the 3rd International Meeting on Thermal Manikin Testing 3IMM at the National Institute for Working Life, 12-13 October, pp. 95-9. Meinander, H. et al., (2004), “Manikin measurements versus wear trials of cold protective clothing (Subzero project)”, European Journal of Applied Physiology, Vol. 92, pp. 619-21. Richards, M.G.M. and Mattle, N.G. (2001), “Development of a sweating agile thermal manikin (SAM)”, Proceedings of the 4th International Meeting on Thermal Manikins, EMPA Switzerland, 27-28 September, pp. 23-6. Zhang, X. (2001), “Heat-storage and thermo-regulated textiles and clothing”, in Tao, X. (Ed.), Smart Fibres, Fabrics and Clothing, Woodhead Publishing Ltd, Cambridge, pp. 34-57. Corresponding author Damjana Celcar can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
IJCST 20,5
258
The single-row machine layout problem in apparel manufacturing by hierarchical order-based genetic algorithm Miao-Tzu Lin
Received 20 July 2007 Accepted 30 April 2008
Department of Fashion Design and Management, Tainan University of Technology, Yongkang City, Taiwan Abstract Purpose – Change of machine layout is often required for small quantity and diversified orders in the apparel manufacturing industry. The purpose of this paper is to use a hierarchical order-based genetic algorithm to quickly identify an optimal layout that effectively shortens the distance among cutting pieces, thereby reducing production costs. Design/methodology/approach – The chromosomes of the hierarchical order-based genetic algorithm consist of the control genes and the modular genes to acquire the parametric genes, a precedence matrix and a from-to matrix to calculate the distance among cutting pieces. Findings – The paper used a men’s shirt manufacturing as an example for testing the results of a U-shaped single-row machine layout to quickly determine an optimal layout and improve effectiveness by approximately 21.4 percent. Research limitations/implications – The manufacturing order is known. The machine layout is in a linear single-row flow path. The machine layout of the sewing department is independently planned. Originality/value – The advantage of the hierarchical order-based genetic algorithm proposed is that it is able to make random and global searches to determine the optimal solution for multiple sites simultaneously and also to increase algorithm efficiency and shorten the distance among cutting pieces effectively according to manufacturing order and limited conditions. Keywords Clothing, Machineability, Programming and algorithm theory Paper type Research paper
International Journal of Clothing Science and Technology Vol. 20 No. 5, 2008 pp. 258-270 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810898872
Introduction Good machine layout and shorten moving distance among materials are important for reducing production costs. Tompkins et al. (1996) indicated that moving non-value added material often takes up 20-50 percent of the total manufacturing costs, and an efficient layout can save 10-30 percent of the total manufacturing costs implying that an optimal layout can improve manufacturing schedules and therefore efficiency. Apparel manufacturing involves small quantities of diversified items that often require changes in machine layouts according to different materials, specifications, and manufacturing processes and methods. If the machine layout is able to be re-arranged quickly, then the change time, labor required, and moving distance can be reduced. To solve problems involving longer search times and difficulties in operations resulting from manufacturing large quantities of product series, this study adopted a hierarchical order-based genetic algorithm to modularize sub-assembly lines and
reduce the order of the search. The chromosomes of the hierarchical order-based genetic algorithm include the control genes and the modular genes that are used to acquire the manufacturing order of the parametric genes; later a from-to matrix and precedence matrix were adopted to calculate the moving distance of cutting pieces. Therefore, the proposed method has the advantages of hierarchical structure and modularization and real-time random and global searches for an optimal solution that also improves algorithm effectiveness. This paper used a men’s shirt manufacturing as an example to quickly determine the optimal machine layout, shorten moving distance among cutting pieces, and improve production efficiency. Literature review A machine layout with a linear single-row flow path exists in different configuration, such as a straight line, U-shaped, serpentine line, and loop. Although modern material handing systems often allow for complex flow path configurations, linear single-row flow paths are still popular in industry. Owing to its ease of construction and control, the linear single-row machine layout is the most commonly used layout (Ho and Moodie, 1998). The machine layout at the plants is one of the issues of this equipment layout study. Optimal algorithms are used to acquire the optimal equipment layout but more time consuming. Another application is the sub-optimal heuristic algorithm that was categorized by Kusiak and Heragu (1987) into four types: construction algorithms, improvement algorithms, hybrid algorithms, and graph theoretic algorithms. In recent years, due to the development of improved computer algorithms, many researchers have proposed meta-heuristic algorithms such as simulated annealing, Tabu search, and genetic algorithms similar to optimal algorithms to determine an acceptable optimal solution within a reasonable time. The optimal algorithms are time consuming while the sub-optimal heuristic algorithm takes solution quality into consideration within the limited time and algorithm’s ability. The linear single-row machine layout problem is identified as a NP-complete problem (Suresh and Sahu, 1993). Accurate mathematical solutions do not exist for such problems. The complexity of such problems increases exponentially with the number of devices. For instance, a system consisting of n machines will comprise a solution space with the size n. For arranging devices in the systems, the number of possible solutions is equal to the number of permutation of n elements (Ficko et al., 2004). Hollier (1963) presented four flow-line analysis (FLA) methods for four single-machine-type problems where only one machine of any type is allowed in a flow line. Bragalia (1996) proposed a combination of simulated annealing and genetic algorithms to minimize total backtracking in the linear ordering machines. It is assumed that the machine locations and facilities are equally space. A two-phase layout procedure combining FLA and simulated annealing is suggested by Ho and Moodie (1998). Phase one modified Hollier’s four FLA methods (methods 1, 2, 3, and 4) and presented two new FLA methods (methods 5 and 6). Method 5 is suitable for bidirectional flow lines and method 6 is suitable for unidirectional flow lines. Ponnambalam and Ramkumar (2001) proposed the two best methods, FLA methods 5 and 6, combining an order-based genetic algorithm to reduce the material handling cost by efficient layout design. Ficko et al. (2004) presented a model of the flexible manufacturing systems in single or multiple rows with the order-based genetic algorithm.
Single-row machine layout problem 259
IJCST 20,5
In studies of machine layouts, the selection of the objective function has been focused on minimal moving time, minimal total moving distance, minimal moving distance among parts, minimal moving costs of materials, and minimal return times (Heragu and Kusiak, 1988; Kumar et al., 1995; Ho and Moodie, 1998; Sarker et al., 1998). The application of the genetic algorithm can convert an objective function into a fitness function to satisfy the survival of the fittest principle.
260 Mathematical model From-to matrix The commonly used linear single-row machine layout in the manufacturing industry is shown in Figure 1. The distance among stations as the from-to matrix is shown in equation (1): 3 2 t 11 · · · t 1j · · · t1I 7 6 . . 6 . . . ... . . . ... 7 7 6 . 7 6 7 6 t i1 · · · tij · · · t iI 7; ð1Þ T IxI ¼ tij IxI ¼ 6 7 6 7 6 . . 6 . . . ... . . . ... 7 7 6 . 5 4 t I 1 · · · tIj · · · t II where I:I is the number of stations and tij:tij is the distance from station i to station j (unit: meter). Precedence diagram and matrix The precedence diagram with circles and arrows to indicate the relationship of the stations is shown in Figure 2. The relationship among stations as the precedence matrix is shown in equation (2): 1
2
3
i 1
2
3
i
Conveyor
I
j
I
AGV
j (a)
(b)
i
1
j
j
2
3
2
1
3
AGV
I
Figure 1. The linear single-row machine layout with respect to different types: (a) loop layout; (b) straight-line layout; (c) U-shaped layout; and (d) serpentine layout
i (c)
(d)
I
i
Single-row machine layout problem
j
I 1
2
261
3
Figure 2. Precedence diagram
2
P IxI
p11
6 . 6 . 6 . 6 6 p ¼ pij IxI ¼ 6 6 i1 6 . 6 . 6 . 4 pI 1 (
pij : pij ¼
···
p1j
···
..
.
.. .
..
···
pij
···
..
.
.. .
..
···
pIj
···
.
.
p1I
3
.. 7 7 . 7 7 7 piI 7; 7 .. 7 7 . 7 5 pII
1
from station i to station j;
0
other:
ð2Þ
The moving distance algorithm among cutting pieces Based on the from-to matrix and the precedence matrix, we are able to acquire the total distance matrix shown as equation (3) and the total distance D shown as equation (4): M IxI ¼ ½mij IxI ¼ ½t ij £ pij IxI ; D¼
I X I X
mij :
ð3Þ ð4Þ
i¼1 j¼1
Hierarchical order-based genetic algorithm Chromosome As shown in the left part of the precedence diagram of Figure 3, according to the manufacturing order of the sub-assembly line, adjacent machine layout has a minimal distance that modularizes the sub-assembly line in the new sequence and order as shown in the right part of the modular diagram of Figure 3. The hierarchical order-based genetic algorithm chromosomes are consisted of the control genes and the modular genes that are used to acquire the parametric genes as shown in Figure 4. Initial population A population pool of chromosomes can be randomly set initially. A conventional binary genetic algorithm was applied to the control genes. The modular genes adopted
IJCST 20,5
1
3 1
2
262
5
6
Figure 3. Precedence relationship of modular order
2
4
3
7
4
8
5
Precedence diagram
Modular diagram
Chromosome A
0 1 1 1 0 3 1 4 5 2 Modular genes
Control genes Control genes
0
1
1
1
0
Modular genes
3
1
4
5
2
Parametric genes
6
7
8
2
1
3
4
5
XA = (6,2,1,7,8,3,4,5)
1 0 1 1 1 3 1 4 5 2
Chromosome B
Control genes
Figure 4. Chromosomes of the hierarchical order-based genetic algorithm
Modular genes
Control genes
1
0
1
1
1
Modular genes
3
1
4
5
2
Parametric genes
6
7
8
1
2
5
XB = (6,1,2,7,8,5,4,3)
4
3
an order-based genetic algorithm to identify an order-based list such as 1, 2, 3, . . . , I. The positions of the chromosomes were selected in accordance with random number from the order-based list and then deleted it from the list. The procedure was repeated until the order-based list was emptied. Fitness function This study convert the total distance into a fitness function by employing a positive number times the reciprocal of the objective function as shown in equation (5); the one with the larger fitness function is considered optimal: 1 ð5Þ FitðDÞ ¼ 1; 000 £ : D Roulette wheel selection Roulette wheel selection is one of the most common techniques being used for such a proportionate selection mechanism. We put the fitness function in order and divided the value of the individual function by the total function value. The ratio acquired was illustrated as a roulette wheel and the fittest chromosome and occupies the largest interval. The numbering of the larger interval was more likely to be selected (Man et al., 2000). Crossover Not all chromosomes were put into crossover, resulting in the definition of Pc, probability of crossover, with a typical value between 0.6 and 1 (Man et al., 2000). This generates offspring from the parents, based on a randomly generated crossover mask. The control genes adopted the uniform crossover. The operation is demonstrated in the left part of Figure 5. The modular genes adopted the uniform order-based crossover. The operation is demonstrated in the right part of Figure 5. Mutation The definition of Pm, probability of mutation, at a typical value less than 0.1, is used to escape from a local optimum (Man et al., 2000). The control genes adopted the conventional binary genetic algorithm mutation. The modular genes used a scramble sub-list mutation as shown in Figure 6. We randomly selected two bits from the bit string and the chromosomes in between were randomly listed (Chan et al., 1998). Termination condition Normally, the termination condition is shown in three aspects. When more operating times are required than evolution times, the algorithm of the generation is set to terminate. The termination also initiates when the objective function reaches its target value. When the group is shown with homogeneity, convergence is presented and it is close to local optimal solution. This study adopted the generation termination algorithm. Flow chart Flow chart of the hierarchical order-based genetic algorithm is shown in Figure 7. Result and discussion To test the effectiveness of this study, we used the same men’s shirt manufacturing in the genetic algorithm study of 2.
Single-row machine layout problem 263
IJCST 20,5
Parent 1
0 1 1
1 0
1 2 3 4 5
Parent 2
1 0 1
1 1
3 5 1 4 2
Mask
0 1 1
0 1
0 1 1 1 0
264
3 5 1 4 2
Crossover operations
0 1 1
1 0
1 0 1
1 1
2 3 4
3
1 5
2
5 1 4 1 2 3 4 5
Figure 5. Crossover of the hierarchical order-based genetic algorithm
Offspring 1
0 0 1
1 1
5 2 3
4 1
Offspring 2
1 0 1
1 0
3 1 4
5 2
Uniform crossover
1
2
3
4
5
Uniform order-based crossover
6
7
8
Original Chromosome
7
8
New Chromosome
Choose two random positions
1
Figure 6. Scramble sub-list mutation
2
3
5
4
6
permute sublist randomly Remark:with the beginning and the end of the selected sublist marked by "" Source: Chan et al. (1998)
Figure 8 shows the precedence relationship diagram of the example. The chromosomes and their structure are shown in Figure 9. The task of each station is tabulated in Table I. The condition of the example is shown as follows: . population size: 50; . probability of crossover (Pc): 0.6; . probability of mutation (Pm): 0.008; and . termination condition: termination until generation 200. According to the manufacturing order, the machine layout is listed as (1-41). However, before applying the proposed algorithm and put order in the equation, we acquired the total distance among cutting pieces, D1 ¼ 70 and fitness value, Fit(D1) ¼ 14.29.
Single-row machine layout problem
START
Create initial random population
265 Parametric genes evaluate from chromosomes
Apply fitness
Termination criterion satisfied?
Y END
N Roulette wheel parent selection and reproduction
Crossover (Pc)
Mutation (Pm)
By applying the order-based genetic algorithm until generation 200, we acquired the machine layout as (1, 41, 40, 39, 3, 4, 2, 5, 7, 8, 9, 10, 25, 26, 15, 16, 17, 21, 22, 23, 20, 19, 18, 24, 32, 11, 12, 13, 14, 27, 28, 29, 31, 6, 36, 38, 37, 35, 34, 33, 30), resulting in a total distance among cutting pieces, D2 ¼ 68 and fitness value, Fit(D2) ¼ 14.7. After the application of the hierarchical order-based genetic algorithm, the chromosomes of the control genes in Figure 10 are shown as (0, 1, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1); the modular genes as (1, 3, 2, 7, 4, 9, 5, 6, 8, 10, 12, 13, 11). From the control genes and the modular genes we acquired the improved order of machine layout as (1, 2, 3, 4, 5, 6, 12, 11, 10, 9, 8, 7, 23, 22, 21, 17, 16, 15, 14, 13, 25, 26, 27, 28, 29, 18, 19, 20, 24, 30, 31, 32, 33, 40, 41, 39, 38, 37, 36, 35, 34), resulting in a total distance among cutting pieces, D3 ¼ 55, and fitness value, Fit(D3) ¼ 18.18. The total distance and fitness value of the generation algorithm acquired by the two methods are shown as in Figure 11 and the assessment of improved effectiveness is shown as in Table II.
Figure 7. Flow chart of the hierarchical order-based genetic algorithm
IJCST 20,5
1
13
18
21
25
34
2
14
19
22
26
35
3
15
23
27
36
5 7 9
4
11
1
266
4
16
28
37
5
17
29
38
6
7
39
20
8
2
6
24
8
9 30
10
3
31
11
10 32
12
33 40
Figure 8. The precedence relationship diagram of men’s shirts manufacturing
12
41
13
Precedence diagram (Chan et al., 1998)
Modular diagram
Chromosome
0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 10111213 Control genes
Figure 9. Control genes Chromosomes of the hierarchical order-based genetic algorithm in men’s Modular genes shirts manufacturing
Parametric genes
Modular genes
0
0
0
0
0 0
0
0
0
0
0
0 0
1
2
3
4
5 6
7
8
9
10
11
1213
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
Discussion Comparison of the hierarchical order-based genetic algorithm and the order-based genetic algorithm When comparing two algorithm methods, we found the convergence of the order-based algorithm in generation 178 and the hierarchical order-based genetic algorithm in generation 39 indicating the shorter moving distance, better fitness value, quicker convergence, and the optimal solution acquired by the hierarchical order-based genetic algorithm. Advantage of the hierarchical order-based genetic algorithm Advantage of the modular genes: after modularizing the sub-assembly line, the number of groups in search is reduced. From the first to the sixth manufacturing steps in
Task No.
Task name
Task No.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Spot fuse collar fall Top fuse collar fall Sew collar stay pocket Runstitch collar fall Trim, turn, and press collar fall Topstitch collar fall Hem collar band Attach collar band Turn and press collar band Topstitch collar band Sew collar band buttonhole Sew collar band button Set centre front placket Hem right front edge Trim neckline Sew centre front buttonhole Sew right front button Hem pocket mouth Crease pocket Set pocket Sew yoke pleats
Task name
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
Set yoke label Set yoke Join shoulder Set sleeve under placket Set sleeve top placket Finish sleeve placket Sew sleeve placket buttonhole Sew sleeve placket button Set sleeve Topstitch armhole Join side seam Hem bottom Hem cuff Runstitch cuff Turn and press cuff Topstitch cuff Sew cuff buttonhole Sew cuff button Set cuff Set and close collar
Source: Chan et al. (1998)
Chromosome
0
1
0
1
1
0
0 0 1
0
0 1
1
Modular genes
1
3
2
7
4
9
5 6 8
10
1213
11
35
36
1 2 3 4 5 6 12 11 10 9 8 7 23 22 21 17 16 15 14 13 25 26 27 28 29 18 19 20 24 30 31 32 33 40 41 39 38 37 36 35 34
37
38
39
41
40
33
32
31
30
24
20
19
18
29
28
27
26
25
1
2
3
Table I. Tasks of each work station in men’s shirts manufacturing
Modular genes
Control genes
34
267
0 1 0 1 1 0 0 0 1 0 0 1 1 1 3 2 7 4 9 5 6 8 10121311 Control genes
Parametric genes
Single-row machine layout problem
4
5
6
12
11
10
9
8
7
23
22
21
17
16
15
14
13
Figure 3 as the example, there are 6! (¼ 720) types of layouts. After modularization as shown in Figure 12(b), there are only 3! (¼ 6) types of layouts that effectively reduce quantities the search. Advantage of the control genes: as shown in Figure 12(c), when “control genes” is 0, it has the increase order (1-2) and when “control genes” is 1, it has the decrease order (5-4-3). Then these two sub-assembly lines can be moved to the main line to shorten the distance.
Figure 10. U-shaped improved machine layout
IJCST 20,5
Hierarchial order-based genetic algorithm fitness value Order-based genetic algorithm fitness value Hierarchial order-based genetic algorithm moving distance value
Fitness value
268
Figure 11. The fitness value and the moving distance value of each generation
1 11 21 31 41 51 61 71 81 91 101111121131141151161171181191 Generation
U-shaped machine layout
Table II. The assessment of improved effectiveness of U-shaped machine layout
1. Machine layout according to manufacturing order 2. Order-based genetic algorithm 3. Hierarchical order-based genetic algorithm
Total distance D (m)
Assessment of improve effectiveness
D1 ¼ 70 D2 ¼ 68 D3 ¼ 55
250 240 230 220 210 200 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0
Moving distance value
Order-based genetic algorithm moving distance value 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Remarks (fitness value) Fit(D1) ¼ 14.29
Improved by 2.9 percent than D1 Improved by 21.4 percent than D1; improved by 19.1 percent than D2
Fit(D2) ¼ 14.7 Fit(D3) ¼ 18.18
Conclusion This study addressed the topic of minimizing the moving distance among cutting pieces during apparel manufacturing and made the following conclusions: Based on the layout composition theory, modularization decreases the number of layout compositions and provides a corresponding increased or decreased order for the control genes to enable these two sub-assembly lines to change direction to the main line in order to shorten the moving distance among cutting pieces.
D=13 1
4
3
6
5
1 4 3 6 5 2
2
Single-row machine layout problem
(a) Order-based genetic algorithm
269 D=7 1
2
3
4
5
2
1
3
6 1 2 3 4 5 6 D=6
1
2
3
6
4
1
3
2
5 1 2 3 6 4 5
(b) Molecular order 0
0
1
1
3
2
D=5 1
2
3
6
5
(c) Heirarchial order-based generic algorithm
4
1 2 3 6 5 4
From the examples of the U-shaped machine layouts, the proposed hierarchical order-based genetic algorithm has proven to increase effectiveness by 21.4 percent thereby matching the limits of manufacturing order possibilities and improving algorithm effectiveness. References Bragalia, M. (1996), “Optimization of a simulated-annealing-based heuristic for single row machine layout problem by genetic algorithm”, International Transactions in Operation Research, Vol. 3 No. 1, pp. 37-49. Chan, C.C.K., Hui, P.C.L., Yeung, K.W. and Ng, S.F.F. (1998), “Handling the assembly line balancing problem in the clothing industry using a genetic algorithm”, International Journal of Clothing Science and Technology, Vol. 10 No. 1, pp. 21-8. Ficko, M., Brezocnik, M. and Balic, J. (2004), “Designing the layout of single- and multiple-rows flexible manufacturing system by genetic algorithm”, Journal of Materials Processing Technology, Vol. 157-158, pp. 150-8. Heragu, S.S. and Kusiak, A. (1988), “Machine layout problem in flexible manufacturing systems”, Operations Research, Vol. 36 No. 2, pp. 258-68. Ho, Y.C. and Moodie, C.L. (1998), “Machine layout with a linear single-row flow path in an automated manufacturing system”, Journal of Manufacturing Systems, Vol. 17 No. 1, pp. 1-22. Hollier, R.H. (1963), “The layout of multi-product lines”, International Journal of Production Research, Vol. 2, pp. 47-57.
Figure 12. Machine layout order analysis
IJCST 20,5
270
Kumar, K.R., Hadjinicola, G.C. and Lin, T.L. (1995), “A heuristic procedure for the single-row facility layout problem”, European Journal of Operational Research, Vol. 87 No. 1, pp. 65-73. Kusiak, A. and Heragu, S.S. (1987), “The facility layout problem”, European Journal of Operational Research, Vol. 29, pp. 229-51. Man, K.F., Tang, K.S. and Kwong, S. (2000), Genetic Algorithms, Springer, Hong Kong. Ponnambalam, S.G. and Ramkumar, V. (2001), “A genetic algorithm for the design of a single-row layout in automated manufacturing systems”, International Journal of Advanced Manufacturing Technology, Vol. 18 No. 7, pp. 512-19. Sarker, B.M., Wilhelm, W.E. and Hogg, G.L. (1998), “Locating sets of identical machines in a linear layout”, Annals of Operations Research, Vol. 77, pp. 183-207. Suresh, G. and Sahu, S. (1993), “Multiobjective facility layout using simulated annealing”, International Journal of Production Economics, Vol. 32, pp. 239-54. Tompkins, J.A., White, J.A., Bozer, Y.A., Frazelle, E.H., Tanchoco, J.M.A. and Trevino, J. (1996), Facilities Planning, Wiley, New York, NY. Corresponding author Miao-Tzu Lin can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Initial validation of point cloud data from a 3D body scanner
Initial validation of point cloud data
Terry Lerch and Sean Anthony Department of Engineering and Technology, Central Michigan University, Mount Pleasant, Michigan, USA, and
Tanya Domina Department of Human Environmental Studies, Central Michigan University, Mount Pleasant, Michigan, USA
271 Received 1 July 2007 Accepted 18 May 2008
Abstract Purpose – The purpose of this paper is to validate the accuracy of point cloud data generated from a 3D body scanner. Design/methodology/approach – A female dress form was scanned with an X-ray computed tomography (CT) system and a 3D body scanning system. The point cloud data from four axial slices of the body scan (BS) data were compared with the corresponding axial slices from the CT data. Length and cross-sectional area measurements of each slice were computed for each scanning technique. Findings – The point cloud data from the body scanner were accurate to at least 2.0 percent when compared with the CT data. In many cases, the length and area measurements from the two types of scans varied by less than 1.0 percent. Research limitations/implications – Only two length measurements and a cross-sectional area measurement were compared for each axial slice, resulting in a good first attempt of validation of the BS data. Additional methods of comparison should be employed for complete validation of the data. The dress form was scanned only once with each scanning device, so little can be said about the repeatability of the results. Practical implications – Accuracy of the point cloud data from the 3D body scanner indicates that the main issues for the use of body scanners as anthropometric measurement tools are those of standardization, feature locations, and positioning of the subject. Originality/value – Comparisons of point cloud data from a 3D body scanner with CT data had not previously been performed, and these results indicate that the point cloud data are accurate to at least 2.0 percent. Keywords Image scanners, Data structures, Data collection Paper type Research paper
Introduction Three dimensional whole body scanning technology is a relatively new method for measuring the surfaces of objects. Older technology employed optical methods; newer technology uses moire´-based light projection or low power lasers. The latter illuminates a narrow strip of the object’s surface while charged coupled device (CCD) cameras detect the white light or laser light. The output from the CCD cameras results The authors are indebted to Barb Wilson and Marybeth Mey of Central Michigan Community Hospital of Mt Pleasant, MI, USA for providing the CT scans of the dress form. They also thank Dr Pat Kinnicutt of Central Michigan University for his assistance with the data analysis. This work was funded from NSF Grant No. BES-0420791.
International Journal of Clothing Science and Technology Vol. 20 No. 5, 2008 pp. 271-280 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810898881
IJCST 20,5
272
in points of the object’s surface which have been triangulated and assigned 3D spatial coordinates. The collection of these surface points forms a point cloud of data that represents the 3D surface of the volumetric object. Body scanners are a special class of 3D scanners that are built specifically to scan the human body. Originally designed for the fashion and garment industries, body scanners have great potential in a number of other fields such as the medical, health, and ergonomic professions. Using the point cloud data as input, body scan (BS) software can compute over 80 anthropometric measurements in as little as 3 min, compared to traditional methods employing a tape measure that can take at least four hours to complete (Paquette, 1996). Figure 1 represents the process employed by most 3D body scanner companies for determining anthropometric data. Currently, two major issues hinder the application of anthropometric measurements derived from body scanners in the apparel industry. First, there is a general lack of standardization of terminology, positioning within the scanner and feature/landmark identification among the different manufacturers of body scanners (Brooke-Wavell et al., 1994; McKinnon and Istook, 2001, 2002; Hwang and Istook, 2004). The “bust” and the “chest” of a female subject are synonymous terms for certain manufacturers, while others define these terms as separate, distinct locations on the female body. Additionally, a subject’s stance and foot positioning can impact the height, inseam and hip measurements and respiration can significantly impact chest measurements (McKinnon and Istook, 2002). Second, there are no accurate and repeatable measurement techniques which can be compared to the anthropometric data produced by body scanning systems. Conventional means of measuring body dimensions are fraught with inaccuracies. A large body of literature exists on the imprecision of using a tape measure for body dimensions (Gordon and Bradtmiller, 1996; Marks et al., 1997; Williamson et al., 1997). Most problematic is observer error, even by trained observers. Positioning of the tape, mechanical tension in the tape, and compliance in the soft tissue of the body all contribute to poor repeatability of the measurements. In comparison, the anthropometric measurements obtained with 3D body scanners are directly dependent upon the accuracy of the raw point cloud data found earlier in the measurement process as described in Figure 1. There is an implicit assumption that the point cloud data is sufficiently accurate as to not negatively affect the derived anthropometric data, but little if any research has been published to prove this assumption. The purpose of this paper is to estimate the uncertainty of the raw point cloud data and to demonstrate that this data is sufficiently accurate so future research may concentrate on issues of standardization. Instead of relying on conventional measurements made with a tape measure as the standard of comparison, X-ray computed tomography (CT) was employed. CT produces measurements of excellent accuracy for both internal and external features of the scanned object and is often referred to as the “gold standard” within the medical community when it comes to dimensional measurements on the human body. However, utilizing human subjects in the validation process can result in significant errors. Scan subject with body scanner
Figure 1.
Create 3D point cloud data set
Compute anthropometric measurements
Note: Process for obtaining anthropometric data using 3D body scanning technology
First, compliance of soft tissue becomes an issue since the subject is typically scanned in a prone position for a CT scan, but must stand upright when scanned with the body scanner. As mentioned earlier, respiration and foot positioning (McKinnon and Istook, 2002) have a significant effect on the accuracy of anthropometric measurements which are computed directly from the point cloud data – the likelihood of the subject maintaining similar chest positions for both scans is minimal. There are also ethical issues to consider with exposing subjects to unnecessary radiation associated with a CT scan. While comparing BS results to CT scan results offered an excellent way to determine the accuracy of the BS point cloud data, using a living human being as the test subject was not practicable; therefore, a dress form was used as a scanning subject. A dress form possesses many of the surface complexities of the human body without the complications of soft tissue, breathing, or ethics. It should be noted that the body scanner provided data for only the surface of an object and not the internal features of an object. Therefore, all internal information pertaining to the dress form from the CT scans was ignored. This paper will proceed with a brief description of the CT and body scanners used for this study. Measurements of length and cross sectional area for selected axial slices of successive CT and BSs will be compared. These comparisons will quantify the accuracy of the point cloud data produced by the body scanner. A brief summary of the results and discussion will conclude the paper. Description of dress forms and scanning systems The dress form used as the scanning subject was made by the Wolf Co. and is constructed by carving wooden pieces with sizes and curvatures similar to a human body and aligning them on a steel rod to form the human figure. Padding is then applied in order to ensure that the dress form’s measurements conform to a particular size. Lastly, a thin layer of fabric is stretched over the dress form surface to hold the padding in place. A woman’s size 4 dress form was used for the CT-BS comparisons. A Toshiba Aquilion 16 CT scanner was used to provide the reference data for the study. This CT scanner was designed for medical imaging of the human body. Typically, the patient is laid on a bed and passes through a tube where the scanning occurs. In this case, the dress form was placed on the bed in a prone position with two foam blocks between the dress form and bed surface. The foam blocks allowed easier identification of the dress form’s back surface on the resulting CT images by creating a gap between the back of the dress form and the bed. The low attenuating foam was not detected by the CT scanner. The voltage and current from the X-ray tube was adjusted to 120 kV and 200 mA to maximize the contrast of the dress form’s surface contour. Axial slices of 1 mm were computed during the reconstruction of the data. The body scanner used in the comparisons was a Human Solutions Vitus Smart 3D laser scanning system. The scanning system consists of three main components: the scanning assembly or booth, a motor controller, and a PC with image reconstruction software. The scanning assembly is 4 in. wide by 4 in. deep by 10 in. high (Figure 2). With a structural frame to keep the device stationary; curtains are hung from the frame to minimize outside light. Located in each of the four corners are the imaging devices which travel in a elevator-type assembly in a vertical column. Each imaging device houses the essential scanning equipment: a low energy laser, and two CCD cameras.
Initial validation of point cloud data 273
IJCST 20,5 Motor controller Laser
274 Figure 2. 3D laser scanner
CCD cameras
A/D signal converter
Desktop Computer
When the system is calibrated correctly, the four elevator assemblies travel down the columns in unison, sweeping the scanning zone with a horizontal plane of laser light. The laser light illuminates the contours of an object standing within the scanning zone and the CCD cameras record discrete points on these contours at each horizontal plane. The entire scan takes approximately 12 s. A computer containing the user interface, data acquisition/reconstruction, and data analysis software, links with the scanner and the motor controller. The computer software acquires data from the A/D converter and triangulates the discrete points for all of the horizontal planes, creating a point cloud representation of the object scanned. This process takes approximately 2 min to complete and can result in as many as 300,000 data points. After the data acquisition/reconstruction program is completed, a 3D image of the object is displayed on the computer screen. The point cloud data can be exported into proprietary and standard file formats (obj, dxf, sdl, ascii) which can be imported into various computer aided design, finite element analysis, and rapid prototyping software packages. Comparisons As mentioned earlier, selected axial slices of CT and BSs were compared. The slices were nominally chosen at four anatomical locations: the shoulder, the bust, the waist, and the hips (Figure 3). For each slice, cross sectional area and point-to-point distances approximating the major and minor axes of the cross section were determined for both the CT and BSs. Comparisons of these measurements validate the size and shape of the axial slices obtained from the body scanner and offer an initial indication of the spatial accuracy of the point cloud data. An idealized BS slice is shown in Figure 4. The CT data were analyzed with ImageJ freeware which inputs dicom format files and computes the two types of measurements. BS data were analyzed in Human Solutions software for distance measurements and Matlab software for cross sectional area measurements. The thicknesses of the slice planes differ between the CT and BSs. The CT images were reconstructed as 1 mm thick slices while the BS images were reconstructed as 3.6 mm thick slices. Since there are over three CT slices that correspond to each BS slice, the second CT slice was compared to its BS counterpart. Resolution of distances measured with the body scanner software is to the nearest mm. For the CT software, the resolution is to the nearest 0.1 mm. Vertical alignment of the CT and BS images was accomplished by locating steel hemming pins in both scans. Steel hemming pins in the dress form were visible near the waist line and were prominent in the CT scans. A thin strip of black plastic
shoulder
bust
waist
hips
Initial validation of point cloud data 275 Figure 3. Axial slice planes chosen for the CT-BS comparisons. The slice planes nominally correspond to the shoulders, bust, waist, and hips of the female dress form
sheeting was taped directly above the hemming pins on the dress form. The plastic sheet was not detected by the body scanner because of its color and sheen and therefore acts as a marker. The dress form was then body scanned. The axial slice plane in the BS point cloud data directly below the plastic marker corresponds to the CT slice plane in which the hemming pins were detected. A ring or band of point cloud data around the dress form’s surface existed for all the BS axial slices (Figures 4 and 5). The uncertainty in the BS measurements was associated with this band. Most bands varied in thickness from a minimum of approximately 0.5 mm to a maximum of 2.0 mm. The BS measurements were calculated from the midpoint of the band of point cloud data. Figure 5 shows the raw point cloud data represented by white dots of the BS overlaid on the CT images corresponding to the shoulder, bust, waist, and hips of the dress form, respectively. As mentioned above, all data analysis was conducted in separate software programs, i.e. measurements were not made from these images. Note the CT images also display the internal features of the dress forms such as the wooden shell and metal rods. Results and discussion The results are summarized in Table I and Figures 6-8. Table I includes the three types of measurements for the four axial slice planes (shoulder, bust, waist, and hips) for both the CT and BS data. In addition, the percent difference between the CT and BS minor axis
major axis
0.5 ≤ t ≤ 2 mm inner perimeter outer perimeter
Notes: Idealized representation of raw point cloud data from BS for an elliptical axial slice. The ring of point cloud data varied between 0.5 and 2.0 mm in thickness
. Figure 4
IJCST 20,5 Internal Parts of Dress Form
276
Internal Parts of Dress Form
LEGEND: 50 mm Body Scan Data
LEGEND: 50 mm Body Scan Data
(a)
Internal Parts of Dress Form
Figure 5. BS point cloud data overlaid on the CT images of the (a) shoulder; (b) bust; (c) waist; and (d) hips slice planes of the dress form
(b)
Internal Parts of Dress Form
LEGEND: 50 mm Body Scan Data
LEGEND: 50 mm Body Scan Data
(c)
(d)
measurements is presented. Figures 6-8 plot the major axis, minor axis, and cross sectional area measurements and their uncertainties, respectively, for both measurement techniques. Estimated measurement uncertainties The major and minor axes measurements from the CT scans were repeated ten times with the resulting averages reported as the “true” values and the standard deviations reported as the uncertainties. The inner and outer perimeters of the point cloud data were used to estimate the uncertainties associated with the BS distance measurements by measuring the extreme minimal and maximum distances and using these as the lower and upper limits. As mentioned earlier, the averages were considered the best estimates for the distances and compared to the “true” values from the CT scans as shown in Table I. Repeatability of the acquisition of the CT and BS data were
Measurement Shoulder Bust Waist Hips
340.00
Major axis (mm) Minor axis (mm) Area (103 mm2) Major axis (mm) Minor axis (mm) Area (103 mm2) Major axis (mm) Minor axis (mm) Area (103 mm2) Major axis (mm) Minor axis (mm) Area (103 mm2)
336
Actual value BS
Actual value CT
Percent difference
336 ^ 2 132 ^ 1 33.4 ^ 1.1 272 ^ 2 197 ^ 1 51.2 ^ 0.8 202 ^ 1 174 ^ 2 29.2 ^ 0.6 313 ^ 1 216 ^ 2 55.8 ^ 0.8
338.6 ^ 1.5 133.5 ^ 1.5 34.13 ^ 0.27 271.2 ^ 1.0 196.9 ^ 0.7 51.17 ^ 0.05 204.1 ^ 1.0 173.3 ^ 0.5 28.93 ^ 0.03 317.9 ^ 0.7 218.0 ^ 0.6 56.36 ^ 0.06
0.7 1.1 2.0 0.3 0.1 0.1 1.0 0.4 0.8 1.5 0.9 1.0
Initial validation of point cloud data 277
Table I. Measurement results for the four slice planes
338.6
325.00 313
310.00
317.9
Length (mm)
295.00 280.00 272 265.00
271.2
250.00 235.00 220.00 205.00
202
204.1
190.00 Shoulder BS Shoulder CT Bust BS
Bust CT
Waist BS
Waist CT
Hips BS
Hips CT
Slice Name
not tested. The dress form was scanned only once with each of the two measurement techniques. The major and minor axes measurements from the BSs were also repeated ten times. The resulting standard deviations were smaller than the uncertainties associated with the inner and outer perimeters of the point cloud data, therefore larger uncertainties associated with the inner and outer perimeters were used when reporting the BS distance measurements. BS cross sectional area uncertainties were estimated with the maximum (2 mm) point cloud thickness around the slice perimeter. The lower limit for area was computed from the inner perimeter and the upper limit was computed from the outer perimeter. In some cases where the point cloud thickness was significantly smaller than 2 mm, such as the waist and bust, this approach may be quite conservative. CT cross sectional area uncertainties were also estimated by varying the slice perimeter to arrive at the lower and upper limits for uncertainty.
Figure 6. Comparisons of the major axis measurements from the CT and BS data
IJCST 20,5
216
215.00
278
Length (mm)
200.00
197
218.0
196.9
185.00 174 170.00
173.3
155.00 140.00 132
Figure 7. Comparisons of the minor axis measurements from the CT and BS data
125.00 Shoulder BS Shoulder CT Bust BS
Bust CT
Waist BS
Waist CT
Hips BS
Hips CT
Slice Name
Area (103 mm2)
Figure 8. Comparisons of the cross-sectional area measurements from the CT and BS data
133.5
56.00 54.00 52.00 50.00 48.00 46.00 44.00 42.00 40.00 38.00 36.00 34.00 32.00 30.00 28.00
55.81 51.22
33.44
56.36
51.17
34.13
29.17 Shoulder BS Shoulder CT Bust BS
Bust CT
Waist BS
28.93 Waist CT
Hips BS
Hips CT
Slice Name
Discussion As can be seen in Table I, the percentage differences in the BS measurements ranged from 0.1 to 2.0 percent when using the CT measurements as references, indicating that the raw point cloud data was accurate to at least 2.0 percent for cross sectional area and 1.5 percent for the axes measurements. Seven of the twelve percent differences fell
below 1.0 percent. Figures 6-8 show that, with the exception of the hips major axis measurement, differences between all measurements were insignificant based on the estimated uncertainties. Of the four axial slices, the BS measurements associated with the shoulder and hips have the largest percentage differences when compared to their CT counterparts. This result may be related to the fact that the locations of major and minor axes were better defined in the bust and waist cross sections due to their shapes. Landmarks such as the bust cleavage may have aided in positioning beginning and ending points for the distance measurements. Also, some difficulty was experienced while detecting the edges of the shoulder in the CT image, which may explain some of the discrepancy in this comparison. An unexpected outcome from the measurement comparisons was that the BS reconstruction program assigns some of the points of the point cloud data inside or under the actual surface of the object being scanned. Initial comparisons of the axes measurements made from the inner perimeter of the point cloud data indicated a consistent under prediction of the major and minor axes distances. This systematic error disappeared when using the average distances between the inner and outer perimeters of the point cloud data. Conclusion Length and cross sectional area measurements of selected axial slices of a female dress form were made with a CT scanner and body scanner. Using the CT measurements as reference, all the BS measurements were found to have percent differences of less than 2 percent. In many cases, the raw point cloud data from the body scanner had percent differences of less than 1 percent. For the uncertainties assigned to the various measurements, differences between the BS and CT measurements were insignificant. While the results presented are preliminary, they indicate that the point cloud data generated by the type of body scanner tested is sufficiently accurate to be used as an anthropometric measurement tool. These results indicate that the main issues for the use of body scanners as anthropometric measurement tools continue to be of standardization, feature locations, and positioning of the subject, and not the accuracy of the raw point cloud data generated by the body scanner. The comparisons presented here should be considered preliminary because they are quite elementary. Cross sectional area and lengths provide a good starting point to compare the CT and BS data sets for size and shape of the axial slices studied, but additional methods should be considered to verify the results presented here. Such methods might include comparisons of axial slice perimeters, volumes of the body, surface areas of portions of the body, and possibly mathematical curve fitting of the surfaces. References Brooke-Wavell, K., Jones, P.R. and West, G.M. (1994), “Reliability and repeatability of 3-D body scanner (LASS) measurements compared to anthropometry”, Annals of Human Biology, Vol. 21 No. 6, pp. 571-7. Gordon, C. and Bradtmiller, B. (1996), “Interobserver error in a large scale anthropometric survey”, American Journal of Human Biology, Vol. 4, pp. 253-63.
Initial validation of point cloud data 279
IJCST 20,5
280
Hwang, S. and Istook, C.L. (2004), “Standardization and integration of body scan data for use in the apparel industry: body scan data connectivity with apparel CAD”, PhD dissertation, North Carolina State University, Raleigh, NC. McKinnon, L. and Istook, C.L. (2001), “Comparitive analysis of the image twin system and the 3T6 body scanner”, Journal of Textile and Apparel Technology and Management, Vol. 1 No. 2, pp. 1-7. McKinnon, L. and Istook, C.L. (2002), “Body scanning – the effects of subject respiration and foot positioning on the data integrity of scanned measurements”, Journal of Fashion Marketing and Management, Vol. 6 No. 2, pp. 103-21. Marks, G., Habicht, J. and Mueller, W. (1997), “Reliability, dependability, and precision of anthropometric measurements”, American Journal of Epidemiology, Vol. 130 No. 3, pp. 578-87. Paquette, S. (1996), “3D scanning in apparel design and human engineering”, IEEE Computer Graphics and Application, Vol. 16 No. 5, pp. 11-15. Williamson, D., Kahn, H., Burnette, C. and Russell, C. (1997), “Precision of recumbent anthropometry”, American Journal of Human Biology, Vol. 5 No. 2, pp. 159-67. Further reading Simmons, K.P. and Istook, C.L. (2003), “Body measurement techniques: comparing 3D body-scanning and anthropometric methods for apparel applications”, Journal of Fashion Marketing and Management, Vol. 7 No. 3, pp. 306-32. Corresponding author Terry Lerch can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
The effect of weave type on dimensional stability of woven fabrics Mehmet Topalbekirog˘lu and Hatice Ku¨bra Kaynak Department of Textile Engineering, University of Gaziantep, Gaziantep, Turkey
The effect of weave type on woven fabrics 281 Received 13 September 2007 Accepted 8 April 2008
Abstract Purpose – Testing the effect of machine washing and drying on dimensional stability produces information about the fabric types that satisfy consumers during end use. At present, it is a known fact that the weave patterns affect the dimensional stability property of woven fabrics. But the essential requirement is to determine the magnitude of this effect for weave types and establish the proper weave types for end use in definite tolerances. The purpose of this paper is to investigate the dimensional stability properties of 100 percent cotton woven fabrics as a function of weave type. Design/methodology/approach – In total, 12 woven fabrics with different weave derivatives are woven with 100 percent cotton and Ne 30/1 combed ring spun yarn for this investigation. These samples are then washed and dried according to domestic washing and drying standard test procedures. The shrinkage values are measured and then expressed as a percentage of the initial dimensions. Findings – It was observed that weave pattern has a significant effect on the dimensional behavior of woven fabrics. Weave patterns with a high number of interlacings have lower shrinkage values. At the same time, lower yarn crimp values restricted the fabric shrinkage and resulted in better dimensional stability. According to one way ANOVA results, the effect of weave type on dimensional stability is found to be significant ( p , 0.01). In addition to these, Pearson correlation analysis showed that there is an important, positive and fair relationship between the number of washing cycles and total shrinkage. Research limitations/implications – The study covers 100 percent cotton woven fabrics with one type of warp and weft sett. The only finishing treatment applied to the sample fabrics was desizing. No dyeing was carried out. Originality/value – Understanding the magnitude of the effect of weave type on dimensional stability of cotton woven fabrics produces more knowledge about products which satisfy the customers with respect to dimensional stability during usage. Keywords Fabric testing, Cotton, Dimensional measurement Paper type Research paper
Introduction Dimensional stability refers to a fabric’s ability to resist a change in its dimensions. A fabric or garment may exhibit shrinkage, i.e. decrease in one or more dimensions or growth, i.e. increase in dimensions under conditions of refurbishing (Collier and Epps, 1999). Changes occur because tensions in some materials that developed during, yarn spinning, fabrication, and finishing may be relaxed when a material is wetted and dried without tension (Kadolph, 1998). Many problems are related to dimensional changes of materials. Poor dimensional stability can create problems with fit, size, appearance, and suitability for end use. Besides, the problems of product fit and appearance, poor dimensional stability also
International Journal of Clothing Science and Technology Vol. 20 No. 5, 2008 pp. 281-288 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810898890
IJCST 20,5
282
effects fabric density and drape. Materials may become more compact and stiff when they shrink (Kadolph, 1998). Woven fabrics account for the majority of the market for cotton goods. Although dimensional instability and distortion after laundering are perceived by consumers as less of a problem in woven cotton fabrics than in cotton knits, these problems do still occur (Higgins et al., 2003). Refurbishing of textile products, including laundering and dry-cleaning, can affect their shape, dimensions, and other properties. Testing for the effect of refurbishing can help predict consumer satisfaction and is used to develop care labels that are required for textile products (Collier and Epps, 1999). When yarns are woven into fabrics they are subjected to considerable tensions, particularly in the warp direction. In subsequent finishing processes such as tentering or calendaring this stretch may be increased and temporarily set in the fabric. The fabric is then in a state of dimensional instability. Subsequently, when the fabric is thoroughly wetted it tends to revert to its more stable dimensions which results in the contraction of the yarns. This effect is usually greater in the warp direction than in the weft direction (Saville, 2002). The dimensional stability of fabric is its ability to resist shrinkage or stretching. While fiber content has some influence on this property, geometric factors are extremely important. One of the most significant elements in dimensional stability is the degree of tension under which yarns are held during fabric construction. Yarns are held taut during weaving, and after removal from the loom they relax. This relaxation is accelerated when the fabric is first subjected to moisture. As the yarns relax, they return to their original length and pull closer together, so that fabric shrinkage results. Extremely compact fabrics with firm yarns and a high fabric or thread count are less subject to size change than those with loose, soft yarns and low thread count ( Joseph, 1972). Purpose of the study The aim of this study is to clarify the relationship between weave type and dimensional stability for 100 percent cotton woven fabrics. Determining the most proper weave types for customer satisfaction is intended. Materials and method Sample fabrics were woven by a Dornier loom with electronic dobby shedding mechanism and rapier weft insertion with 450 rev/min loom speed. The warp sheet was prepared by a sample warping machine, and sized with a sample sizing machine. After sizing straight draft, was applied to warp sheet with a sample drawing machine. Twelve frames were used for all sample fabrics with straight draft. The only finishing treatment applied to the samples was desizing. Fabrics were treated to enzymatic desizing for 6 h. Then they were washed with 60 m/min washing speed in washing machine which have five vats with the liquor temperatures of 95, 95, 85, 65, and 308C, respectively. Drying operation of fabrics was done at 1208C and 140 cm fabric width by 30 m/min drying speed. Sizing, desizing, and washing recipes are given in Table I. These samples were produced as men’s shirts. Weft sett is 28 wefts/cm and warp sett is 46 warps/cm for all samples. The component yarn used for both warp and weft is Ne 30/1, 100 percent cotton combed ring yarn. Detailed information about the yarn used to produce the fabric samples with relevant structures is given in Table II.
Structural views of 12 woven fabrics tested in this experimental study are given in Figure 1 and fabric weight (g/m2), yarn crimp (percent), fabric thickness (mm) properties of samples are shown in Table III. In this study, we report an experimental investigation of dimensional stability properties of different weave types of woven fabrics. The shrinkage of fabrics were tested according to ISO 6330 Textiles – domestic washing and drying procedures for
The effect of weave type on woven fabrics 283
Sizing recipe
Desizing recipe
Washing recipe
150 l liquor 20 kg EMSIZE CMS 60 10 kg BP20 (PVA) 500 g Glissofil extra (oil)
2.5 g/l Torozym NT 2 g/l Schnellnetzer KE 1 g/l R. Entlu¨fter BK 1 g/l Emulgator BE-O
2 g/l Sevalin D 1 g/l Schnellnetzer KE 0.6 g/l Optiderm BS-L
Yarn twist Yarn strength Hairness Unevenness percent Thick þ 50 percent/km Neps þ 200 percent/km Fiber length Fiber fineness
P1
919 turns/m 23 Rkm 5.67 9.69 20 71.7 29.4 mm 4.25 micronaire
P2
P3
Table I. Sizing, desizing, and washing recipies
Table II. Properties of the yarn used in the experimental study
T1
T2
T3
T4
T5
T6
T7
T8
S1
Figure 1. Structural views of weave types. Pl, regular weft rib; P2, irregular weft rib combined rib; P3, basket weave; Tl, balanced twill; T2, unbalanced twill; T3, weft faced twill; T4, warp faced twill; T5, warp faced twill; T6, twill 2/1; T7, twill 2/2; T8, twill 3/1; S1, sateen
IJCST 20,5 Weave type Plain
284
Table III. Fabric properties of samples
Twill
Sateen
Sample code
Fabric weight (g/m2)
P1 P2 P3 Tl T2 T3 T4 T5 T6 T7 T8 S1
150 150 146 150 148 150 149 148 148 149 149 141
Yarn crimp (percent) Warp Weft 1.6 1.7 4.7 5.3 4.9 4.9 4.9 4.4 6.8 6.0 4.9 2.9
16.7 16.9 13.5 13.6 14.8 14.3 14 13.8 13.6 14.1 12.5 11.5
Fabric thickness (mm) 0.48 0.38 0.42 0.31 0.29 0.30 0.33 0.33 0.28 0.30 0.30 0.41
textile testing and ISO 3759 Textiles – preparation, marking and measuring of fabric specimens and garments in tests for determination of dimensional change. The shrinkage was determined after washing by measuring the final dimensions and are then expressed as percentages of the initial dimensions for both warp and weft direction. In order to understand the statistical importance of weave pattern on dimensional stability of woven fabrics, a one way ANOVA was performed. To determine the groups of weave pattern types and the effects of these groups on dimensional stability Tukey multiple comparison test was used, In addition to these, Pearson correlation analysis was done to show the relationship between shrinkage and number of washing from statistical approach. For this aim the statistical software package SPSS 8.0 was used to interpret the experimental data. All test results were assessed at significance levels of p , 0.05 and p , 0.01. Results Table IV and Figure 2 show the mean shrinkage values of different weave types for both warp and weft direction after first washing. In weft direction samples P1, P2, and P3 exhibits considerably high shrinkage values. This is a result of low number of interlacings of these samples in weft direction in comparison to other samples. Sample P1 has the highest shrinkage value of 6.3 percent and lowest number of interlacings in weft direction. Sample P1 is followed by samples P2 and P3 with the values of 6 and 5.9 percent, respectively. It is obvious from Figure 2 that twill weave samples which have high number of interlacings in weave structure for weft direction exhibit fairly low shrinkage values. Sample T1 which is a twill weave derivative has the lowest shrinkage value of 0.9 percent. The highest weft direction shrinkage value among twill weave types 1.9 percent belongs to sample T5 is even an acceptable value. In warp direction, the lowest shrinkage value belongs to sample P2 (2.3 percent) and it is followed by sample P1 (3.6 percent). These samples have the same interlacing number but different shrinkage values in warp direction. Because in sample P2 the
Shrinkage (percent) Weave type
Sample code
Weft
Warp
P1 P2 P3 Tl T2 T3 T4 T5 T6 T7 T8 S1
9.0 6.0 5.9 0.9 1.4 1.2 1.7 1.9 0.9 1.7 1.1 4.7
3.6 2.3 6.0 5.7 5.8 5.6 7.0 7.6 5.9 5.9 5.6 4.9
Plain Twill
Sateen
The effect of weave type on woven fabrics 285
Table IV. Shrinkage percentage of weave types after first washing
10 weft
warp
Shrinkage (%)
8 6 4 2 0 P1
P2
P3
T1
T2
T3 T4 Samples
T5
T6
T7
T8
S1
interlacing points in weft direction restricts the warp yarn movement and so shrinkage in warp direction. Samples T4 and T5 have the highest shrinkage values because of their lower number of interlacing in warp direction. For our samples, it is obvious that increasing the number of interfacings in the weave structure causes lower shrinkage values. But sample S1 which has a sateen weave structure exhibits lower shrinkage values than expected for both warp and weft direction contrary to this mentioned assumption. This situation is a result of lower warp and weft yarn crimp values than other samples. Because during fabric shrinkage yarns swell by increasing their diameter via shortening their length simultaneously. So increasing the yarn length in fabric structure by increasing crimp helps fabric to shrink more easily. In other words decreasing yarn crimp value restrict the yarns to shorten and fabrics to shrink. According to one way ANOVA results of dimensional stability test, weave type has a significant effect on weft direction shrinkage (F(n 2 24) ¼ 387.98, p , 0.01), on warp direction shrinkage (F(n 2 24) ¼ 64.17, p , 0.01) and on total shrinkage
Figure 2. Shrinkage percentage of weave types after first washing
IJCST 20,5
286
(F(n 2 24) ¼ 56.34, p , 0.01). When the weave types used in this study are observed very different structures are seen. Tukey post hoc test created groups of samples according to the effect of weave type on average shrinkage values. These results showed that at the end of the dimensional stability test, the weave constructions of samples P1 ðX ¼ 6:30Þ and P3 ðX ¼ 5:97Þ consisted the group that has the highest effect ( p , 0.05) on shrinkage while samples Tl ðX ¼ 3:30Þ, T6 ðX ¼ 3:37Þ, T8 ðX ¼ 3:37Þ, T3 ðX ¼ 3:40Þ consisted the group that cause the lowest increase on shrinkage value ( p , 0.05). Figure 3 exhibits the shrinkage values of samples in both warp and weft direction after ten washing cycles. Shrinkage values increase by successive washings but dimensional stability behavior of samples do not differ so much. The highest and the lowest shrinkage values are belong to same samples in first washing. The Pearson correlation analysis showed that there is an important, positive and fair relationship between number of washing and shrinkage values (r ¼ 0.557, p , 0.01). The calculated r 2 value is 0.310. This means increasing number of washing cycles cause higher shrinkage. Figure 4 exhibits the scatter diagram that shows the relationship between number of washing and shrinkage. Discussion and conclusion This experimental study on dimensional stability property of different weave structures showed that weave type has a significant effect on shrinkage values. Weave types with high number of interfacings showed better dimensional stability results. In this respect samples Tl, T3, T6, and T8 exhibited acceptable average shrinkage values which are lower than 5 percent. These samples all have high number of interlacings in both warp and weft direction. In addition to this sample S1 which is a sateen weave type having the lowest number of interlacing in both directions have a good dimensional stability with 4.8 percent average shrinkagein contrary to expectations. This situation is a result of low crimp values of this sample. So this experimental study brings the importance of crimp values to light with respect to dimensional stability. Also according to performed one way ANOVA results the effect of weave type on dimensional stability is found to be significant ( p , 0.01). Tukey post hoc test grouped samples Tl, T3, T6, and T8 as the best group and samples P1 and P3 as the worst group. 16
Shrinkage (%)
weft
warp
12
8
4
Figure 3. Shrinkage of different weave types after ten washings
0 P1
P2
P3
T1
T2
T3 T4 Samples
T5
T6
T7
T8
S1
The effect of weave type on woven fabrics
14
Shrinkage (%)
12 10
287 8 6 4 2 0
1
2
3
4 5 6 7 Number of washing
8
9
10
It is clear from test results that for a good dimensional stability woven fabrics must have a good number of interlacing in both weft and warp direction. It is not enough to have a good interlacing number in one direction because of this reason woven fabric must have a balanced structure. In addition to this yarn crimp has an important effect on dimensional stability fabrics with low crimp values have good dimensional stability. Recommendations for future research In this study, the samples are produced with only one weft sett and only one warp sett. The further study should be detailed on the effect of yarn sett to ensure more balanced weave structures. By this way the effects of yarn sett on dimensional stability could be investigated. An other point that attracts attention, related to yarn sett is the effect of crimp on dimensional stability because different yarn setts result different crimp values. As a result of this research, we observed that low crimp values restrict shrinkage even if the number of interlacing is very low. So by varying the sett values on some weave types the mentioned effect of crimp can be observed in a detailed manner. Our sample fabrics are planned to be used as summer shirts. So from this point of view, keeping the cloth comfort in mind 100 percent cotton yarn is chosen as raw material, In future study, different types of synthetic blended cotton yarns should be used. By this way the effect of the type of synthetic fiber and the effect of blend ratio should be determined in addition to weave pattern. In this study, the sample fabrics are neither dyed nor finished. The only treatment is enzymatic desizing. The fabric types which are determined to have good dimensional stability at the end of this study should be chosen and treated with different finishing and dyeing operations. By this way it can be achieved to develop dimensional stability of these chosen weave types via finishing and dyeing.
Figure 4. Scatter diagram of samples for Pearson bivariate correlation analysis
IJCST 20,5
288
References Collier, B.J. and Epps, H.H. (1999), Textile Testing and Analysis, Prentice-Hall, Englewood Cliffs, NJ. Higgins, L., Anand, S.C., Holmes, D.A. and Hall, M.E. (2003), “Effects of various home laundering practices on the dimensional stability, wrinkling, and other properties of plain woven cotton fabrics: experimental overview, reproducibility of results, and effect of detergent”, Textile Research Journal, Vol. 73, pp. 357-66. Joseph, M.L. (1972), Textile Science, Holt, Rinehart & Winston, San Diego, CA. Kadolph, S.J. (1998), Quality Assurance for Textiles and Apparel, Fairchild, New York, NY. Saville, B.P. (2002), Physical Testing of Textiles, Woodhead, Cambridge. Corresponding author Mehmet Topalbekirog˘lu can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
Study on heat transfer of liquid cooling garment based on a novel thermal manikin Kai Yang School of Fashion and Art, Zhongyuan University of Technology, Shanghai, People’s Republic of China
Ming-Li Jiao
Study on heat transfer of liquid cooling garment 289 Received 30 January 2008 Revised 24 May 2008 Accepted 24 May 2008
Department of Materials and Chemistry, Zhongyuan University of Technology, Shanghai, People’s Republic of China, and
Yi-Song Chen, Jun Li and Wei-Yuan Zhang College of Fashion, Donghua University, Shanghai, People’s Republic of China Abstract Purpose – The purpose of this paper is to explore the heat transfer and establish a heat transfer model of an extravehicular liquid cooling garment based on a thermal manikin covered with soft simulated skin. Design/methodology/approach – The thermal manikin applied in this study was a copper manikin, typical of which was its soft simulated skin – a newly thermoplastic elastomer material. Based on this novel thermal manikin, the heat transfer analysis of an extravehicular liquid cooling garment was performed. To satisfy the practical engineering application and simplify analysis, the hypotheses were proposed, and then the heat transfer model was established by heat transfer theory, in which the heat exchange equation of the liquid cooling garment with the thermal manikin and with the air layer, and the garment’s total heat dissipating capacity were derived. Findings – The verification experiments performed in a climatic chamber by a thermal manikin wearing a liquid cooling garment at different surface temperatures of the thermal manikin show that the modeling value fits well with the experimental value, and the heat transfer model of the liquid cooling garment has a high accuracy. Meanwhile, the relationship between the heat-dissipating capacity of the liquid cooling garment and its design parameters – inlet temperature and liquid velocity – is suggested as being based on the heat transfer model. Originality/value – The paper shows that it is an effective method to control the heat-dissipating capacity of a liquid cooling garment by changing the inlet temperature to some degree, but not by changing the liquid velocity. Keywords Liquidity, Cooling, Heat transfer, Thermal measurement, Elastomers Paper type Research paper
Introduction As the development of manned spaceflight, astronaut’s extravehicular activity is becoming more and more frequent. Extravehicular environment is very bad: high vacuum, strong radiation, and tremendous temperature changes. In that environment, astronaut cannot live or work, and they must be protected in the closed environment of extravehicular spacesuit. Extravehicular spacesuit is composed of many layers, such as outside cover, vacuum insulation layer, air retaining layer, ventilation structure, liquid cooling garment, and underwear (Skoog et al., 2002; Moyers et al., 2006). Among them, liquid cooling garment is used to exhaust most of energies of astronaut
International Journal of Clothing Science and Technology Vol. 20 No. 5, 2008 pp. 289-298 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810898908
IJCST 20,5
290
in extravehicular activity and make astronaut comfortable (Koscheyev et al., 2005; Doerr, 2001), so the research on liquid cooling garment is very important for development of extravehicular spacesuit. Liquid cooling garment is equivalent to heat exchanger as to its function, which is usually combined with cooling source to compose a complete refrigeration system, and to dissipate human’s metabolic heat and maintain human’s dynamic heat balance. So its heat dissipation capacity is the key index in designing and evaluating liquid cooling garment. Liquid cooling garment’s design is focusing on different heat dissipation distribution in different region. And in evaluating liquid cooling garment’s heat dissipation capacity, human must maintain high and steady metabolic rate, which is very difficult for human. So the research of liquid cooling garment always applies thermal manikin to replace real body, which can solve garment design parameter’s selection, cooling capacity’s distribution and improve garment’s heat dissipation performance. Present thermal manikin is always consisted of hard metal material (Cheong et al., 2006; Liu et al., 1999), such as copper (Gonzalez et al., 1998) and aluminum, which surface hardness is much higher than real body’s skin. In addition, metal material’s heat transfer is very rapid, so manikin’s surface temperature will change greatly with environment temperature, which affects thermal manikin’s stability and further affects garment’s measurement. The manikin “Walter” developed by Hong Kong Polytechnic University is enclosed by a kind of fabric (Chen et al., 2003). Fabric is generated through spinning, twisting, weaving or knitting, which makes it an anisotropic material. And fabric’s physical properties such as compress elasticity is very different with real body’s skin, which leads to contact area of liquid cooling garment with thermal manikin is smaller than with real body, and further affects garment’s heat dissipation research. So heat transfer analysis of liquid cooling garment based on a soft material thermal manikin is much of practical significance. In this study, a novel copper thermal manikin was applied, which was developed by Donghua University and Research Institute of China Space Medico-Engineering together. The typical characteristic was its soft simulated skin covered on the outer surface of copper layer, which was a soft thermoplastic elastomer material (Steller et al., 2006), and had good physical properties (Pinchuk et al., 2008; Rajan et al., 2004) much similar to human skin. Based on this thermal manikin and heat transfer theory, research on liquid cooling garment’s heat transfer was performed and the heat transfer model was established, in which the heat exchange equation of liquid cooling garment with thermal manikin and with air layer, and garment’s total heat dissipating capacity were derived. Experimental verification of garment’s total heat dissipating capacity by thermal manikin wearing liquid cooling garment was performed, which indicated that the model has a high accuracy. Finally, based on the heat transfer model, the relationship between liquid cooling garment’s heat dissipating capacity and its design parameter (inlet temperature and liquid velocity) was studied. Hypothesis of model Heat transfer (Yigit, 1998; Ghaddar et al., 2005) of liquid cooling garment with human (or thermal manikin) is complicated, which produce the following heat flow (Figure 1): heat conduction flow from human body to underwear (qs1), heat conduction flow from underwear to garment’s tube wall (qs2), mass transfer flow from human to garment (qM), human’s radiant heat flow (qr1), convective heat flow between outer surface of tube wall and environment (qc), radiant heat flow of extravehicular spacesuit’s shell (qr2). So, to
Study on heat transfer of liquid cooling garment
qr1 qc qs1
qs2 qr2
291
qM
human body
underwear
liquid cooling garment
Figure 1. Heat flow diagram of liquid cooling garment
air
satisfy practical engineering application and simplify analysis, the following hypotheses are proposed in establishing heat transfer model of liquid cooling garment by thermal manikin: (1) Liquid cooling garment’s heat transfer is analyzed only on steady state, which simplifies the analyzing and decreases the actual calculation greatly. (2) Heat transfer of basic clothing in liquid cooling garment and the heat transfer of thermal manikin are along their normal direction. (3) The radius of tube in liquid cooling garment is much smaller than curvature radius of all sections in thermal manikin, so liquid cooling garment’s basic clothing and thermal manikin are all treated as homogeneous plate. Their physical parameters are homogeneous. (4) Underwear is cotton or cotton/flax blended knitting-fabrics, which has little influence on heat transfer between liquid cooling garment and thermal manikin. So the underwear is neglected in analysis. (5) To simplify heat transfer analysis, liquid cooling garment is in outside environment. In addition, the novel manikin is a dry thermal manikin, so sensible heat transfer is considered only. Heat transfer model Taking liquid cooling garment as research object, its heat transfer is cooling liquid’s heat exchange with thermal manikin and air layer by convection and conduction. The heat circuit diagram is shown in Figure 2. Take micro-unit dl in liquid cooling garment’s tube, and its temperature variable quantity is dt. According to heat transfer theory, the heat (dq) taken away by cooling liquid in micro-unit dl is: dq ¼
p 2 d urC p dt; 4
ð1Þ
copper layer
simulated skin
tube wall
cooling liquid
tube wall
air
d1 l1
d2 l 0 (1 + bt)
D D 1n 2l 3 d
D a 1d
D D 1n 2l 3 d
1 1–e + a2 e
Figure 2. Liquid cooling garment’s heat circuit diagram
IJCST 20,5
292
where d is the inner radius of tube (m); u, liquid velocity (m/s); r, liquid density (kg/m3); Cp, liquid specific heat at constant pressure ( J/(kg 8C)). According to Figure 4, the heat exchange resistance between liquid cooling garment and thermal manikin includes copper layer’s heat conductive resistance, simulated skin’s heat conductive resistance, tube wall’s heat conductive resistance, convective resistance between cooling liquid and tube wall. So the heat exchange quantity (dq1) between micro-unit dl in liquid cooling garment and thermal manikin is: dq1 ¼ pDmdl d1 l1
1 þ
d2 l2
þ
D 2 l3
ln Dd þ aD1 d
ðt w3 2 tÞ
1 ¼ pDmdl d1 ðtw3 2 tÞ; d2 D D D l1 þ l0 ð1þbtÞ þ 2l3 ln d þ a1 d
ð2Þ
where D is the external diameter of tube (m); m, effective area ratio of liquid cooling garment covering thermal manikin; d1, thickness of copper layer (m); l1, thermal conductivity of copper layer (W/(m 8C)); d2, thickness of simulated skin layer (m); l2, thermal conductivity of simulated skin layer (W/(m 8C)); l0, thermal conductivity of simulated skin layer at 08C (W/(m 8C)); b, constant; l3, thermal conductivity of tube wall (W/(m 8C)); a1, convective heat transfer coefficient between cooling liquid and tube wall (W/(m 8C)); tw3, interface temperature of simulated skin and tube wall (8C); t, temperature of cooling liquid (8C). Let:
d1 D D D ¼ K1; þ ln þ l1 2l3 d a1 d then equation (2) can be simplified to the following equation: dq1 ¼ pDmdl
1 d2 K 1 þ l0 ð1þbtÞ
ðt w3 2 tÞ:
ð3Þ
The heat exchange resistance between liquid cooling garment and air layer includes convective resistance between cooling liquid and tube wall, tube wall’s heat conductive resistance, convective and radiative resistance between tube wall and air layer. So the heat exchange quantity (dq2) between micro-unit dl in liquid cooling garment and air layer is: dq2 ¼ pDð1 2 mÞdl
D a1 d
1 ðt f 2 tÞ; þ 2Dl3 ln Dd þ a12 þ 121 1
ð4Þ
where a2 is the convective heat transfer coefficient between tube wall and air layer (W/(m 8C)); 1, blackness of tube wall; tf, temperature of air layer (8C). Let: 1 D a1 d
þ
D 2l3
ln Dd
þ a12 þ 121 1
¼ K 2;
then equation (4) can be simplified to the following equation: dq2 ¼ pDð1 2 mÞK 2 ðtf 2 tÞdl:
ð5Þ
According to the first law of thermodynamics, we get: dq ¼ dq1 þ dq2 :
ð6Þ
Put equations (1), (3), (5) into (6), and we get:
p 2 1 d urC p dt ¼ pDm ðt w3 2 tÞdl þ pDð1 2 mÞK 2 ðt f 2 tÞdl: d2 4 K 1 þ l ð1þbtÞ
ð7Þ
0
Separate the variables, we get: dl ¼
X 1 þ X 2t dt; X 3 þ X 4t þ X 5t 2
ð8Þ
where: X 1 ¼ durC p ðK 1 l0 þ d2 Þ; X 2 ¼ durC p K 1 l0 b; X 3 ¼ 4ðml0 t w3 þ K 1 K 2 tf l0 þ K 2 t f d2 2 K 1 K 2 t f l0 m 2 K 2 mtf d2 Þ; X 4 ¼ 4ðml0 bt w3 2 ml0 þ K 1 K 2 l0 bt f 2 K 1 K 2 l0 2 K 2 d2 2 K 1 K 2 ml0 bt f þ K 1 K 2 ml0 þ K 2 md2 Þ; X 5 ¼ 4ðK 1 K 2 ml0 b 2 ml0 b 2 K 1 K 2 l0 bÞ: X 1 þ X 2t X 3 þ X 4t þ X 5t 2 can be decomposed to the style of:
A C 2 tþB tþD
so make integral to equation (8) and get: Z l Z t A C dl ¼ 2 dt; tþD 0 in t þ B
ð9Þ
where tin is the inlet temperature of cooling liquid (8C). So: l ¼ ½A lnðt þ BÞ 2 C lnðt þ DÞ 2 E;
ð10Þ
where E ¼ A lnðt in þ BÞ 2 C lnðt in þ DÞ. After lots of calculation, we summarized that A < C and B < 0. So equation (10) can be simplified to the following equation:
Study on heat transfer of liquid cooling garment 293
IJCST 20,5
l ¼ A ln
t 2 E: tþD
ð11Þ
So, the relationship of cooling liquid’s temperature t varied with tube length l is: t¼
294
D 2lþE A
e
21
:
ð12Þ
Put equation (12) into equation (2) and make integral, we get the total heat exchange between liquid cooling garment and thermal manikin is: Z L 1 pD m ðt w3 2 tÞdl q1 ¼ d2 þ K 0 1 l0 ð1þbtÞ Z L 1 D ð13Þ pD m 2 t ¼ dl; w3 d2 2lþE A þ K 21 e 0 1 l0
1þ e
bD 2lþE A 21
where L is the tube length (m). Make variable substitution m ¼ e2ððLþEÞ=AÞ , and we get: q1 ¼
Z
2LþE A
e
e
2E A
ð2pdml0 AÞ
ðtw3 m 2 þðDbt w3 22t w3 2DÞmþðt w3 2Dbt w3 þD 2bD 2 Þ dm: ðK 1 l0 þ d2 Þm 3 þðK 1 l0 Db22K 1 l0 22d2 Þm 2 þðK 1 l0 þ d2 2K 1 l0 DbÞm ð14Þ
Fraction: t w3 m 2 þðDbt w3 22t w3 2DÞmþðtw3 2Dbt w3 þD 2bD 2 Þ ðK 1 l0 þ d2 Þm 3 þðK 1 l0 Db22K 1 l0 22d2 Þm 2 þðK 1 l0 þ d2 2K 1 l0 DbÞm can be decomposed to the style of: G I K þ þ ; mþF mþH mþJ so make integral to equation (14) and we get: " # LþE LþE LþE e2 A þF e2 A þH e2 A þJ þI ln E þK ln E q1 ¼ pdml0 A Gln E e2A þF e2A þH e2A þJ
ð15Þ
Put equation (12) into equation (5) and make integral, we get the total heat exchange between liquid cooling garment and air layer is: Z L Z L D q2 ¼ pDð12 mÞK 2 ðtf 2tÞdl ¼ pDð12 mÞK 2 tf 2 lþE dl: ð16Þ e2 A 21 0 0
Make variable substitution m ¼ e2ððLþEÞ=AÞ , and we get: " # LþE e2 A 21 L q2 ¼ pdð12 mÞK 2 A Dln E þ ðtf þDÞ : e2A 21 A
ð17Þ
Study on heat transfer of liquid cooling garment
So, liquid cooling garment’s total heat dissipating capacity q is: q ¼ q1 þq2 "
LþE A
e2
þF
e2
LþE A
þH
e2
LþE A
þJ
þI ln E þK ln E E e2A þF e2A þH e2A þJ " # 2LþE e A 21 L þ pdð12 mÞK 2 A Dln E þ ðtf þDÞ : e2A 21 A
¼ pdml0 A Gln
295
# ð18Þ
Experimental verification The verification experiments were performed in climatic chamber by thermal manikin wearing liquid cooling garment. The environmental temperature was 218C, relative humidity was 60 percent, and wind velocity was less than 0.1 m/s. When thermal manikin’s surface temperature was stabilized at 30, 33, 37, 40, 45, and 508C, respectively, the experiments of adjusting liquid cooling garment’s different inlet temperature were performed (garment’s liquid velocity was 2 l/min). After experiment reached stable state, liquid cooling garment’s outlet temperature tout was recorded, and the total heat dissipating capacity q was calculated by the following formula: p ð19Þ q ¼ d 2 urC p ðt out 2 tin Þ: 4 Heat dissipating capacity’s experimental value was compared with modeling value, and the comparative diagram was shown in Figure 3. From Figure 3 we can see that the modeling value is fitting well with the experimental value, and after calculation the maximum deviation is less than 3.95 percent, which indicates that the liquid cooling garment’s heat transfer model has a high accuracy. Application Liquid cooling garment’s design parameters, such as inlet temperature and liquid velocity have great influence on garment’s heat dissipating capacity. So based on the heat transfer model above, the relationship study on heat dissipating capacity and design parameters was performed. Figure 4 shows the diagram of heat dissipating capacity changes with inlet temperature when the liquid velocity is in the range of 0.1-4 l/min. We can see that heat dissipating capacity has an approximately linear relation with inlet temperature, and it decreases with the increase of inlet temperature. When liquid velocity changes, especially increases to certain degree, the slope of variation curve changes slightly. So, the effect of inlet temperature on heat dissipating capacity is obvious and it is an effective method to control heat dissipating capacity by changing inlet temperature in some degree. Figure 5 shows the relationship between heat dissipating capacity changes with liquid velocity when inlet temperature is in the range of 5-208C. We can see that when liquid velocity is smaller, heat dissipating capacity increases with the increase of liquid velocity; while when liquid velocity increases to certain degree, heat dissipating
IJCST 20,5
1,800 1,700 1,600
296
heat dissipating capacity q (W)
1,500 1,400 1,300 1,200 1,100 1,000 900 800 700 600 500 400 300 200 30
Figure 3. Comparative diagram of modeling value and experimental value
35 40 45 thermal manikin's surface temperature (°C)
modeling value when tin = 5°C modeling value when tin = 10°C modeling value when tin = 15°C modeling value when tin = 20°C
50
experimental value when tin = 5°C experimental value when tin = 10°C experimental value when tin = 15°C experimental value when tin = 20°C
Notes: Reproduced from the only available original
capacity tends to be stable and its variable quantity changing with liquid velocity becomes smaller and smaller. When inlet temperature changes, the variation curves of heat dissipating capacity – liquid velocity change significantly. So, there needs very wide range of liquid velocity if control heat dissipating capacity only by liquid velocity. While in practical application, very big liquid velocity is not allowed because of the structure and strength of tube and the pressure of pump. Therefore, it is not advisable to control heat dissipating capacity by changing liquid velocity. Conclusion This study investigates heat transfer of extravehicular liquid cooling garment based on a novel thermal manikin covering with soft simulated skin. By heat transfer theory and basic hypotheses, the heat transfer model of liquid cooling garment is established, in which the heat exchange equation of liquid cooling garment with thermal manikin and with air layer, and garment’s total heat dissipating capacity are derived. To evaluate the accuracy of model, the verification experiments are performed in climatic chamber by thermal manikin wearing liquid cooling garment at different manikin’s surface temperature, which show that the modeling value is fitting well with the experimental value, and the maximum deviation is less than 3.95 percent, so the heat transfer model of liquid cooling garment has a high accuracy. Finally, the relationship between heat dissipating capacity of liquid cooling garment and its design
Study on heat transfer of liquid cooling garment
1,500 1,400 1,300 u = 0.1L/min u = 0.3L/min u = 0.5L/min u = 1L/min u = 2L/min u = 3L/min u = 4L/min
heat dissipating capacity q (W)
1,200 1,100 1,000 900 800 700
297
600 500 400 300 200 100 4
6
8
10 12 14 inlet temperature tin (°C)
16
18
20
Figure 4. Diagram of heat dissipating capacity changes with inlet temperature
Notes: Reproduced from the only available original
1,400 1,300
heat dissipating capacity q (W)
1,200 1,100 1,000 900 800 700 600 500 400 300 200 100
tin = 20°C
tin = 15°C
0
tin = 10°C
tin = 5°C
–100 0
5
10 15 liquid velocity u (m/s)
Notes: Reproduced from the only available original
20
Figure 5. Diagram of heat dissipating capacity changes with liquid velocity
IJCST 20,5
298
parameters (inlet temperature and liquid velocity) is studied based on the heat transfer model. It indicates that the effect of inlet temperature on heat dissipating capacity is obvious and it is an effective method to control heat dissipating capacity by changing inlet temperature in some degree, while it is not advisable by changing liquid velocity. References Chen, Y.S., Fan, J. and Zhang, W. (2003), “Clothing thermal insulation during sweating”, Textile Research Journal, Vol. 73 No. 2, pp. 152-7. Cheong, K.W.D., Yu, W.J., Kosonen, R., Tham, K.W. and Sekhar, S.C. (2006), “Assessment of thermal environment using a thermal manikin in a field environment chamber served by displacement ventilation system”, Building and Environment, Vol. 41 No. 12, pp. 1661-70. Doerr, D.F. (2001), “Development of an advanced rocket propellant handler’s suit”, Acta Astronautica, Vol. 49 Nos 3-10, pp. 463-8. Ghaddar, N., Ghali, K., Harathani, J. and Jaroudi, E. (2005), “Ventilation rates of micro-climate air annulus of the clothing-skin system under periodic motion”, International Journal of Heat and Mass Transfer, Vol. 48 No. 15, pp. 3151-66. Gonzalez, R.R., Endrusick, T.L. and Levell, C.A. (1998), “Biophysical properties and skin wettedness of garments with variable moisture vapor transmission rates (MVTR)”, Journal of Thermal Biology, Vol. 23 No. 1, pp. 41-8. Koscheyev, V.S., Leon, G.R. and Coca, A. (2005), “Finger heat flux/temperature as an indicator of thermal imbalance with application for extravehicular activity”, Acta Astronautica, Vol. 57 No. 9, pp. 713-21. Liu, X., Abeysekera, J. and Shahnavaz, H. (1999), “Subjective evaluation of three helmets in cold laboratory and warm field conditions”, International Journal of Industrial Ergonomics, Vol. 23 No. 3, pp. 223-30. Moyers, M.F., Saganti, P.B. and Nelson, G.A. (2006), “EVA space suit proton and electron threshold energy measurements by XCT and range shifting”, Radiation Measurements, Vol. 41 Nos 9-10, pp. 1216-26. Pinchuk, L., Wilson, G.J., Barry, J.J., Schoephoerster, R.T., Parel, J.M. and Kennedy, J.P. (2008), “Medical applications of poly(styrene-block-isobutylene-block-styrene) (“SIBS”), Biomaterials, Vol. 29 No. 4, pp. 448-60. Rajan, G.S., Vu, Y.T., Mark, J.E. and Myers, C.L. (2004), “Thermal and mechanical properties of polypropylene in the thermoplastic elastomeric state”, European Polymer Journal, Vol. 40 No. 1, pp. 63-71. Skoog, A.I., Abramov, I.P., Stoklitsky, A.Y. and Doodnik, M.N. (2002), “The Soviet-Russian space suits: a historical overview of the 1960s”, Acta Astronautica, Vol. 51 Nos 1-9, pp. 113-31. Steller, R., Z˙uchowska, D., Meissner, W., Paukszta, D. and Garbarczyk, J. (2006), “Crystalline structure of polypropylene in blends with thermoplastic elastomers after electron beam irradiation”, Radiation Physics and Chemistry, Vol. 75 No. 2, pp. 259-67. Yigit, A. (1998), “The computer-based human thermal model”, International Communications in Heat and Mass Transfer, Vol. 25 No. 7, pp. 969-77. Corresponding author Kai Yang can be contacted at:
[email protected] To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0955-6222.htm
The preliminary studies of influence of garments on human beings’ corona discharge Izabela L. Ciesielska and Jozef Masajtis The Faculty of Textile Engineering and Marketing, The Institute of Architecture of Textiles, The Technical University of Ło´dz´, Ło´dz´, Poland
Influence of garments on human beings 299 Received 2 February 2008 Accepted 15 May 2008
Abstract Purpose – The purpose of this paper is to analyze the corona discharge films (CDFs) taken from the fingertips of human subjects who had contact for a long period of time with two sets of clothes, in order to establish in what way a long period of contact with textiles influences life’s parameters: the heart beat (HB), the blood pressure (BP), and the volunteers’ level of comfort. Design/methodology/approach – Three volunteers took part in the experiments. They were placing a fingertip in the area of a strong electrical field of high voltage (10 kV) and high frequency (1,024 Hz) to register a CDF. A digital camera placed within the area of corona discharges records this phenomenon. Findings – The paper finds that there is no statistical difference between the parameters of a CDF taken from the fingertips of volunteers after 5 h of wearing two sets of clothes. There is a connection between the level of comfort of the volunteers and their CDF. Originality/value – The CDF shows the consequence of the different factors, impact on human subjects. The authors are moderate in their opinion about the influence of extreme textiles-related feelings. Keywords Clothing, Textile products, Human biology, Individual psychology Paper type Research paper
1. Introduction This experiment focused on the question weather garments that were used by human subject may influence his/her comfort in measurable way. It is already well-known that garments may impact humans, e.g. from an air permeability point of view (Shishoo, 2005; Scott, 2005). Moreover, people’s mood may change after change the garment, e.g. from a tracksuit for an evening-dress some may improve their mood. There are many persons who after coming back from work wear so-called “home clothes.” Wearing them gives a feeling of relief, and they may help in “regeneration” after work. Is it possible to measure such feelings? The aspects mentioned above refer to physiology and psychology of organisms in textiles. The literature review carried out for this research needs points out that the analysis of garments use comfort generally concerns the physiological comfort, and very rare it concerns all aspects of all kinds of comfort. It is probably connected with the lack of the The authors would like to express their gratitude to the Polish State Committee for Scientific Research for support.
International Journal of Clothing Science and Technology Vol. 20 No. 5, 2008 pp. 299-316 q Emerald Group Publishing Limited 0955-6222 DOI 10.1108/09556220810898917
IJCST 20,5
300
interdisciplinary scientific research on an aesthetical, psychological, thermal, and physiological (thermophysiological) comfort. Both, answering questions about what constitutes comfort in general as well as the feelings connected with usage of textiles are complex and difficult. The simplest way of analysing the matter of consumers’ behaviour is to judge garments preferences, and carry out marketing analysis which focus on receiving straight responses of product consumer (Mullen and Johnson, 1990), e.g. does this product fulfil your expectations? A few simple questions like that may not reveal the complexity of the issue concerning fulfilling all the functions by the product with providing a satisfaction to the user, which can be understood as a thermophysiological wear comfort, skin sensorial wear comfort, psychological wear comfort, connected closely with aesthetical comfort, and maybe even social and finally ergonomic wear comfort. A certain way of comfort estimation was created to broaden the analyzed issues. It takes into consideration the analysis of the preferences and the skin sensorial wear comfort (smoothness of the textile material, fabric sensation in different environmental conditions) as well as electrostatic abilities of textiles surfaces, measuring the physiological parameters of user. The most often analyzed parameters are: the temperature and humidity of microclimate between the users’ skin and the garment they wear; the frequency of muscle contraction both while resting and during physical effort; the basic physiological life parameters of textiles’ users like heart beat (HB), BR; and rectal temperature (Scott, 2005). This kind of research develops and helps to better understand the issue of comfort pertaining to different kinds of garments. Let’s analyze a sportswear as an example. The wear comfort of sportswear is an important quality criterion. It affects not only the well-being of the wearers but also their efficiency and performance. If an active sportsperson wears sports clothes with poor air permeability, breathability, rectal temperature, and HB may increase much more rapidly than while wearing breathable sportswear (Umbach, 2001, 2002). As a consequence, the wearers of the breathable clothing outperforms other athletes because they can withstand high activity levels for a longer period of time. Another approach to textiles comfort was proposed by American scientists (Gwosdow et al., 1986). They propose to judge what external conditions allow easier acceptance of chosen textiles. The research focuses on the influence of skin friction by textiles on the total feeling of comfort. It was proven that when the skin wetness and, consequently, the pressure of textiles on skin rise, the subject considers those textiles less pleasant, rougher. It decreases the total acceptance for this textile. Those feelings against the skin were not noted in normal conditions. It was concluded that the higher skin wetness increased the skin friction on textiles and the feeling of roughness. As a result, lack of acceptance for the tested textiles increased. The area where textiles come in contact with the skin as well as the perception and textiles preferences in different external conditions were studied by researchers from Japan (Kim et al., 1995) and Hong Kong (Wong and Li, 2004). The research concerns exposure of a person to cold (Kim et al., 1995). The recent studies showed that the subject exposed to cold dressed faster with thicker clothing in the morning than in the evening. Next, they endeavoured to answer the question what kind of textile was chosen by a volunteer – thin or thick when they feel cold. It was proven that with exposure to the same temperature volunteers were choosing thick garments in the morning and thin in the evening. The researchers pursued their
investigation and came to the conclusion that when the temperature of the environment decreased from the 33 to 258C, the volunteers chose a soft piece of cloth instead of a rough one. It was explained as the behavioural thermoregulation of the organism. The touch of a softer piece of cloth evokes the feeling of warmth. The similar research trend was presented by other authors (Wong and Li, 2004). They analyzed the relationship among human physiological and psychological thermal and moisture responses in tight-fitting aerobic wear. Results show that both physiological and psychological responses are significantly influenced by the period of contact with the garments, the garment themselves, the body location and some of their interactions. A multiple regression was used to investigate the connections between the skin microclimate temperature and the humidity (the physiological reaction) and the sensation of changes of the temperature, the humidity and the overall comfort by the users (psychological reaction). The authors tried to estimate, which of the above-mentioned parameters influence overall comfort. The results show that overall comfort depends mostly on the perception of differences in the temperature. The subjective perception of clothing comfort can be predicted on the basis of human physiological and psychological responses in relation to the temperature and the humidity. The essence of textile garments’ perception and the possibilities of gaining the information about it were also studied with a different approach (Lee and Choi, 2004). Those researches were directed on the analysis of the reaction on two sets of garments: the first – a shirt with short sleeves and trousers with short legs, and the second – a shirt with long sleeves and trousers with long legs. The influence of those sets of garments on the metabolism, thermoregulation, and subjective feelings of users were studied. The expenditure of energy was higher when using the first set of garments in the conditions of 198C. Despite this low temperature, male volunteers still felt comfortable, whereas some female volunteers expressed their discomfort and their HB was unstable when using the first set of garments. The analysis of the experiment results showed that not only the subjective feeling of textiles in certain external conditions plays an important role, but also, according to researchers, the reaction of the organism on adaptation to external conditions. External conditions determine the perception of textiles by the volunteers. It means that the fact of real cooling down of the organism was not always consistent with the feeling of cold. The difference between the real energy expenditure, in case of the first set of garments, in cool external conditions and the sensation of those conditions was observed. Other authors (Li et al., 2005) confirmed that the sensitivity of the human body to the perception of cold varied over sections of the body. The wear trials conducted for their research demonstrate that different locations on the body respond differently to cold stimuli. Another example of that approach is the research (Hatch et al., 1992) that focuses on the response of human skin to fabric. These responses included ecyematous dermatitis, percutaneous absorption of chemicals, hydration, water evaporation, changes in bacterial flora of the skin, blood flow, and neural responses. In general, the presented types of researches acknowledge the difference between what is measurable, e.g. the thermal comfort, and what is impossible to measure, like feelings. The attempts of scientific judgements between measurable and non-measurable aspects give the possibility of broadening the knowledge and discovering the nature of reaction on external stimuli. The presented literature review shows only certain possibilities of carrying out the research, and possibilities that are
Influence of garments on human beings 301
IJCST 20,5
302
seen by scientists in the area of feelings connected with textiles. It was noticed that the fact that an organism reacts to textiles garments, e.g. by intensified HB does not always go with feelings change and energy lost in the volunteer. The declaration of the volunteers who use the textiles does not always correspond with a real reaction of their organisms. How can this divergence be explained? The unilateral scientific approach, e.g. only an analysis of the feelings, or perception, without considering the physiological parameters change will not guarantee a success and will not let for making a step forward the better understanding of the influence of textiles on humans. Moreover, it may lead to faulty conclusions. It is believed that the method able to link both factors would allow for analyzing the interactions between the perception and physical and physiological phenomena, which occur as a result of the contact between humans and textiles. As the majority of human activities aims at improving the existence of human beings, it is necessary to identify the needs of a contemporary textile user and analyze the relations between the textile user and textile products, the experienced feelings while using them, as well as physical and physiological phenomena arising from the closeness of textiles. That is why an innovative approach involving an unconventional method has been proposed to investigate the effects of textiles exerted on their users. This unconventional method, known as corona discharge photography (CDP), registers the phenomena created as the effect of recombination of electrons in the air and sweat secreted from the human body whose part is placed within the high voltage and high frequency electrical field in order to observe this recombination. Electrons that come from a live matter may reflect the state of the matter (Korotkov, 2002; Iovine, 1994). The occurrence of the phenomenon – a discharge around body parts – has been well-known for over two centuries, but it was rendered famous only in the 1950s by Kirlian’s photography, owing to the studies carried out in the Soviet Union by Semyon Dawidowicz Kirlian (1900-1980). The phenomenon itself has already been noticed by Georg Christoph Lichtenberg (1742-1799), a German scientist who discovered in 1777 that electrical discharges, which exist around dust particles, were characterized by highly differentiated shape and nature. Although those lighting shapes around the objects were termed Lichtenberg’s figures, he could not find any application for them in practice. One of the Polish explorers, Jacob Jodko-Narkiewicz (1847-1905), a photographer and a medical doctor who carried out studies on electromagnetism also became famous in the field of discharges. In 1896, he reported his precursory findings on electrical manifestations – the “aura” – as he called it. The analysis of this phenomenon was successfully continued by Kirlian who, while working in a hospital as an electro-technician and repairing a massage device, suddenly noticed electrical discharges between the patient’s skin and the electrode of the repaired device. He managed to develop an apparatus able to register the observed phenomenon. Then, together with his wife, Valentine, he continued his studies on electrical discharges, and in 1949 he obtained a patent for the device able to register electrical discharge photography also called “corona discharge photography” The results of their research and observations were published in the Soviet scientific journal (Kirlian and Kirlian, 1961), where they described the method developed to register the phenomenon of non-electric transformations of living and non-living objects into electrical transformation.
According to the most-up-to-date knowledge, the area of capturing the image of the discharge is being now developed and computerized. Currently CDP, also known as bioelectrophotography, is described as a method used to reveal and record corona discharges created around an object, e.g. human fingertips, in two forms – photography and film. An image can be induced when a strong electrical field of high voltage (10 kV) and high frequency (1,024 Hz) is produced. A digital camera placed within the area of corona discharges records this phenomenon. A corona discharge is an electrical discharge observed in gases, which occurs on the surface of charged conductors. An electrical impulse transmitted to the plate after placing an object on it, stimulates the response of the object, called inter alia, by the movement of charge carriers. Their movement is constantly accelerated in the conditions of a stable ionisation process evoked only by charging a high frequency current (Loeb, 1965; Pehek et al., 1976). The occurrence of ionisation collisions creates an electron avalanche. The ionisation (electron dissociation from atom) in the electrical field results from the collision of neutral atoms (or those previously excited) with free electrons accelerated by the field forces (Opalin´ski, 1979). A free electron re-association is followed by a visible glow. This phenomenon is recorded around the fingertip (Figure 1). Numerous recent studies, involving corona discharges, have provided evidence that discharges generated around non-living objects, e.g. coins or stones do not change their form or shape over time. Corona discharges generated around different liquids (Skarja et al., 1998; Korotkov et al., 2004) as well as around living organisms are characterized by variations over time (Korotkov, 2002; Iovine, 1994; Kirlian
Influence of garments on human beings 303
1 3 5
4
2 7
6 8
9 Notes: (1) study object; (2) optic glass plate connected to a power source; (3) discharge; (4) voltage generator; (5) optic system; (6) CCD camera (charge-coupled device system with charge-coupling, technology involving light-sensitive elements); (7) image transducer; (8) image of discharges on the computer monitor; (9) processed and analysed image of discharges. An object is placed on the optic plate (1), the voltage generator is on (4), while the generated current parameters are 10 kV and 1024 Hz. In the induced electrical field, charge carriers (free electrons and positive or negative ions) (3) present in the air surrounding the surface of optic plate (2) and the object (1) collide with other charge carriers interacting with one another (they attract themselves and mutually neutralize the excess of different charges) Reproduced from the only available original
Figure 1. The system used to visualize CDFs and CDPs along with the processed photography
IJCST 20,5
304
and Kirlian, 1961). This observation suggests that exogenous environmental events, lifestyles, garments and the physical state (also the mental state in humans) of the study objects may influence corona discharges generated around a given animated object. Although the change of garments is often accompanied by the change in mood or comfort (Section 1) those feelings are difficult to define. This kind of research attempts to determine what factors influence people’s feelings towards clothing textiles and, if possible, to estimate them. 2. Methodology Three volunteers (1 man, aged 28 year, and 2 women, aged 28-32 years, mean 30 ^ 2 years) were eligible for the study. All of them were informed about the purpose of the study and the way the experiment was going to be carried out. They were also trained how to place the finger on the plate in the same manner with the same pressure. All the tests were performed in the conditions in the room temperature of 17-258C, the relative humidity of 26-40 percent and the atmospheric pressure of 980-1,012 hPa. On the day of the experiment, they neither smoked nor used medication. The examination methods comprised: a questionnaire-based rating of the respondents’ feeling of comfort, including a 15-item scale with categories of responses ranging from extremely poor to extremely good feeling; HB, blood pressure (BP), measurements of the microclimate (temperature and humidity) between the volunteer skin and the top garment; measurements of the differences in the potential between a volunteer wearing one of the set of garments and the ground; registration of corona discharge film (CDF) from ring fingertips of the right hand. The ring finger of the right hand was placed on the plate of the GDV device and a 32-s-long film of corona discharge around fingertip was recorded. On the first day, a volunteer was wearing a set of garments made of cotton, on the second day, a set of garments made of acrylic, polyester, and/or polyamide. This procedure was repeated ten times in total in order to confirm the results. Altogether, the volunteers were wearing the cotton garment for ten days and the synthetic garment for another ten days. CDFs were recorded with a GDV Camera and film frames parameters were analyzed with the use of the GDV Processor Program. A multiple regression and t-test were used to analyze the effect of the subjects’ examination conditions, their level of comfort, the menstrual cycle, HB, BP on CDF of respondents. The study design was approved by the Bioethics Committee of the Medical University in Łodz, Poland. The reasons why the authors proposed to carry out the study in the described way, involving the ring finger, were as follows: there have been to date no reports on the effect of the long-term contact with textiles on corona discharges created around the human fingertips or other parts of the human body; the shapes of fingertip imprints are ellipse-like and therefore easy to analyze from the mathematical point of view; and there is an easy access to fingers (almost always uncovered). The authors decided to register 32-s-long films from corona discharges around the fingertip of the ring finger of the right hand because this finger seems to be the “most sensitive” according to some neurophysiologists (Schweizer et al., 2000). The time of the effect exerted by textiles on a given person (60 s in this experiment) is an important element – hence the idea of the corona discharge registration around only one finger in order to eliminate confounding factors of the human organism
reaction to textiles. This idea was adopted in the presented study. A rationale was found in the neuroanatomy of the right hand ring finger and the fact that the surface of the finger skin is very rich in sweat glands (consideration of wetness of the skin surface). In addition, having performed the test with von Frey hair, it was proved that the ring finger is more sensitive than others (Schweizer et al., 2000), and the innervation of the ring finger makes it more sensitive than the remaining ones. The decision to use raw materials in the experiment was based on the following prerequisites: a common use of raw materials in garments, the accumulation of electrostatic charges during material friction, the ability of materials to generate electrostatic charges, and the ability of the hand to feel differences between materials.
Influence of garments on human beings 305
3. Results As a result of the performed experiments a number of 32-s-films were produced (Figures 2-19).
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the female aged 28 years wearing: 100 percent cotton garments
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the female aged 28 years wearing: 100 percent cotton garments
Figure 2. Frame no. 1 from the corona discharge film registered around a fingertip of a volunteer
Figure 3. Frame no. 160 from the corona discharge film registered around a fingertip of a volunteer
IJCST 20,5
306 Figure 4. Frame no. 320 from the corona discharge film registered around a fingertip of a volunteer
Figure 5. Frame no. 1 from the corona discharge film registered around a fingertip of a volunteer
Figure 6. Frame no. 160 from the corona discharge film registered around a fingertip of a volunteer
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the female aged 28 years wearing: 100 percent cotton garments
Note: 100 percent acrylic, polyester, polyamide garments
Note: 100 percent acrylic, polyester, polyamide garments
Influence of garments on human beings 307
Note: 100 percent acrylic, polyester, polyamide garments
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the male aged 28 years wearing: 100 percent cotton garments
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the male aged 28 years wearing: 100 percent cotton garments
Figure 7. Frame no. 320 from the corona discharge film registered around a fingertip of a volunteer
Figure 8. Frame no. 1 from the corona discharge film registered around a fingertip of a volunteer
Figure 9. Frame no. 160 from the corona discharge film registered around a fingertip of a volunteer
IJCST 20,5
308 Figure 10. Frame no. 320 from the corona discharge film registered around a fingertip of a volunteer
Figure 11. Frame no. 1 from the corona discharge film registered around a fingertip of a volunteer
Figure 12. Frame no. 160 from the corona discharge film registered around a fingertip of a volunteer
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the male aged 28 years wearing: 100 percent cotton garments
Note: 100 percent acrylic, polyester, polyamide garments
Note: 100 percent acrylic, polyester, polyamide garments
Influence of garments on human beings 309
Note: 100 percent acrylic, polyester, polyamide garments
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the female aged 32 years wearing: 100 percent cotton garments
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the female aged 32 years wearing: 100 percent cotton garments
Figure 13. Frame no. 320 from the corona discharge film registered around a fingertip of a volunteer
Figure 14. Frame no. 1 from the corona discharge film registered around a fingertip of a volunteer
Figure 15. Frame no 160 from the corona discharge film registered around a fingertip of a volunteer
IJCST 20,5
310 Figure 16. Frame no. 320 from the corona discharge film registered around a fingertip of a volunteer
Figure 17. Frame no. 1 from the corona discharge film registered around a fingertip of a volunteer
Figure 18. Frame no. 160 from the corona discharge film registered around a fingertip of a volunteer
Note: The chosen frames from the CDF registered around a fingertip of the ring finger of the right hand of the female aged 32 years wearing: 100 percent cotton garments
Note: 100 percent acrylic, polyester, polyamide garments
Note: 100 percent acrylic, polyester, polyamide garments
Influence of garments on human beings 311
Note: 100 percent acrylic, polyester, polyamide garments
They were analyzed taking into account the following parameters: Area. The number of pixels of glow image, the number of pixels not removed from the picture during the filtration process and glow area – the number of pixels with non-zero intensity. AVI. The average value of intensity for all the pixels from glow image, the scale of glow from 0 – black to 255 – white (maximum intensity of glow) was used. ET. The entropy by the isoline defined as: SðM Þ ¼ 2
j#M X
P j ðM ÞlnðP j ðM ÞÞ;
j¼1
where Pj(M) ¼ Nj/NM denotes the function of allocation of the radius of dots by the isoline of the picture, e.g. probability of value j (Nj – a number of dots with equal value by the isoline) among dots of the isoline length M (NM – number of all dots on the isoline). The variation between values may be very high; that is why the division into even intervals was used. Form C.
The form coefficient – the quotient of the isoline length and the average radius multiplied by 2p, reflects a degree of irregularity of the glow contour.
LENG.
The length of isoline – the length of the curve of the glow contour corresponding to the average intensity.
AVR.
The average radius – the average value of the function R(a) ¼ Rmin 2 Rmax by the angle of inclination of the beam a in [08; 3608], drawn from the glow center to the nearest dot Rmin and the most distant from the center Rmax with the intensity 1 more than the noise level (set on 30, the best for registration the CDFs of fingertips of humans according to Prof. K. Korotkov, DSc).
Figure 19. Frame no. 320 from the corona discharge film registered around a fingertip of a volunteer
IJCST 20,5
312
NRF.
The number of free fragments of which glow area consists.
NRMN.
The normalized deviation of isoline radius – the quotient deviation of the function R(a) to the average radius AVR of the glow.
SD Area.
The standard deviation from the area calculated on the basis of the frames registered during film (1 s ¼ 10 frames).
SD AVI.
The standard deviation from the AVI.
SD Form C. The standard deviation from the Form C. SD LENG.
The standard deviation from the LENG.
SD AVR.
The standard deviation from the AVR.
SD NRF.
The standard deviation from the NRF.
A multiple regression analysis estimated the impact of the level of the volunteers’ comfort on AVR, AVI, NRF of the CDFs and the influence of such factors as the temperature, atmospheric pressure, air humidity on all parameters of CDFs’ recorded during the contact between volunteers of both genders and both sets of garments. The extreme feelings of malaise and well-being increase standard deviation from AVR, AVI (well-being), and the standard deviation from the NRF (malaise). Further analysis was carried out using the t-test. All the gathered data were divided into three groups (each representing one volunteer) according to differences between individuals. The first group consisted of data gathered to analyze the 32-s-films of corona discharges registered in volunteer 1, a 28-year-old woman, wearing the set of garments made of cotton raw material and the set of garments made of polyamide, polyester, acrylic raw materials as well as the life parameters data, the microclimate data, and the data concerning the potential generated on the surface of garments’ top. The t-test did not reveal statistically significant differences between the mean values of parameters from 32-s-films registered when volunteer 1 was wearing garments made of natural and synthetic raw materials. There were no statistically significant differences between HB ( p ¼ 0.50) and BP ( p ¼ 0.60) measured while the volunteer was wearing both sets of garments. There were statistically significant differences in the microclimate temperature measured between the volunteers skin and the top garment made of natural raw materials and synthetic materials ( p ¼ 0.01); the differences in potentials were measured between the volunteer wearing a set of garments made of natural raw materials and a set of garments made of synthetic material ( p ¼ 0.01). The second group comprised the data gathered to analyze the 32-s-films of corona discharges registered from the fingertip of volunteer 2, a 28-year-old man, and the data concerning his life parameters, the microclimate data and the potential measured between the volunteer wearing both sets of garments and the ground. The t-test did not show statistically significant differences between the mean values of parameters from a 32-s-film registered when volunteer 2 was wearing garments made of natural raw materials and synthetic. There were no statistically significant
differences between HB ( p ¼ 0.70) and BP ( p ¼ 0.60) monitored at the time of wearing those sets of garments. There were statistically significant differences in the microclimate temperatures measured between the volunteer’s skin and the top garment made of natural raw materials and synthetic ( p ¼ 0.01). There was also a statistically significant difference between the potential generated on the surface of the sets of garments made of cotton and synthetic ( p ¼ 0.0007). The third group embraced the data gathered to analyze a 32-s-film of corona discharge registered from the fingertip of volunteer 3, a 32-year-old woman, and the data on her life parameters, the microclimate data and the potential measured between the volunteer wearing sets of garments and the ground. The t-test did not indicate statistically significant differences between the mean values of parameters from a 32-s-film registered when volunteer 3 was wearing garments made of natural raw materials and synthetic. There were statistically significant differences between HB ( p ¼ 0.02), BP ( p ¼ 0.0460), and microclimate temperatures ( p ¼ 0.04) monitored during wearing these two sets of garments. There was also a difference between the potential generated on the surface of the sets of garments made of cotton and synthetic ( p ¼ 0.450). The difference in potentials between the ground and the volunteer wearing the set of garments made of synthetics raw materials was higher than those between the ground and the volunteer wearing the set of garments made of cotton raw materials. This results from the tri-biological property of garment materials. This observation applies to all the three volunteers. The correlation (r ¼ 2 0.20) was observed between the level of comfort felt by all volunteers and the difference in potentials between the ground and the volunteer wearing sets of garments. The higher difference in the potentials between the ground and the volunteer the worse the well-being among volunteers. An electrization-related discomfort could influence the self-rating reported in the questionnaire. In such a case the authors could assume the effect of textiles on the mood of individual persons. The influence of textiles on physiological parameters of the subjects is another aspect of the discussed issue. The type of garment sets influenced the level of HB and BP in two of the three volunteers. The highest values of these parameters were observed while the subject was wearing the set of the garments made of synthetic raw materials. These parameters may also have affected the mood of the subjects. It should be emphasized that electrization is not the only factor contributing to the well-being or aggravation, the thermal discomfort caused by synthetic raw materials (microclimate temperature) should also be taken into consideration. There were no statistically significant differences between the mean value of CDF parameters registered while the subjects were wearing the set of garments made of natural raw material like cotton, or synthetic raw material such as polyamide, polyester, and acryl. 4. Discussion This study confirmed the effect of such factors as temperature, atmospheric pressure, air humidity, and, partly, HB and BP on CDF parameters, which has also been reported by other authors (Ciesielska, 2007; Korotkov, 2002; Iovine, 1994; Kirlian and Kirlian, 1961; Loeb, 1965; Pehek et al., 1976; Opalin´ski, 1979; Szosland, 2003). Moreover, we noticed that the set of garments influences only the microclimate temperature
Influence of garments on human beings 313
IJCST 20,5
314
generated between the volunteer’s skin and the worn garments, while the microclimate humidity was not affected. The experiments did not reveal differences in CDF parameters dependent on the sets of garments during a long-term contact in the described conditions and without intensive physical activity. This means that a long-term use of textiles has no impact on CDF parameters that could change them, depending on the set of textiles. This can be associated with the adaptation of the human body to a given textile set. However, one must bear in mind that those parameters were registered after a 5-h use of the set. The authors assume that the textiles themselves give a weak impulse that vanishes over time due to the accommodation process. It is also possible that they provoke a sensors reaction only at the beginning of the contact and that it diminishes after certain period of time depending on the set of garments, so that it is not possible to register this reaction by the means of CDF. Additionally, the adaptation of receptors to the effect of an external factor depends on a certain condition: if the stimulus evokes a constant stress, receptors stop to signal it (pain is the only exception). The adaptation process is characteristic of the skin receptors, which depends on the decrease in the generation potential over the time. The ability to detect tactile stimuli is not essential to human survival. It is difficult to estimate the adaptation time to an external impulse, such as textiles. The volunteer’s body probably got used to the textiles to such an extent that it was not visible on CDF. These findings must be regarded as preliminary ones on account of the small number of volunteers participating in the experiment. However, the analysis of the results allows the conclusion that the level of both extremely high and extremely low comfort felt by the respondents influenced their CDF. The most likely reason for this observation is the emotion-related intense perspiration of hands. It is well documented that sweat (and liquids in general) as well as air humidity exert an effect on the discharge (Korotkov, 2002; Iovine, 1994; Skarja et al., 1998; Korotkov et al., 2004; Naidu and Kamaraju, 1996). 5. Conclusions The interpretation of the results of this study is difficult. It is one of the first reports on the effect of textiles on the CDF variability in humans. The CDP and CDF seem to be a source of information about the effect of life parameters of the human body. The authors are moderate in their opinion about the influence of extreme textiles-related feelings. The analysis shows that the level of comfort felt by volunteers as well as the factors, such as temperature, atmospheric pressure, and air humidity influence all CDF parameters recorded during the contact of the volunteers (both genders) with both sets of garments. The extreme feelings of malaise or well-being increase the standard deviation of the fallowing parameters: AVR, AVI (well-being), and NRF (malaise). There are statistically significant differences in microclimate temperature measured between the skin of the volunteer and top garments made of natural raw materials and synthetic raw material, but no statistically significant differences were noted between the mean value of CDF parameters registered after 5 h of wearing the set of garments made of natural raw materials and synthetic raw materials. No statistically significant differences between the HB and BP levels after this period of time in two of the three volunteers wearing sets of garments were found either.
References Ciesielska, I.L. (2007), “The analysis of the influence of textiles on corona discharges”, doctoral dissertation, The Technical University of Ło´dz´, Ło´dz´. Gwosdow, A.R., Stevens, J.C., Berlund, L.G. and Stolwijk, J.A.J. (1986), “Skin friction and fabric sensation in neutral and warm environments”, Textile Research Journal, Vol. 56 No. 9, pp. 574-9. Hatch, K.L., Markee, N.L. and Maibach, H.I. (1992), “Skin response to fabric – a review of studies and assessment methods”, Clothing and Textiles Research Journal, Vol. 10 No. 4, p. 54. Iovine, J. (1994), Kirlian Photography, A Hands-on Guide, TAB Books, Blue Ridge Summit, PA. Kim, H.E., Tokura, H., Nagashima, R. and Nishizawa, K. (1995), “Do cooled women select soft or hard clothing?”, Journal of Thermal Biology, Vol. 20 No. 4, pp. 327-32. Kirlian, S.D. and Kirlian, W.Ch. (1961), “Fotografirovanie i wizualnoje nabludenie pri posredstwie took wysokiej czastoty”, Z˙urnal naucznoj i prokładnoj fotografii i kinematografii, Vol. 6, pp. 397-403. Korotkov, K. (2002), Human Energy Field, Backbone, New York, NY. Korotkov, K., Krizhanovsky, E., Borisova, M., Korotkin, D., Hayes, M., Matravers, P., Momoth, K.S., Peterson, P., Shiozawa, K. and Vainshelboim, A. (2004), “Time dynamics of the gas discharge around drops of liquids”, Journal of Applied Physics, Vol. 95 No. 7. Lee, J.Y. and Choi, J.W. (2004), “Influences of clothing types on metabolic, thermal and subjective responses in a cool environment”, Journal of Thermal Biology, Vol. 29, pp. 221-9. Li, J., Barker, R.L., Wang, Y. and Zhang, W. (2005), “Cold sensitivity differences between body sections under clothing”, Textile Research Journal, Vol. 75 No. 3, pp. 208-12. Loeb, L.B. (1965), Electrical Coronas, University of California Press, Berkeley, CA. Mullen, B. and Johnson, G. (1990), The Psychology of Consumers’ Behavior, Lawrance Erlbaum Associates, Hillsdale, NJ. Naidu, M.S. and Kamaraju, V. (1996), High Voltage Engineering, 2nd ed., McGraw-Hill, New York, NY. Opalin´ski, J. (1979), “Kirlian – type images and the transport of thin-film materials in highvoltage corona discharges”, Journal of Applied Physics, Vol. 50 No. 1, pp. 498-504. Pehek, J., Kyler, H. and Faust, D. (1976), “Image modulation in corona discharge photography”, Science, Vol. 194, p. 4262. Schweizer, R., Maier, M., Braun, C. and Birbuumer, N. (2000), “Distribution of mislocalizations of tactile stimuli on the fingers of the human hand”, Somatosensory and Motor Research, Vol. 17 No. 4, pp. 309-16. Scott, R.A. (2005), Textiles for Protection, The Textile Institute, Woodhead Publishing Limited, Cambridge. Shishoo, R. (2005), Textiles for Sport, The Textile Institute, Woodhead Publishing Limited, Cambridge. Skarja, M., Berden, M. and Jerman, I. (1998), “Influence of ionic composition of water on corona discharge around water drops”, Journal of Applied Physics, Vol. 84 No. 5. Szosland, J. (2003), “The corona discharges in researches of the system human being – garments. Part 1”, Przegla˛d Wło´kienniczy, Vol. 12. Umbach, K.H. (2001), “Optimization of the wear comfort by suitable fibre, yarn and textile construction”, paper presented at the 40th International Man-Made Fibres Congress, Dornbirn, Austria.
Influence of garments on human beings 315
IJCST 20,5
Umbach, K.H. (2002), “Measurement and evaluation of the physiological function of textiles and garments”, paper presented at 1st Joint Conference Visions of the Textile and Fashion Industry, Seoul, South Korea. Wong, A.S.W. and Li, Y. (2004), “Relationship between thermophysiological responses and psychological thermal perception during exercise wearing aerobic wear”, Journal of Thermal Biology, Vol. 29, pp. 791-6.
316 Corresponding author Izabela L. Ciesielska can be contacted at:
[email protected]
To purchase reprints of this article please e-mail:
[email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints
International Journal of Clothing Science and Technology
ISSN 0955-6222 Volume 20 Number 6 2008
International textile and clothing research register Editor-in-Chief
Professor George K. Stylios
Access this journal online _________________________
2
Editorial advisory board __________________________
3
Editorial _________________________________________
4
Research register _________________________________
6
Research index by institution______________________
112
Research index by country ________________________
113
Research index by subject_________________________
114
Research index by principal investigator ___________
116
Access this journal electronically The current and past volumes of this journal are available at:
www.emeraldinsight.com/0955-6222.htm You can search more than 175 Emerald journals in Emerald Management Xtra (www.emeraldinsight.com) See page following contents for full details of what your access includes.
CONTENTS
www.emeraldinsight.com/ijcst.htm As a subscriber to this journal, you can benefit from instant, electronic access to this title via Emerald Management Xtra. Your access includes a variety of features that increase the value of your journal subscription.
How to access this journal electronically
To benefit from electronic access to this journal, please contact
[email protected] A set of login details will then be provided to you. Should you wish to access via IP, please provide these details in your e-mail. Once registration is completed, your institution will have instant access to all articles through the journal's Table of Contents page at www.emeraldinsight.com/0955-6222.htm More information about the journal is also available at www.emeraldinsight.com/ ijcst.htm Our liberal institution-wide licence allows everyone within your institution to access your journal electronically, making your subscription more cost-effective. Our web site has been designed to provide you with a comprehensive, simple system that needs only minimum administration. Access is available via IP authentication or username and password. Emerald online training services Visit www.emeraldinsight.com/training and take an Emerald online tour to help you get the most from your subscription.
Key features of Emerald electronic journals Automatic permission to make up to 25 copies of individual articles This facility can be used for training purposes, course notes, seminars etc. This only applies to articles of which Emerald owns copyright. For further details visit www.emeraldinsight.com/ copyright Online publishing and archiving As well as current volumes of the journal, you can also gain access to past volumes on the internet via Emerald Management Xtra. You can browse or search these databases for relevant articles. Non-article content Material in our journals such as product information, industry trends, company news, conferences, etc. is available online and can be accessed by users. Reference linking Direct links from the journal article references to abstracts of the most influential articles cited. Where possible, this link is to the full text of the article. E-mail an article Allows users to e-mail links to relevant and interesting articles to another computer for later use, reference or printing purposes. Structured abstracts Emerald structured abstracts provide consistent, clear and informative summaries of the content of the articles, allowing faster evaluation of papers.
Additional complimentary services available
Your access includes a variety of features that add to the functionality and value of your journal subscription: Xtra resources and collections When you register your journal subscription online you will gain access to additional resources for Authors and Librarians, offering key information and support to subscribers. In addition, our dedicated Research, Teaching and Learning Zones provide specialist ``How to guides'', case studies and interviews and you can also access Emerald Collections, including book reviews, management interviews and key readings. E-mail alert services These services allow you to be kept up to date with the latest additions to the journal via e-mail, as soon as new material enters the database. Further information about the services available can be found at www.emeraldinsight.com/alerts Emerald Research Connections An online meeting place for the world-wide research community, offering an opportunity for researchers to present their own work and find others to participate in future projects, or simply share ideas. Register yourself or search our database of researchers at www.emeraldinsight.com/connections
Choice of access
Electronic access to this journal is available via a number of channels. Our web site www.emeraldinsight.com is the recommended means of electronic access, as it provides fully searchable and value added access to the complete content of the journal. However, you can also access and search the article content of this journal through the following journal delivery services: EBSCOHost Electronic Journals Service ejournals.ebsco.com Informatics J-Gate www.j-gate.informindia.co.in Ingenta www.ingenta.com Minerva Electronic Online Services www.minerva.at OCLC FirstSearch www.oclc.org/firstsearch SilverLinker www.ovid.com SwetsWise www.swetswise.com
Emerald Customer Support
For customer support and technical help contact: E-mail
[email protected] Web www.emeraldinsight.com/customercharter Tel +44 (0) 1274 785278 Fax +44 (0) 1274 785201
EDITORIAL ADVISORY BOARD Professor Mario De Araujo Minho University, Portugal Professor H.J. Barndt Philadelphia College of Textiles & Science, Philadelphia, USA Professor Dexiu Fan China Textile University, Shanghai, China Professor Carl A. Lawrence University of Leeds, Leeds, UK Professor Gerald A.V. Leaf Heriot-Watt University (Hon), UK Professor P. Grosberg Shankar College of Textile Technology and Fashion, Israel Professor Trevor J. Little North Carolina State University, USA
Professor David Lloyd University of Bradford, Bradford, UK Professor Masako Niwa Nara Women’s University, Nara, Japan Professor Issac Porat School of Textiles, UMIST, UK
Editorial advisory board
3
Professor Ron Postle The University of New South Wales, Australia Professor Rosham Shishoo Swedish Institute for Fibre and Polymer Research, Mo¨lndal, Sweden Professor Paul Taylor University of Newcastle, Newcastle upon Tyne, UK Professor Witold Zurek Ło´dz´ Technical University, Poland
International Journal of Clothing Science and Technology Vol. 20 No. 6, 2008 p. 3 # Emerald Group Publishing Limited 0955-6222
IJCST 20,6
4
International Journal of Clothing Science and Technology Vol. 20 No. 6, 2008 pp. 4-5 Emerald Group Publishing Limited 0955-6222
Editorial The International Textiles and Clothing Research Register championing the research efforts of the community The International Textile and Clothing Research Register (ITCRR) is now in its 14th year of publishing the research efforts of our community. You can see the breadth of activity, the large funding involved, the hot topics and the research players in the field of textile and clothing research. By so doing, ITCRR provides a platform of participation and dissemination to those working in our discipline and avoids duplication of effort. Again, as you will see in this new edition, textile and clothing research and practice is increasing in volume, in quality and in diversity, all good news for all of us involved in the field. We try to capture as many projects as possible, but I understand that there maybe projects not having been registered. I will welcome them in our next issue and I invite them to send me their project details anytime during the year for the next ITCRR issue. Research, development and innovation can, without doubt, give us more wisdom, enable our industries to become more competitive, and contribute to our quality of life. I believe that registering research projects will provide the due credit to originators of the research and contribute to the future development of this field. Groups of expertise can be identified in this manner, repetition and re-invention can be avoided leading to best utilisation of time and funding for faster and better directed research. Textiles and clothing originate from the physiological need to protect ourselves from the environment. New challenges however are already upon us with nanotextiles; nanofibres nanocoatings, with multifunctional and smart textiles and clothing, and with wearable electronics. In this year’s Issue 5 of the IJCST we published a special issue on smart textiles and clothing. We continue to welcome contributions from textile and clothing aesthetics, design and fashion highlighting our belief that design and technology go hand to hand, and we are working with a special issue in Textile and Clothing Design and Technology Interface. IJCST was set up 20 years ago as a platform for the promotion of scientific and technical research at an international level. Our original statement that the manufacture of clothing needs to change to more technologically advanced forms of production and retailing, still stands, IJCST has however evolved further by also providing opportunities in the new research areas of Nanotextiles, SMART textiles and clothing and in wearable electronics. The journal, now fully indexed in SCI continues with its authoritative style to accredit original technical research, adhering to our refereeing processes, however difficult these may prove at times. IJCST will be instrumental in continuing to support conferences and meetings from around the world in its effort to promote the science and technology of clothing. I would like to thank the our research community and those authors in particular who have contributed to this volume, our editorial board for their continuous support and our colleagues who have acted in a refereeing capacity and have given us their free q Professor George K. Stylios
time and expertise to progress our research efforts. I take the liberty to list some of those names below (apologies in advance if anyone has accidentally been omitted from this list): . Professor Paul Taylor, University of Newcastle . Professor Isaac Porat, UMIST . Professor R.H. Wardman, Heriot-Watt University . Professor R. Christie, Heriot-Watt University . Professor Ron Postle, The University of New South Wales . Dr Taoruan Wan, University of Bradford . Professor David Lloyd, University of Bradford . Professor G.A.V. Leaf, Heriot-Watt University . Dr David Brook, University of Leeds . Dr Jaffer Amirbayat, UMIST . Dr David Tyler, Manchester Metropolitan University . Professor Jintu Fan, Hong Kong Polytechnic University . Dr Jelka Gersak, University of Maribor . Dr Hua Lin, Nottingham University . Dr Sharon Lam Po Tang, Heriot-Watt University . Dr Lisa McIntyre, Heriot-Watt University . Dr T. Wan, Heriot-Watt University Correspondence address: Heriot-Watt University, School of Textiles, Netherdale, Galashiels, Selkirkshire TD1 3HF, Scotland, UK. E-mail:
[email protected];
[email protected] George K. Stylios Editor-in-Chief
Editorial
5
Research register
IJCST 20,6
Athens, Greece 6
TEI Piraeus, P. Ralli and Thivon 250, GR-12244 Athens, Greece Tel: +30 210 5381224; Fax: +30 210 5450965; E-mail:
[email protected] Principal investigator(s): Lect. Savvas Vassiliadis, Prof. Christos Kotsios, Prof. Anthony Primentas Research staff: Prof. Thanos Peppas, Argyro Kallivretaki
Strengthening the university-industry links in Uzbekistan Other Partners: Academic
Industrial
TechMinho (P), Logotech S.A. (GR) Tashkent Institute of Textile and Light Industry (UZ), Namangan Engineering and Economic Institute (UZ) Project started: 1 September 2005 Project end date: 31 August 2008 Project budget: e300,000 Source of support: European Commission Keywords: Industrial liaison office, Distance learning During the three years long project the situation of the university-industry links in Uzbekistan will be analysed. In parallel the need of the distance learning will be investigated taking in acount the local conditions. After the first definition phase, the project foresees the development of structures serving the university-industry links, like the liaison offices in the Universities and related actions. The distance learning activities will result in the operation of the necessary infrastructure and the creation of many modules for use between the universities and the industries.
Project aims and objectives The project aim is to analyse the current conditions in Uzbekistan and to develop structures and materials related to the strengthening of the university-industry links and distance learing courses. The objective is to enhance the interaction between universiteis and industries using modern technological tools and methods.
Research deliverables (academic and industrial) Establishment of structures supporting the university-industry links in Uzbekistan and development of distance learning courses. Publications Not available.
Athens, Greece National Technical Univesity of Athens, Iroon Polytechniou 9, Zografos, GR-15773 Athens, Greece Tel: +30 210 7721520; Fax: +30 210 7722347; E-mail:
[email protected] Principal investigator(s): Christopher Provatidis Research staff: Argyro Kallivretaki
Computational simulation of the mechanical performance of textile products Other Partners: Academic
Industrial
None None Project started: 5 November 2007 Project end date: 4 November 2009 Project budget: e15,000 Source of support: Committee for the Basic Research of the National Technical University of Athens Keywords: Micromechanical, Macromechanical, Finite element method, Simulation The current research project focuses in the mechanical analysis of textile structures implementing the Finite Element Method (FEM). Three stages of analysis are adopted for the integrated modelling of a textile structure. The micromechanical analysis of yarns is implemented for the evaluation of the yarn apparent properties in the first stage. The second stage of modelling corresponds to the mesomechanical modelling of the textile structure for the calculation of the apparent properties of the unit cell. The third stage supports the simulation of complex deformations (drape test) of the macromechanical models of the textile structures.
Project aims and objectives The research project aims at the development of a reliable computational method for the evaluation of the mechanical performance of a textile structure considering the fibres’ properties, the yarn and the fabric microstructure.
Research deliverables (academic and industrial) Computational tools for: .
The parametric modelling of a typical yarn based on the structural characteristics of the fibre assembly (fibre elastic properties, number of fibres, fibre diameter, yarn diameter).
.
The parametric mesoscale modelling for a series of textile structures (basically woven and knitted structures).
Research register
7
IJCST 20,6
.
The simulation of micro- meso- and macro- scale models in the appropriate deformations.
Comparison with experimental data.
8
Publications Kallivretaki, A., Vassiliadis S. and Provatidis, C. (2008), “Computational modelling of fibrous assemblies”, CIRAT 3 International Conference of Applied Research In Textiles, November 2008, Sousse, Tunisia. Vassiliadis, S., Kallivretaki, A., Grancaric, A.M., Giannakis S. and Provatidis, C. (2008), “Computational modelling of twill and satin woven structures”, 8th Autex World Textile Conference, June 2008, Biella, Italy. Vassiliadis, S., Kallivretaki, A., Kavagia, X., Provatidis, C., Mecit D. and Roye, A. (2008), “Computational modelling of spacer fabrics”, 8th Autex World Textile Conference, June 2008, Biella, Italy.
Bolton, UK University of Bolton, Centre for Materials Research and Innovation, University of Bolton, Deane Road, Bolton BL3 5AB, UK Tel: +44 1204 903559; Fax: +44 1204 399074; E-mail:
[email protected] Principal investigator(s): Dr S. Rajendran, Prof. S.C. Anand, Co-investigator Research staff: Dr Alister Rigby
Design and development of novel compression therapy regimes for the treatment of venous leg ulcers Other Partners: Academic None
Industrial Vernon-Carus Ltd, Rossendale Combining Company Ltd Project end date: 23 October 2008
Project started: 24 October 2005 Project budget: £152,142 Source of support: EPSRC Keywords: Compression therapy, Leg ulcers, Bandages
Venous leg ulcers are the most common type of ulcers and their prevalence increases with age. In the UK alone about 1 per cent of the adult population suffers from active ulceration during their life time. The total cost to the National Health Service in the UK for venous leg ulcers treatment is about £650 million per annum, which is 1-2 per cent of the total healthcare expenditure. Costs per patient have recently been estimated to be between £1,200 and £1,400. Venous leg ulcers are chronic and there is no medication or surgery to cure the disease other than the compression therapy. A sustained graduated compression mainly enhances the flow of blood back to the heart, improves the functioning of valves and calf muscle pumps, reduces oedema and prevents the swelling of veins. In the UK four layer bandaging system is widely used whilst in Europe and
Australia the non-elastic two layer short stretch bandage regime is the standard treatment. Both the two layer and four layer systems require padding bandage that is applied next to the skin and underneath the short stretch or compression bandages. It is generally agreed by the clinicians that four layer bandages are too bulky for patients and the cost involved is high. A wide range of compression bandages is available in the Drug Tariff but each of them has different structure and properties and this influences the variation in performance and properties of bandages. The research carried out at the University of Bolton showed that there are significant variations in properties of commercial padding bandages, more importantly the commercial bandages did not distribute the pressure evenly at the ankle as well as the calf region. When pressure is applied using compression bandages, the structure of the nonwoven padding bandages collapsed and the bandages could not impart cushioning effect to the limb. In view of the above mentioned limitations and problems, it is vital that research and development work should be carried out to design, develop and characterise novel single layer bandages that would effectively fulfil the requirements of both padding and compression bandages. It is recognised that spacer is the right technology to produce novel compression bandages that meet the prerequisites of both ideal padding and compression bandages. In three-dimensional (3D) spacer fabrics, two separate fabric layers are combined with an inner spacer yarn or yarns using either warp knitting or weft knitting route. It is possible to produce low modulus spacer fabrics by making use of elastic yarns. Elastic compression could be achieved by altering the structures. It should be mentioned that 3D structure allows greater control over elasticity and these structures can be engineered to be uni-directional, bi-directional and multi-directional. Uni-directional elasticity is one of the desired properties for compression bandages. The three-dimensional nature of spacer fabrics makes an ideal application next to the skin because they have desirable properties that are ideal for the human body. 3D fabrics are soft, have good resilience that provides cushioning effect to the body, breathable, ability to control heat and moisture transfer. For venous leg ulcer applications, such attributes together with improved elasticity and recovery promote faster healing.
Project aims and objectives .
To study in-depth the current practices and problems in managing venous leg ulcers.
.
To test and characterise the currently available commercial bandaging systems.
.
To design and develop single layer bandage that would replace the conventional multiplayer padding and compression bandages using warp knitting technology.
.
To design and develop single layer bandage that would replace the conventional multiplayer padding and compression bandages using weft knitting technology.
.
To study the feasibility of using environmentally and human skin friendly biodegradable fibres in designing the single layer bandage.
.
To test and characterise the properties of novel compression bandages and bandaging regimes.
.
To mathematically model and verify the performance and properties of the developed structures.
Research register
9
IJCST 20,6
.
To quantify, predict and optimise the characteristics of the novel compression therapy regimes.
Research deliverables (academic and industrial) None
10
Publications “Venous Leg Ulcer Treatment and Practice”, submitted to Journal of Wound Care.
Bornova/Izmir, Turkey Ege University, Ege Universitesi, Muhendislik Fak. Tekstil Muh. Bol. Bornova Izmir Turkey. Tel: 90 232 339 92 22; Fax: 90 232 339 92 22; E-mail:
[email protected] Principal investigator(s): Kerim Duran Research staff: Ays¸ egu¨l Ekmekci Ko¨rlu¨, Seher Perincek and M. ˙Ibrahim Bahtiyari
The use of ultrasound UV for oxidative treatment of textile materials, for the acceleration of processes and treatment of textile wastewaters Other Partners: Academic
Industrial
None Alenka Majcen Le Marechal, University of Maribor, Faculty of Mechanical Engineering, Textile Department Project started: 1 December 2006 Project end date: 1 December 2009 Project budget: 41,035 Eur (78,296 YTL) Source of support: TUBITAK Keywords: Ultrasound, UV, Pretreatment, Finishing, Ozone The objectives of the proposed project are the wastewater treatment with AOP processes where full decolouration, maximum reduction of surfactants and toxic compounds, maximum recycling of wastewaters, strong reduction of process sludges and extraction of salts will be achieved by combining of different AOP treatment technologies (US, UV, H2O2 and combinations). The use of experimental design and artificial inteligence should be included for the optimization of the wastewater treatment processes (minimum consumption of the energy, water, chemicals, direct toxicity assessment, . . .). So, the second part of the project will focus on the decolorization and cleaning up treatment of wastewaters resulting from the first part of the project. The aim is to use the same AOP processes for the wastewater treatment as for finishing processes (UV/H2O2, thermal/H2O2, US/H2O2 or O3). In the first step we will study the degradation and decolorization of model wastewater. We’ll study the influence of different operating
conditions for each AOP, such as: temperature, intensivity of UV irradiation, the amount of hydrogen peroxide added, treatment duration, initial concentration of the dye, concentration of other chemicals needed for the dyeing process itself, pH, frequency of the sound wave and power. In the second step we would like to degrade or decolorize the wastewater from the Turkish and Slovene textile factories. Enzymatic treatments supplemented with ultrasonic energy results in shorter processing times, less consumption of expensive enzymes, less fiber damage and better uniformity of treatment to the textiles. The effect of ultrasound power is an important technique increasing the mass transfer towards to the textile material. In the ozonation process, it is expected that there will be an increase in the whiteness degree of fabrics. Because of the combined effects of ultraviolet light in the presence of hydrogen peroxide, the hydroxyl radicals, much more active then the conventional oxidizing agents will be generated. The textile finishing wastewater treatment is expected to achieve the following characteristics: .
Maximal decolorization of the wastewater treated (up to 99 per cent).
.
Maximal reduction of environmental parameters according to Slovenian and Turkish environmental legislations (TOC, COD, BOD5, toxicity).
.
Reduction of surfactants and toxic compounds (more than 95 per cent).
The combination of different AOP wastewater treatment processes is expected to achieve a complete decolorization of the process waters for every type of wet process (finishing, bleaching, and dyeing).
Project aims and objectives In this study, it has been aimed to investigate the effects of ultrasound, UV and ozone on textile pretreatment processes. Conventional methods are compared with alternative methods which are above. The objectives of the proposed project are the wastewater treatment with AOP processes where full decolouration, maximum reduction of surfactants and toxic compounds, maximum recycling of wastewaters, strong reduction of process sludges and extraction of salts will be achieved by combining of different AOP treatment technologies (US, UV, H2O2 and combinations). The use of experimental design and artificial inteligence should be included for the optimization of the wastewater treatment processes (minimum consumption of the energy, water, chemicals, direct toxicity assessment, . . .). So, the second part of the project will focus on the decolouring and cleaning up treatment of wastewaters resulting from the first part of the project. The aim is to use the same AOP processes for the wastewater treatment as for finishing processes (UV/H2O2, thermal/H2O2, US/H2O2 or O3). In the first step we will study the degradation and decolouration of model wastewater. We’ll study the influence of different operating conditions for each AOP, such as: temperature, intensivity of UV irradiation, the amount of hydrogen peroxide added, treatment duration, initial concentration of the dye, concentration of other chemicals needed for the dyeing process itself, pH, frequency of the sound wave and power. In the second step we would like to degrade or decolorize the wastewater from the Turkish and Slovene textile factories.
Research register
11
IJCST 20,6
12
Research deliverables (academic and industrial) Experiments are current and in progress. Publications Presentation in Autex 2008. World Textile Conference, June 24-26, 2008, Biella, Italy. Simona Vajnhandl, Alenka Majcen Le Marechal, Darinka Fakin, Kerim Duran, Aysegul E. Korlu, M. ˙Ibrahim Bahtiyari and Seher Perincek. The use of ultrasound in pretreatment and dyeing processes and wastewater treatment.
Bornova/Izmir, Turkey Ege University, Ege Universitesi, Muhendislik Fak. Tekstil Muh. Bol. Bornova Izmir Turkey. Tel: 90 232 339 92 22; Fax: 90 232 339 92 22; E-mail:
[email protected] Principal investigator(s): Ays¸ egu¨l Ko¨rlu¨ Research staff: Kerim Duran, Seher Perincek and M. ˙Ibrahim Bahtiyari
Enzymatic finishing and the effects on cellulosic textile materials Other Partners: Academic
Industrial
Alina Popescu, 16 Lucretiu Patrascanu None Street, sector 3, Bucharest, 030508, Romania, The Research Development National Institute for Textile and Leather Project started: 1 March 2008 Project end date: 01 September 2009 Project budget: 10,487Eur (20,000 YTL) Source of support: TUBITAK Keywords: Enzyme, Textile, Finishing, Cellulose Modern biotechnology often utilizes genetically modified micro-organisms which are characterized by enhanced productivity of specific metabolites, enzymes among them. The range of novel applications of cellulolytic enzymes includes the modification of fibres and fabrics from regenerated cellulose, in order to improve their utility properties and wearing comfort, as well as to maintain their mechanical parameters. In the enzymatic modification of fibres and fabrics, the quantitative and qualitative composition of cellulolytic complex, and so the type of micro-organism being produced, plays the key role. This kind of complex, which is a tailored mixture of specific enzymes, may serve as the ideal solution for creating effective, economical and technologically simple methods for modifying cellulosic fibres. Enzymes have been used for over fifty years to remove starch-based sizes in the textile industry. During the last decade, cellulolytic enzymes have replaced the
traditional stone-washing of denim garments and found applications in finishing fabrics and clothing from cotton, linen and regenerated cellulose. In the modern textile industry, finishing processes which are based on biodegradable and environment-friendly enzymes can fully substitute for a wide range of chemical and mechanical operations so far used to improve the quality of textile products, and save on the energy and consumption of chemicals. Within the context of the project, especially cotton fabrics will be treated enzymatically. The aim of project is environmentally textile finishing and the effects of conventional and bio finishing processes on textile material to be compared with each other. During the experiments laccase enzyme, this is not as common as cellulase in textile finishing, will be used. Turkish side will apply enzymatic finishing to fabric and compare with conventional methods. The properties of washed fabrics will be investigated. Romanian partner will elaborate: (i) schemes and plans for enzyme products treating of textile materials of cellulose fibers, in successive or cumulated phases, as alternatives to the classical processing treatments, (ii) experiments in lab and pilot scale on 100% cotton fabrics, (iii) establishing the influence of enzymatic treatments over treated textile materials, and (iv) their efficiency through physical-mechanical, physical-chemical analyses and color measures.
Project aims and objectives Especially cotton fabrics will be treated enzymatically. The aim of project is environmentally textile finishing and the effects of conventional and bio finishing processes on textile material to be compared with each other.
Research deliverables (academic and industrial) Experiments are current and in progress. Publications Not available.
Bouca/I˙zmir, Turkey Dokuz Eylul University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90 232 4127211; Fax: +90 232 4127210; E-mail:
[email protected] Principal investigator(s): Prof. Dr Nilufer Erdem Research staff: Asisst Prof. Dr Aysun Cireli & Research Assistant U. Halis Erdogan
Investigation of production probabilities of chemical fibers having different characteristics in laboratory conditions
Research register
13
IJCST 20,6
14
Other Partners: Academic
Industrial
None None Project started: 1 January 2005 Project end date: 31 December 2008 Project budget: e90,000 Source of support: TUBITAK (The Scientific and Tecnological Research Council of Turkey) Keywords: Chemical fibers, Polymers, Fiber spinning Today Turkey is one of the leading chemical fiber producer in the world with it’s approximately 1.100.000 metric ton chemical fiber production per year. At first chemical fibers were produced to meet increasing fiber demand, because the production of natural fibers was not enough for demands in an increasing world population. However, nowadays with the developments in polymer technology and fiber spinning technologies chemical fibers gather additional physical and chemical properties. Conventional chemical fibers have round cross-sections. Recently, fibers with hollow, tri-lobal, delta and various types of cross-sections are produced to advance chemical fibers. These fibers help to increase the fabric appearance, tactile and performance (such as strength, air and heat transfer) properties. New fibers play an important role either gathering additional properties to traditional fabric types or designing different textile structures. So the application areas of chemical fibers increases rapidly, for example chemical fibers are now widely used in the field of military textiles, sport textiles, medical textiles; aerospace technology, etc. Keeping these conditions in mind, our country has to follow recent developments like other chemical fiber producer countries. It is necessary to make additional researches in both conventional chemical fiber production and production of new fibers with advanced properties.
Project aims and objectives .
To produce either conventional chemical fibers with round cross section or new fibers with hollow, tri-lobal, delta and various types of cross-sections in the BCF/FDY form and investigate properties of these fibers.
.
To observe and evaluate the effect of various production parameters (viscosities, drawing velocities, type of douse etc) and conditions (temperatures etc) in fiber spinning process to fiber characteristics such as strength, humidity, etc.
.
To emphasize the advantages and the importance of these functional new fibers for special application areas such as technical textiles.
.
And also to discuss experimental result of researches in papers with other researchers and producers.
Research deliverables (academic and industrial) Pilot scale fiber production, nano-composite fibers.
Publications Erdogan, U.H., Erdem, N. and Cireli, A.A. “Polypropylene/silica nano-particle composite fibers”, In 7th AUTEX Annual Textile Conference Tampere-Finland, June 2007.
Buca/I˙zmir, Turkey
15
Dokuz Eylul University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +902324127211; Fax: +902324127210; E-mail:
[email protected] Principal investigator(s): Assoc. Prof.DrEnder Yazgan Bulgun ¨ zge S¸ahin, DrOzan Kayacan Research staff: Assist Prof.Dr O
Thermo-controlled smart garment design Other Partners: Academic
Research register
Industrial
None Dokuz Eylul University Electrical and Electronics Engineering Dept. Project started: May 2007 Project end date: July 2008 Project budget: e10,000 Source of support: Dokuz Eylul Univ. Scientific Research Fund, Turkish Scientific and Technical Research Council Keywords: Smart garments, Electro-textiles, Textile based conductive materials, Steel yarns, Heating “Smart Systems and Materials” are getting more and more attention in recent years and have a great potential in the field of textiles. The smart/interactive textile structures integrate electronics and textile materials. These products, called ‘the garments of the future’, involve different functions such as protection, actuation, communication, etc. The garments, which can heat the body, will possibly be one of the most widely used products for future use in daily life. These products are developed especially for the use of the people who work outside during their day. The thermal heating occurs in the thermal panels that are placed inside the garment. The procedures used to design this kind of garments can be grouped into two major topics. The first is to fulfill the needs of the comfort properties as it is an ordinary textile product while the second is to meet the functional requirements as a warming system. Heating amount, durability, sufficient working time and determining the optimum power source are among the major parameters in point of designing of a portable structure. In this study, developments about smart/interactive garments having electronic functions were investigated. Steel based conductive yarns were used to produce heating panels within the study being one of the first about interactive electronic garment design in Turkey. Portable power supplies were applied to fabric based panels to obtain heating function. Beside an electronic circuit, a functional garment containing all system was
IJCST 20,6
designed and produced. Performance of the heating garment prototype was evaluated on a thermal mannequin by testing under the cold weather environments.
Project aims and objectives
16
The aim of the project is to design a smart garment containing stainless steel fabric based heating panels and test all system under cold weather conditions. Single and multi-ply steel fabrics are applied to electrical current and their heating behaviors are observed and compared.
Research deliverables (academic and industrial) . .
Comparisons about different portable batteries. Integration of electronic circuit and textile based materials on the garment.
Publications O. Kayacan and E. Bulgun, “Heating behaviors of metallic textile structures”, 3rd International Technical Textiles Congress Proceeding Book,62-70, Istanbul, December 2007.
Buca/Izmir, Turkey Dokuz Eylu¨l University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 00902324127211; Fax: 00902324127210; E-mail:
[email protected] Principal investigator(s): Prof. Dr Ays¸ e Okur Research staff: Res. Ass. Musa Kilic¸
Analyses of unevenness and hairiness on blended yarns Other Partners: Academic
Industrial
None None Project started: May 2008 Project end date: May 2010 Project budget: e52,000 Source of support: Dokuz Eylu¨l University Keywords: Blended yarns, Hairiness, Unevenness, Modal, Tencel, Promodal Yarns made of regerated cellulosic fibres such as modal, tencel, promodal and blends of these with cotton have wider use especially in recent years. However, when the literature is checked, it is found that there are not adequate scientific researches on unevenness and hairiness of these kinds of yarns. So, in this project it is aimed to analyze the unevenness and hairiness of blended yarns made of cotton/modal, cotton/tencel and cotton/promodal. In the project, the effect of linear density and twist, the effect of blend type and the effect of spinning system on hairiness and unevenness will be analyzed by using yarns made of 100 per cent cotton, 67 per cent-33 per cent, 50 per cent-50 per cent,
33 per cent-67 per cent cotton-regenerated cellulosic fibre and 100% regenerated cellulosic fibre. Also, performance properties of fabrics made of these yarns will be analyzed within the context of the project.
Research register
Project aims and objectives In this project it is aimed to analyze the unevenness and hairiness of the yarns statistically, determine the relationships between unevenness-hairiness and offer a more realistic formula to calculate the limit irregularity of these blended yarns. It is hoped that the results of the project which has a fairly wide experimental study will be useful for practical uses.
Research deliverables (academic and industrial) The effect of blend ratio, linear density, twist, blend type, spinning system on hairiness and unevenness of blended yarns will be observed at the end of the study. Also, relationships between hairiness and unevenness of these yarns will be analyzed. Within the context of the project, performance properties of the fabrics made of these yarns will be analysed. A new formula is aimed to be derived from the statistical analyses for the limit irregularity of the blended yarns. Furthermore, different measuring principles for the yarn hairiness will be compared. It is thought that this project will contribute to the further studies on hairiness and unevenness of blended yarns made of regenerated cellulosic fibres with cotton. Publications Not available.
Buca/Izmir, Turkey Dokuz Eylu¨l University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90 232 4127211; Fax: +90 232 4127210; E-mail:
[email protected] Principal investigator(s): Prof. Dr M. Sevil Yesilpinar Research staff: Zeynep Ezgi Senuyar
Development of a computer-aided program on the preparation of garment pattern Other Partners: Academic
Industrial
Yrd. Doc¸. Dr Vecdi Aytac¸ (Ege University) Project started: 1 July 2008 Project end date: 1 July 2009 Project budget: e10,360 Source of support: (TUBITAK) The Scientific and Technological Research Council of Turkey Keywords: Garment, Garment simulation, E-learning
17
IJCST 20,6
18
It’s so difficult to imagine for a student what the lecturer explains on the board about pattern and the view of the sewn product. The student should understand first the shape of 2D patterns and then the constitution of the garment. In somebody’s opinion understanding the subjects explained on 2D plane is easy but for the other it has different difficulties. The goal of this project is to develop an education program that provides to combine the patterns prepared on 2D plane as if they are sewn. The aim of this program is to support the use of e-learning, to visualize the garment patterns and to help students and professionals who begin pattern learning. Another purpose of this program is to explain and let cognize the basics of garment pattern in an easier and in a faster way by simulations which are still 3D as this is the case with most of detail avoiding computer drawings and photographs used in this field. On the Project different skirt and trouser models are chosen as material. The 2D patterns of the skirt and trouser models will be obtained on 2D CAD system. The most appropriate softwares are chosen to compose the 3D materials for the program and to design the interface to constitute the target educational garment pattern CD. In this context, the programs 3DS Max, Cult3D, Adobew Flashw CS3 Professional which work on Windows Vista operating system will be used. In this project with the help of the technological developments effective, easy and fast teaching of the preparation of garment patterns is aimed. Therefore, realistic objects, animations and visual materials will be taken. So the students will have the advantage of viewing on computer the 3D view of the 2D patterns as if they’re sewn. For example, on a skirt model they will be able to understand the order of the parts like pocket flap, pocket bag, front pattern. Consequently, they will have a thorough knowledge of the sewing process. Furthermore, they will watch the combining of garment patterns on a 3D and interactive computer environment. They will see the front, back and side views of the garment. They will be able to watch the virtual garment alone or on a mannequin with the fit measures. And this will accelerate the learning period.
Project aims and objectives The 3D systems used in the product development projects became an international research field in clothing industry for the last 10 years. Companies have used 2D CAD systems in all processes from creating the garment pattern to preparing the cutting layout. However, there are just a few “accurate” 2D CAD systems which are able to prepare cutting layout by considering the kind of material. These programs have a feature that allows to make graphical presentations of samples, colors and surfaces. In this context to be in need of being used of these garment simulations in educational area cannot negate. That means these programs provide to learn easily and consequently to explane in a shorter time. As is known, the future of a country and industry depends on the skilled workers and personnel. The better people are improved during the education the sooner they would take an active and important part in the business life. This improving hangs on not only the learning person but also the teaching person and the type of education. The aim of this project is to develop an education program that provides to combine the patterns prepared on 2D plane as if they are sewn. The aim of this program is to support the use of e-learning, to visualize the garment patterns and to help students and professionals who begin pattern learning. Another purpose of this program is to explain
and let cognize the basics of garment pattern in an easier and in a faster way by simulations which are still 3D as this is the case with most of detail avoiding computer drawings and photographs used in this field.
Research register
Research deliverables (academic and industrial) Publications Not available.
19
Budapest, Hungary Budapest University of Technology and Economics, Department of Polymer Engineering, 9 Muegyetem rkp. Budapest H-1111 Hungary. Tel: 36-1-463-1529; Fax: 36-1-463-1527; E-mail:
[email protected] Principal investigator(s): La´szlo´ M. Vas Research staff: Pe´ter Tama´s, Pe´ter Nagy, Veronika Nagy, Zsolt Ra´cz, Zolta´n L. Simon, Zolta´n Gombos
Structural and strength modelling and complex testing of fibrous structures and fiber reinforced composites Other Partners: Academic
Industrial
None BUTE Department of Information Engineering Project started: 2005 Project end date: 2008 Project budget: e43,000 Source of support: Hungarian Scientific Research Fund – OTKA T049069 Keywords: Modelling, Testing, Fiber bundles, Fibrous structures, Reinforcements, Composites Besides the fibers, fiber groups called fiber bundles as intermediate structural elements play an important role in the macro-scale behavior and properties of the fibrous structures such as slivers, rovings, yarns, webs, fleeces, fabrics, and reinforcements of polymeric composites. The Department of Polymer Engineering has dealt with testing textiles and fibrous reinforcing systems using image processing and with the theoretical fundamentals of their computational structural-mechanical modeling essentially for more than one decade. In the framework of some earlier projects (OTKA I/3 821, 1991-1994; OTKA I/5 T7652, 1993-1994) a novel modeling method [1-4] was developed applying the so-called idealized fiber-bundle-cells as structured statistical mechanical model elements to modeling fiber flows such as slivers or fiber bundles to flat bundle test or twisted fiber
IJCST 20,6
20
structures such as filament or spun yarns. In addition, in co-operation with the Research Institute of Technical Physics and Materials Science Budapest some geometrical and mechanical testing techniques [5-8] for measuring the diameter of fibers or yarns and the fiber orientation on webs or yarn surface as well as the local deformation of fabrics during tensile test were elaborated using image processing technology. The idealized fiber-bundle- cells [1,2,12] are statistical because the strength parameters of the fibers (tensile strength, breaking strain, initial tensile stiffness), and the bundle properties of the fiber cells (pretension, order), as well as the parameters for describing the grip of the fibers or fiber chains (slippage load, slippage length) are stochastic variables. From the sub-bundles combined, so-called composite bundles or fiber bundle chains can be formed. According to the modeling concept a finite composite fiber bundle or fiber bundle chains or network built up of randomly or deterministicly oriented fibers ungripped or gripped at one or two ends can be the structural-mechanical model of a real textile considered concerning not only the increasing load but the damage process as well [1-4, 12-13, 16-17]. Another research work (OTKA T022077, 1997-1999) extended to the fiber reinforced composites focused on modeling the structure of glass fiber mats and testing the structural-mechanical properties of the one layer glass mat reinforced unsaturated polyester (UP) resin composite sheets. Besides the structural-geometrical model of short fiber structures developed [9-10] a testing method based on image processing was worked out for studying the local deformation and crack propagation of transilluminated composite specimens during tensile test. Using the mobile CCD camera image processing system developed in the framework of this project [5-6] the cracks coming into being during the tensile test were detected as fibers or edges and by that the fiber deformation and crack propagation processes could be described numerically by evaluating the fiber orientation histograms recorded [11]. The next project in this topic (OTKA T038220, 2002-2004) aimed mainly at testing and modeling the damage and fracture process of unidirectional carbon fiber reinforced epoxy resin composites during flexural test as well as analyzing the effect of the geometry. According to experiences the modeling method using statistical fiber-bundlecells can be describe the expected value of the bending load in an acceptable way even if the simplest fiber-bundle-cell was used and the breakage of the stretched layers were modeled [12-13]. The fiber-bundle-cells method was also applied to modeling the mechanical behavior of braided carbon fiber reinforced composite tubes subjected to tensile load [15]. Two PhD Theses made use of the results. The current research project (OTKA T049069, 2005-2008) undertook to summarize and complete the results of the structural-mechanical testing and modeling methods and work out some new applications, as well as – in collaboration with the Department of Information Engineering BUTE – develop a software package for modeling and a complex testing method including geometrical and mechanical tests, acoustic measurements and image processing [14,18]. This is the topic of a PhD Thesis and a DSc Thesis.
Project aims and objectives The aim of the research is, based on the theory of the fiber flows and fiber bundles as well as the results of experimental and FEM examinations, to develope a modelling method and software suitable for estimating and characterizing statistically the
structural-geometrical and strength properties of different fibrous structures (fiber bundles, slivers, rovings, yarns, fiber mats, and fabrics), unidirectional composites and specimens cut from laminated composite sheets in different directions as well as their damage processes during tensile and flexural tests. On the other hand, it is planned to develope a complex testing and avaluating system based on a universal tensile tester, some CCD camera image processing equipment and an acoustic emission device suitable for studying, analysing the statistical behaviour, the damage process, and the scale effects of single fibers, rovings, fiber reinforced composite specimens. Objectives and tasks: .
Developing image and AE processing aided methods for the complex testing of the strength properties of unidirectional and laminated composites specimens.
.
Experimental and FEM examinations of the relationships between the structural and the strength properties of unidirectional and laminated composite specimens cut in different directions.
.
Developing statistical structural-strength basic models and modelling methods for describing the tensile and flexural strength properties of the unidirectional and laminated composite specimens.
.
Developing some applications of the structural-strength models.
.
Developing a statistical structural-strength modelling software package based on the results obtained above.
Research deliverables (academic and industrial) The realized results of the project have been expected as follows: .
Testing and evaluation methods using image processing.
.
Windows based modelling software realized in Delphi Code.
.
2-3 papers published in periodicals annually.
.
2-3 papers published at conferences annually.
.
2-3 MSc Theses annually.
In the framework of the research three PhD Theses have been elaborated as well as the results have been expected to be used by undergraduate and PhD students for preparing papers, MSc Theses and other PhD Theses. Publications and outputs
Earlier publications: 1. Vas, L.M. (1990), “The statistical fiber bundle strength and its application in testing fibers and yarns (in Hungarian)”, Magyar Textiltechnika, Vol. 43 No. 4, pp. 165-85. 2. Vas, L.M. (1992), “Latest results in the tensile test theory of fiber and yarn bundles ordered into plane (in Hungarian)”, Magyar Textiltechnika, Vol. 45 No. 3, pp. 71-5, (5/6), 137-42, (7/8), 187-191. 3. Vas, L.M. and Csa´szi, F. (1993), “Use of composite-bundle theory to predict tensile properties of yarns”, Journal of the Textile Institute, Vol. 84 No. 3, pp. 448-63. 4. Vas, L.M. and Hala´sz, G. (1994), “Modelling the breaking process of twisted fiber bundles and yarns”, Periodica Polytechnica Ser. Mech. Eng., Vol. 38 No. 4, pp. 325-50.
Research register
21
IJCST 20,6
22
5. Vas, L.M., Hala´sz, G., Taka´cs, M., Eo¨rdo¨gh I., Sza´sz, K. (1994), “Measurement of yarn diameter and twist angle with image processing system”, Periodica Polytechnica Mech. Eng., Vol. 38 No. 4., pp. 277-96. 6. Hala´sz, G., Taka´cs, M., Vas, L.M. (1994), “Image processing system for measuring geometrical properties of fibres and yarns”, Fibres & Textiles in Eastern Europe, January/February, pp. 30-33. 7. Vas, L.M., Hala´sz, G. (1994), “Untersuchung der Vera¨nderungen in Fadendiameter und DrehungsWinkel bei der Zug- und Drehbeanspruchung”, Periodica Polytechnica Mech. Eng., Vol. 38 No. 4, pp. 297-324. 8. Vas, L.M., Hala´sz, G., Nagy, P., Eo¨rdo¨gh, I., Juha´sz, G., Sza´sz, K. (1997), “Testing the deformation of textile fabrics using image processing system (in Hungarian)”, Magyar Textiltechnika, Vol. 50, 1.sz., pp. 19-25. 9. Vas, L.M., Balogh, K., Nagy, P. and Gaa´l, J. (1998), “Structural modeling and testing of glass fiber mats (in Hungarian)”, Magyar Textiltechnika, Vol. 51 (In-Tech-Ed’98 Conference Budapest, October 21-22. 1998, special issue), pp. 67-71. 10. Vas, L.M., Balogh, K. (2000), “Testing fiber orientation and its effect on glass mats by using image processing system”, VI. International Conference IMTEX’2000, Lodz, June 5-6 (Zeszty Naukowe No. 845. Wlo´kiennictwo No.58.) Proceedings, pp. 69-78. 11. Vas, L.M. and Balogh, K. (2002), “Investigating damage processes of glass fiber reinforced composites using image processing”, Journal of Macromolecular Sciences Part B – Physics, Vol. B41. Nos. 4/6, pp. 977-89. 12. Vas, L.M. and Ra´cz, Z. (2004), “Modeling and testing the fracture process of impregnated carbon fiber roving specimens during bending, Part I. Fiber bundle model”, Journal of Composite Materials, Vol. 38 No. 20, pp. 1757-85. 13. Vas, L.M., Ra´cz, Z. and Nagy, P. (2004), “Modeling and testing the fracture process of impregnated carbon fiber roving specimens during bending Part II. Experimental studies”, Journal of Composite Materials, Vol. 38 No. 20, pp. 1787-801. 14. Ra´cz, Z., Simon Z.L. and Vas, L.M. (2004), “Analysing the flexural strength properties of unidirectional carbon/epoxy composites”, COMPTEST 2004, 2nd International Conference on Composites Testing and Model Identification, 21-23 September. University Bristol, UK (Poster paper, CD)
Publications of the current research: 15. Simon, Z. and Vas, L.M. (2005): “Relationship between the flexural properties and specimen aspect ratio in laminated composites”, 22nd DANUBIA-ADRIA Symposium on Experimental Methods in Solid Mechanics. Monticelli Terme/Parma, Italy, September 28-October 1 (Poster). 16. Ra´cz, Z. and Vas, L.M. (2005), “Relationship between flexural strength and size effects in unidirectional carbon/epoxy”, Composite Interfaces, Vol. 12. pp. 325-39. 17. Nagy, V. and Vas, L.M., “Testing polyester yarns with overtwisting and estimating the pore sizes (in Hungarian)”, Magyar Textiltechnika, Vol. LVIII. e´vf. Part I: 2005/1. 1-3, Part II: 2005/2. 26-27 18. Gombos, Z., Vas, L.M. and Gaa´l, J. (2005), “Testing and evaluation of resin take-up processes (in Hungarian)”, Anyagvizsga´lo´k Lapja, Vol. 15 No. 3, pp. 97-9. 19. Gombos, Z. and Vas, L.M. (2005), “Determining the pore size of glass fiber mats on the basis of the statistical fiber mats model (in Hungarian)”, Magyar Textiltechnika, LVIII, pp. 89-91. 20. Simon, Z.L. and Vas, L.M. (2005), “Analyzing the flexular properties of laminated composites produced by RTM technology (in Hungarian)”, Anyagvizsga´lo´k Lapja, Vol. 15 No. 4, pp. 119-21. 21. Zsigmond, B. and Vas, L.M. (2005), “Tensile testing of carbon fiber reinforced braided tubes (in Hungarian)”, Mu˝anyag e´s Gumi, Vol. 42 No. 12, pp. 488-91. 22. Nagy, V. and Vas, L.M. (2005), “Pore characteristic determination with mercury porosimetry in staple yarns”, Fibers and Textiles in Eastern Europe, Vol. 13 No. 3, July/September, pp. 21-6. 23. Gombos, Z., Nagy, V., Vas, L.M. and Gaa´l, J. (2005), “Investigation of pore size and resin absorbency in chopped strand mats”, Periodica Polytechnica, Ser. Mech. Eng., Vol. 49 No. 2, pp. 131-48.
24. Vas, L.M. (2006), “Strength of unidirectional short fiber structures as a function of fiber length”, Journal of Composite Materials, Vol. 40 No. 19, pp. 1695-734. 25. Vas, L.M. and Cziga´ny, T., “Strength modeling of two-component hybrid fiber composites in case of simultaneous fiber failures”, Journal of Composites Materials, Vol. 40 No. 19. pp. 1735-62 (IF ¼ 0,671). 26. Zsigmond, B. and Vas, L.M. (2006), “Examination of the tensile state of fibers in braided fiber reinforced composite tubes”, Periodica Polytechnica Ser. Mech. Eng., Vol.50 No.1, pp. 67-76. 27. To¨ro¨k, P. and Gombos, Z. (2006), “Determination of the gel point from the exotherm effect of UP resin (in Hungarian)”, Anyagvizsga´lo´k Lapja, Vol. 16 No. 1, pp. 14-18. 28. Gyivicsa´n, P., Gombos, Z. and Vas, L.M. (2006), “Examination of pore size distribution of chopped strand mats with image analysis system (in Hungarian)”, Magyar Textiltechnika, Vol. 59 No. 5, pp. 146-49. 29. Vas, L.M. and Tama´s, P. (2006), “Fiber-bundle-cells method and its application to modeling fibrous structures”, GE´PE´SZET 2006, 5th Conf. on Mech. Eng. Budapest, May 25-26, Proceedings (CD – Fulltext) ISBN 963 593 465 3. 30. Vas, L.M. (2006), “Idealized fiber bundles and their application to modeling fibrous structures (in Hungarian)”, Scientific Session of Committee of Fiber and Composite Technology Hungarian Academy of Sciences, Budapest, 28 February. 31. Vas, L.M. and Tama´s, P. (2006), “Fiber-bundle-cells method and its application to modeling fibrous structures”, GE´PE´SZET 2006, 5th Conf. on Mech. Eng. Budapest, May 25-26, Proceedings (CD – Fulltext) ISBN 963 593 465 3. 32. Gombos, Z. and Vas, L.M. (2006), “Temperature dependence of resin absorption of various chopped strand mats”, GE´PE´SZET 2006, 5th Conf. on Mech. Eng. Budapest, May 25-26, Proceedings (CD – Full-text), pages: 6. ISBN 963 593 465 3. 33. Meggyes, G., Gombos, Z. and Vas, L.M. (2006), “Analysing the orientation and size effects in composite sheets in tensile tests”, GE´PE´SZET 2006, 5th Conf. on Mech. Eng. Budapest, May 25-26, Proceedings (CD – Full-text), pages: 6. ISBN 963 593 465 3. 34. Simon, Z., Szabo´, L. and Vas, L.M. (2006), “Determination of flexular modulus by image processing”, GE´PE´SZET 2006, 5th Conf. on Mech. Eng. Budapest, May 25-26, Proceedings (CD – Full-text) ISBN 963 593 465 3. 35. To¨ro¨k, P., Gombos, Z. and Vas, L.M. (2006), “The effect of curing temperature upon the mechanical properties of unsaturated polyester resin”, GE´PE´SZET 2006, 5th Conf. on Mech. Eng. Budapest, May 25-26, Proceedings (CD – Full-text), pages: 6. ISBN 963 593 465 3. 36. Nagy, V. (2006), “Examination and modeling of porosity in polyester twisted fibrous structures”, PhD thesis, BUTE Budapest. 37. Ra´cz, Z. (2006), “Analyzing the bending characteristics and damage processes of unidirectional composite beams (in Hungarian)”, PhD thesis, BUTE Budapest. 38. Vas, L.M. (2006), “Statistical modeling of unidirectional fiber structures. macromolecular symposia”, Special Issue: Advanced Polymer Composites and Technologies, Vol. 239 No. 1, pp. 159-75. 39. Gombos Z., Nagy V., Kosˇta´kova´, E. and Vas, L.M. (2006), “Absorbency behaviour of vertically positioned nonwoven glass fiber mats in case of two different resin viscosities”, Macromolecular Symposia, Vol. 239 No. 1, pp. 227-31. 40. Nagy, P. and Vas, L.M. (2006), “Investigating the time dependent behavior of thermoplastic polymers under tensile load”, Macromolecular Symposia. Special Issue: Advanced Polymer Composites and Technologies, Vol. 239 No. 1, pp. 176-81. 41. Simon, Z. and Vas, L.M. (2007), “Relationship between bending modulus and test sizes of laminated glass/polyester composites”, Materials Science Forum, Vol. 537-538, pp. 71-9. 42. Vas, L.M., Poloskei, K., Felhos, D., Deak, T. and Czigany, T. (2007), “Theoretical and experimental study of the effect of fiber heads on the mechanical properties of non-continuous basalt fiber reinforced composites”, Express Polymer Letters, Vol. 1 No. 2, pp. 109-121-91.
Research register
23
IJCST 20,6
24
43. Vas, L.M. (2007), “Modelling polymer composites by using fiber bundle theory based methods (in Hungarian)”, Day of Materials Science. Scientific Session of Committee of Materials Science and Technology Hungarian Academy of Sciences, Budapest, 11 May. 44. Vas, L.M. and Tama´s, P. (2007), “Modeling method based on idealized fiber bundles”, 3rd ChineEurope Symposium on Processing and Properties of Reinforced Polymers, 11-15 June, Budapest (Poster). 45. Gombos, Z., Vas, L.M. and Hruza, J. (2007), “The effect of layer number on air permeability in case of glass fiber mats”, 3rd Chine-Europe Symposium on Processing and Properties of Reinforced Polymers, 11-15 June, Budapest (Poster).
Budapest, Hungary Budapest University of Technology and Economics, Budapest XI, ˝ egyetem rkp. 3 Postal A. H-1521 Budapest, Hungary. MU Tel: +36-1-463-1376; Fax: +36-1-463-1376; E-mail:
[email protected] Principal Investigator(s): Professor Judit Borsa Department or Laboratory: Department of Plastics and Rubber Technology, Department of Physical Chemistry, Department of Chemical Technology
Modification of cellulose fiber for extension of its application Other Partners: Academic
Industrial
Johannes Kepler University, Linz, Austria, Johan Be´la National Center for Dr Habil. Ildiko Tanczos Epidemiology, Inst. for Isotops and Surface Chemistry of the Hungarian Academy of Sciences Project started: 1 January 2005 Project end date: 31 December 2008 Source of Support: Hungarian National Research Fund (OTKA), Governmental Fund (GVOP) Keywords: Cellulose, Cotton, Hemp, Swelling, Chemical modification, Carboxymethylcellulose, Functional textile, Antimicrobial textile, Textile for hospital use Cellulosic fibers are modified by physical and chemical methods: .
Interaction of cotton cellulose with quaternary ammonium hydroxide (tetramethylammonium hydroxide) is studied in comparison with sodium hydroxide.
.
Cotton fiber is modified by slight carboxymethylation. Effect of technology parameters on the properties of the modified fiber, theoretical aspects of modification and some possible application of modified fiber (e.g. antibacterial textile for hospital use) are investigated.
.
Delignification and refinement of various kinds of Hungarian hemp are studied.
Publications Borsa, J., La´za´r, K. and La´szlo´, K., “Antibacterial effect of carboxymethylated cotton fiber”, The Fiber Society Spring Conference, StGallen, Switzerland, May 2005.
Edinburgh, Scotland, UK Heriot-Watt University, School of Engineering and Physical Sciences, Riccarton, Edinburgh, Scotland, UK. EH14 4AS. Tel: 00 44 131 451 3034; Fax: 00 44 131 451 3473; E-mail:
[email protected];
[email protected] Principal investigator(s): Professor J.I.B. Wilson and Dr R.R. Mather Research staff: Ms. A.H.N. Lind
Solar cells in textiles Other Partners: Academic
Industrial
None None Project started: 2001 Project end date: 2010 Keywords: Thin film silicon, Solar energy, Photovoltaics We are developing thin-film silicon solar cells on low cost textile substrates, using chemical vapour deposition (CVD) technology, based on previous thin-film diamond expertise. The CVD technology employs a proprietary microwave plasma system (developed at Heriot-Watt University) with silane/hydrogen/dopant gas mixtures to produce the sequence of layers that forms the active part of these cells. We have shown that relatively low deposition temperatures of 200 degrees C and the active plasma conditions of the process do not affect our textile substrates, whether of woven or nonwoven construction. In addition, solutions have been determined to the problem of providing reliable electrical contacts over fibrous, flexible substrates, together with a conventional transparent conducting oxide as the top contact in the cell “sandwich” structure. Effective “first barrier” encapsulation may also use our deposition technology.
Project aims and objectives Flexible solar cells for a variety of applications:, e.g. building fac¸ades, use in remote areas, emergency use in disaster relief, camping/leisure industry, portable chargers.
Research deliverables (academic and industrial) Working prototype. Publications “Textiles make solar cells that are flexible and lightweight”, Technical Textiles International, December 2002, pp. 5-6. “Solar textiles: production and distribution of electricity coming from solar radiation”, Applications in Intelligent Textiles and Clothing, ed. H. Mattila, Woodhead Publishing Limited, Cambridge, to be published in 2006.
Research register
25
IJCST 20,6
26
Galashiels, UK Heriot-Watt University, School of Textiles and Design, Netherdale, Galashiels, Scotland TD1 3HF. Tel: +44 (0) 1896 892140; Fax: +44 (0) 1896 756701; E-mail:
[email protected] Principal investigator(s): R H Wardman
Verification of algorithm for the numerical specification of standard depths of colour Other Partners: Academic
Industrial
None None Project started: May 2008 Project end date: May 2011 Project budget: £12,000 Source of support: Society of Dyers and Colourists Keywords: Standard depths, Colour, ISO, Textiles
Standard depths of colour are used to determine the fastness properties of dyed materials. The existing standard depths were established over fifty years ago and there is general agreement hat they are not of uniform depth.The project involves the visual assessment of fabric samples dyed to standard depths of shade defined by an algorithm developed by the PI. The visual assessments will be carried out by observers in at least five different countries and their results used to verify the algorithm. The algorithm will then be written into a new proposal for an ISO standard that defines standard depths of colour.
Project aims and objectives To prepare samples according to the algorithm previously published and verify its accuracy by visual assessments of a panel of colourists in five different countires.
Research deliverables (academic and industrial) A new ISO standard for the numerical specification of standard depths of colour. Publications 1. Chen, C.C., Wardman, R.H. and Smith, K.J. (2006), “The mapping of a surface of constant visual depth in CIELAB colour space”, Coloration Technol., Vol. 118, p. 281. 2. Wardman, R.H., Islam S., and Smith, K.J. (2006), “Proposal for a numerical definition of standard depths”, Coloration Technol., Vol. 122, p. 350.
Galashiels, UK Heriot-Watt University, School of Textiles and Design, Netherdale, Galashiels, Scotland TD1 3HF, UK. Tel: +44 (0) 1896 892140; Fax: +44 (0) 1896 756701; E-mail:
[email protected] Principal investigator(s): R M Christie and R H Wardman Research staff: R Shah
Digital fast patterned microdisposal of fluids for multifunctional protective textiles Other Partners: Academic University of Manchester, Hogeschool Gent, University of Lodz, University of Twente
Project started: May 2006 Project budget: e12.6 m Source of support: EU FP6 Keywords: Inkjet printing, Textiles
Industrial Ten Cate, Grado Zero Espace, B&B Corporate Knitwear, Guantenor S.L., J Sarens, N.V., Iris DP S.r.L, JPC S.P., Zoo, Liebaert, Skalmantas, SKA Polska S.P. Zoo, Vexed Generation Ltd, Xennia Technology Ltd, D’Appolonia SpA, Lamberti SpA, Xaar PLC, Saxion Project end date: May 2010
To develop breakthrough technology based on digitally microdisposing fluids on textiles enabling high-speed protective functionalisation, continuous processing and customised production. Digital microdisposal has the ability of exact localisation and patterning of functionalities in multilayer textile substrates integrating advanced thermo and hydro regulation, sensorics, actuating and controlled release functions, based on nano-technology and multifunctional materials.
Project aims and objectives To realise a generic functionalisation technology for making multifunctional protective textiles on the basis of localised and multilayered compounds. To use low temperature and low water processessing by inkjet application of functional compounds. Develop innovations in mechatronics and micro-fluids, nanotechnology and multifunctional materials as well as in micro-nano metrology. Set new standards of performance in personal protective equipment.
Research deliverables (academic and industrial) On-line process for the microdisposal of active fluids onto textile fabrics. Monitoring system integrating different inspection techniques.
Research register
27
IJCST 20,6
Publications
None in the public domain.
Galashiels and other site partners, UK 28
Research Institute for Flexibel Materials, School of Textiles and Design, Heriot Watt University, Galashiels, Selkirkshire, TD1 3 HF, UK. Tel: 01896 892135; Fax: 01896 758965; E-mail:
[email protected] Principal investigator(s): Prof G K Stylios, RIFleX, Heriot Watt University, Dr K Lee, Unilever Research (Industrial), Dr R Potluri, Manchester University and Prof Long, Nottingham University Research staff: L Luo plus others in partner universities and Industrial Partners
Multi-scale integrated modelling for high performance flexible materials Other Partners: Academic
Industrial
University of Nottingham and Manchester Unilever, OCF Plc, TechniTex Faraday Ltd, Crode International Plc, ScotWeave University Ltd, Airbags International Ltd, Moxon of Huddersfield Ltd, Carrington Carrer and Workwear Ltd Project started: 1 January 2007 Project end date: 31 December 2010 Project budget: £1.7 Million Source of support: Department of Trade and Industry DTI Keywords: Modelling, Yarn, Fabric, Garment, Hagh performance This is a flagship proposal for the UK. It is based on integrating micro, meso and macro scale structure/property and deformation models for high performance flexible materials. The outputs will be industry targeted solutions for predicting the properties and behaviour of high performance flexible materials in deformed states during usage including garments. The proposal stems from the modelling achievements of the three academic partners, combining their complimentary work, and after integration, adapting them for industry use, with Unilever, the lead partner, and 12 companies covering diverse applications of the outputs, attempting to represent this sector. OCF and Scotweave will help commercialising the output in the form of a software product or licence. The main aim of the project is to develop the models such that they can be used primarily by the high performance textiles and garment industry, but also by other industries dealing with flexible materials. The key strength of the proposal is that for the first time an attempt is made to model the behaviour of flexible materials in a 3D manner, taking into account the dynamic changes of performance related properties with physical changes during use:
.
Bridging the gap between (a) industry needs for predictive and development tools and (b) academic modelling efforts on two levels: structure/property micro/meso scale and macro-scale whole flexible structures such as high performance clothing.
.
Modelling of overall properties and performance of simple deformed textile structures such as draped/creased/folded fabrics, predicting the dynamic changes of performance related to physical deformations. This will be an innovative engineering tool for industries involved in designing, developing and manufacturing flexible materials, initially targeted at fabrics, but with applications to paper and thin films.
.
Modelling simple high performance whole garments based on structural/geometrical parameters of the flexible materials, to predict the changes in whole garment performance caused by yarn and fabric changes. No model that can perform this function for the high performance clothing industry is yet available. This aspect of the work will be a powerful tool for the high performance clothing and garment industry (high performance medical wear, protective clothing, sportswear, etc.)
The project will lead to two way modelling: predicting properties and performance of deformed textile materials and whole garments from structural, mechanical and geometrical parameters, or vice versa, i.e. generating structural, mechanical and geometrical requirements for specific end-product characteristics. Focusing on the high performance clothing sector, the work will provide a first-in-its-kind tool for design, development, engineering and optimisation of high performance flexible materials and exploited by a consortium of diverse companies, three of which are technology providers and the rest users of the technology. Different mechanisms (e.g. interfacial modification) can be exploited in the manufacture, modification, cleaning and care of high performance multi-component textiles. This will permit rational design of these materials and associated products, reducing the need for experimental development and testing. This project will push the boundaries in multi-scale modelling by building on leading edge expertise in fibre to yarn scale modelling at Manchester University, yarn to textile scale modelling at Nottingham University, and textile to whole flexible structures and clothing at Heriot-Watt University. By developing and interfacing these areas closely coupled tools will allow completely predictive models to be developed. It will also be the first time that modelling techniques have been used to understand the effects of interfacial properties on textile performance, leading to increasing innovation in high performance textile products.
Project aims and objectives .
Combining the strong modelling expertise of the four academic partners to create multi-scale integrated models for predicting structural properties and performance of flexible materials.
.
Optimising the integrated models for industry use, focusing in the first instance at high performance clothing and technical textiles.
Research register
29
IJCST 20,6
30
.
Increasing time and cost efficiency of industry in product design and fabrication, starting with the high performance clothing industry.
The objectives address industry needs for new and improved product design and fabrication aids for high performance materials, e.g. in view of the London 2012 Olympics (high performance sportswear, anti-terrorism protective clothing, etc.). They will enable the prediction of properties and performance before manufacturing, hence accelerating the development stage, reducing costs and increasing competitiveness.
Research deliverables (academic and industrial) A Virtual Testing capability will be developed and transferred into industry, reducing the amount of experimentation required for new product development (new textiles, treatments and laundry products designed and manufactured) and thereby reducing the associated costs and waste generation (improving sustainability of production). Direct beneficiaries of this work will be manufacturers of apparel, textile, technical and allied industries. Indirect beneficiaries will include manufacturers of carpets, non-wovens, composites, paper and structural materials, as well as retailing and software (after commercialising): .
.
.
Bridging the gap between (a) industry needs for predictive and development tools and (b) academic modelling efforts on two levels: structure/property micro/meso scale and macro-scale whole flexible structures such as high performance clothing. Modelling of overall properties and performance of simple deformed textile structures such as draped/creased/folded fabrics, predicting the dynamic changes of performance related to physical deformations. This will be an innovative engineering tool for industries involved in designing, developing and manufacturing flexible materials, initially targeted at fabrics, but with applications to paper and thin films. Modelling simple high performance whole garments based on structural/geometrical parameters of the flexible materials, to predict the changes in whole garment performance caused by yarn and fabric changes. No model that can perform this function for the high performance clothing industry is yet available. This aspect of the work will be a powerful tool for the high performance clothing and garment industry (high performance medical wear, protective clothing, sportswear, etc.).
Publications
Too early.
Galashiels, Scotland, UK Research Institute for Flexibel Materials, School of Textiles and Design, Heriot Watt University, Galashiels, Selkirkshire TD1 3HF, UK. Tel: 01896 892135; Fax: 01896 758965; E-mail:
[email protected]
Principal investigator(s): Prof G K Stylios Research staff: X Zhao
Research register
Integration of CFD and CAE for design and performance assessment of protective clothing Other Partners: Academic None
31 Industrial Tilsatec Ltd, TechniTex Faraday Ltd, Camira Fabrics Ltd, St Jame’s University Hospital, Remploy Ltd, Pil Membranes Ltd, Altair Engineering Ltd Project end date: 31 May 2010
Project started: 1 June 2007 Project budget: £600,000 Source of support: Engineering and Physical Sience Rerearch Council EPSRC and Department of Trade and Industry DTI Keywords: CFD, CAE, Protective fabric, Garment, Apparel, Modelling high performance This collaborative proposal aims at improving and developing new textiles for protective clothing by integrating Computational Fluid Dynamics (CFD) and Computer Aided Engineering (CAE). A new industrial tool for predicting the diffusion of chemical and/or biological (CB) agents through multilayer, non-homogenous flexible porous materials such as fabrics and whole garments will also be established. Multi-layer textile structures (flat and shaped) and simple garments will be modelled, using equations of mass, heat and momentum balance, integrated with human computational representation. The outcome will be optimisation of current commercial fabrics/garments and developing new protective clothing whilst, at the same time offering a new objective tool for product designers, engineers and developers to predict and evaluate performance of CB protective products.
Project aims and objectives .
Develop and integrate CAE/CFD for modelling of a clothed human to predict the performance of textiles and garments in CB protective applications, linking with fabric design and manufacturing.
.
Establish an objective measure protocole for CB protection end uses.
.
Improve protection performance of (PPE) and (PC) through new product development for extreme conditions.
.
Develop new materials by better understanding of the complex interactions between the flow of CB agents and textile/materials properties.
Research deliverables (academic and industrial) With a worldwide focus on CB agents, the project is timely for industry, considering the legislative demands for public health, safety and security. Growing concerns from
IJCST 20,6
32
government, the public and private sector about threats of terrorist attacks or epidemic outbreaks have led to increased performance and evaluation requirements. Increasing emphases on security for international events (e.g. London 2012) are also drivers for the above innovations. The proposed project addresses these issues by development key technologies for immediate and long-term use. The consortium experience can achieve the project objectives and lead to a major breakthrough in PC. Publications
Too early.
Galashiels, Scotland, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, UK. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail:
[email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Mohamed Basel Bazbouz
Investigating the spinning of yarn from electro-spun nanofibres Other Partners: Academic
Industrial
None None Project started: November 2004 Project end date: October 2009 Project budget: N/A Source of support: N/A Keywords: Electro-spinning, Polymers, Nanofibre, Alignment, Yarn, Composite Our laboratory is using a process called electrospinning which has the ability to produce a wide variety of polymeric fibres with diameters from a couple of micrometer down to the nanometer scale. In this case different structures can be made from electrospun fibers to suit the needs of various industries. Electrospinning, a fibre spinning technique that relies on electrostatic forces to produce fibres in the nanometer to micron diameter range, has been extensively explored as a simple method to prepare fibres from polymer solutions or melts. Under the influence of the electrostatic field, a pendant droplet of the polymer solution at the spinneret is deformed into a conical shape (Taylor cone). If the voltage surpasses a threshold value, electrostatic forces overcome the surface tension, and a fine charged jet is ejected. As these electric static forces increase, the jet will elongate and accelerate by the electric forces. The jet undergoes a variety of instabilities, dries, and deposits on a substrate as a random nanofibre mat.
In our work, nonwoven electrospun mats of nylon 6 produced from solutions with formic acid with different concentrations are examined. Each nonwoven mat with average fibre diameters from 200 to 1300 nm was prepared under controlled electrospinning process parameters. Effects of electric field and tip-to-collection plate distances of various nylon 6 concentrations in formic acid on fibre uniformity, morphology and diameters were measured. Processing parameters effects on the morphology such as fibre diameter and its uniformity of electrospun polymer nanofibres was investigated. A process optimization summarized the effects of solutions properties and processing parameters on the electrospun nanofibre morphology was issued. In our work, we Control the electrospinning process to move away from just collecting random fibre mesh to enabling the accurate deposit of fibres at any predetermined position. This will be by using a simple method of getting a fibres bundle made of aligned nanofibres between two known points. This collection process has been termed as the ‘gap method of alignment’ involves grounding two circular disks from the spinneret. We have demonstrated that it is possible to produce continuous fiber yarn made out of electrospun nanofibres. The current process has the potential to spin nanofibre at a commercially viable rate.
Project aims and objectives .
Understanding of the electrospinning process.
.
Optimizing its process parameters to electrospin polymers into nanofibres with desired morphology.
.
Discussing the models proposed for jet forming, jet travel, processing instabilities.
.
Tensile testing of polymeric nanofibres.
.
Controlling the spatial alignment of electrospun fibres.
.
A technique for making continuous fibre bundle yarns from electrospun fibres.
.
Investigating the other methods for producing continuous fibre bundle yarns, Core-yarn and laminate composite consisting of aligned fibres in different Directions.
Research deliverables (academic and industrial) .
Nonwoven electrospun mats of nylon 6 produced from solutions with formic acid with different concentrations are examined.
.
Each nonwoven mat with average fibre diameters from 200 to 1300 nm was prepared under controlled electrospinning process parameters.
.
Effects of electric field and tip-to-collection plate distances of various nylon 6 concentrations in formic acid on fibre uniformity, morphology and diameters were measured.
.
Processing parameters effects on the morphology such as fibre diameter and its uniformity of electrospun polymer nanofibres were investigated.
Research register
33
IJCST 20,6
.
A process optimization summarized the effects of solutions properties and processing parameters on the electrospun nanofibre morphology was issued.
.
The well-known Chauchy’s inequality is applied to prediction the velocity of the end of the jet in electrospinning.
.
A critical relationship between radius r of jet and the axial distance z from nozzle is obtained for the straight jet.
.
Draw ratio between the jet and the final fibres was pridicted theoritically. Manufacturing of aligned fibres array was easily achievable.
34 . .
Processing parameters effects on the aligned fibres such as gab distance and collection time were investigated.
Publications and outputs
In preparation: . Systematic parameter study for ultra- fine nylon 6 fibre produced by electrospinning technique. .
Electrospinning of aligned nanofibres with cotrolled deposition.
Conferences Paper and Poster: . June 2006 (Electrospinning of nanofibres: potential scaffolds for medical applications) a presentation presented in Research in Support of Medicine, Health and Safety Conference, Edinburgh, UK. . June 2006 (Systematic parameter study for ultra- fine nylon 6 fibre produced by electrospinning technique), Poster presentation, Research in Support of Medicine, Health and Safety Conference, Edinburgh, UK.
Galashiels, Scotland, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, UK. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail:
[email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Mohammad Mahfuzur Rahman Chowdhury
Investigating nanofiber production by the electrospinning process Other Partners: Academic
Industrial
None
None
Project started: July 2004 Project end date: June 2009 Project budget: N/A Source of support: N/A Keywords: Electrospinning, Electrospinning process, Parameters, Polymer, Nanofibre application
Research register
35 Electrospinning is a unique way to produce novel polymer nanofibres with diameter typically in the range of 10 nm to 500 nm. Using this process, the polymer nanofibres can be made from a variety of polymer solutions or melt to produce fibres for a wide range of applications. Electrospinning occurs when the electrical force at the surface of a polymer solution or melt overcomes the surface tension and causes an electrically charged jet to be ejected. When the jet dries or solidifies, an electrically charged fibre remains. This charged fibre can be directed or collected or accelerated by electrical forces, then collected in sheets or other geometrical forms. This research project is an investigation of the electrospinning process and the effect of process variables on orientation, crystallinity, microstructure and mechanical properties of the nanofibres produced. Some of the polymeric parameters investigated are polymer type, solvent type, molecular weight, solution properties, viscosity, conductivity and surface tension. In the case of process parameters, the electric potential, flow rate, concentration, distance between capillary and collection screen, ambient parameters are important.
Project aims and objectives .
. .
To investigate process-structure-property relationships in polymer fibres with nanosize diameters produced by electrospinning. To innvestigate the morphology and properties of the polymer nanofibres. To produce fibres at uniform diameters
Research deliverables (academic and industrial) . . . . .
Nanofibres of uniform diameter. Defined mechanical and physical properties. Process-structure-property relationships. Detailed understanding of the electrospinning process. Nanofibres suitable for applications such as air filtration, protective clothing, fibre reinforced support, and Biomedical.
Publications “Nano fibre and its medical application”, Poster presentation in Research in support of Medicine, Health and Safety”, Conference, Heriot-Watt University, Scotland, UK.
IJCST 20,6
36
Galashiels, Scotland, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, UK. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail:
[email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Dr Taoyu Wan
Novel micro-channel membranes for controlled delivery of biopharmaceuticals Other Partners: Academic None
Industrial Stryker UK Ltd, St James’s University Hospital, Leeds General Infirmary, Camira Fabrics Ltd, Dinsmore Textile Solutions Ltd Project end date: March 2009
Project started: April 2006 Project budget: £650,000 Source of support: DTI Technology Programme Keywords: Micro-channel, Micro-porous, Membrane, Controlled delivery, Drug release
This project, which stems out of research findings of an EPSRC-funded research, aims at developing micro channel structures (coatings, membranes, foams, etc.) with encapsulated biopharmaceuticals capable of controlled release by changes in temperature, pH, magnetic field or voltage. A technique for encapsulating biopharmaceuticals into micro-channels of a polymer matrix structure and controlling their subsequent release will be developed. Driven by the consortium, the technologies will be veered towards various pay-load bearing applications, e.g. in self-supporting materials for delivering bone growth hormones (INN eptotermin alpha) in bone fractures, or in coated textiles for the personal hygiene, healthcare, treatment and protective clothing industries.
Project aims and objectives The main objectives for the projects are to: .
Establish criteria for channel size and distribution control; engineer the structure and morphology of the porous material to suit the encapsulation of biopharmaceuticals and their slow release. This involves both self-supported materials and coated textiles.
.
Investigate UV-based techniques for the encapsulation of specific biopharmaceuticals, with particular reference to bone growth hormones.
.
Investigate strategies for the controlled release of the biomaterials from the channels within the developed membrane or coating.
Characterise and test laboratory and pilot-scale samples – study release rates of the biopharmaceuticals, degradation/absorption rates and efficacy of the system. . Investigate alternative exploitation routes as coated textiles with encapsulated biopharmaceuticals. The project will investigate possibilities both for implantable and non-implantable payload bearing materials, both as self-supported materials, and as coatings or membranes supported by a base fabric. Research work on implantable applications will however only be performed in laboratory scale with the aim to prove the concepts. .
Research deliverables (academic and industrial) .
.
Pilot-scale production of membranes, coatings, foams, or any other forms of channel-containing materials with engineered, controllable and tailor-made channel sizes, shapes and distributions. Defined characteristics, properties and performance of the as-produced materials.
.
A technique to encapsulate biopharmaceuticals within the membrane channels using a UV cross-linking technique, which eliminates the need to use high temperature treatments for encapsulation. This enables the biopharmaceuticals to maintain their effectiveness.
.
The final expected result is the development of a system for activated release, e.g. following a change in stimuli or external conditions, such as temperature, pH, magnetic field or voltage. This will enable the controlled and targeted release of the biopharmaceutical as, when and where required for optimum treatments after fractures.
Publications Stylios, G.K., Giannoudis, P.V. and Wan, T., “Applications of nanotechnologies in medical practice”, Injury, Vol. 36S, pp. S6-S13, 2005. Stylios, G.K., Wan, T. and Giannoudis, P.V., “Present status and future potential in the enhancement of bone healing using nanotechnology”, Accepted for publication in Injury, 2006.
Galashiels, Scotland, UK Heriot-Watt University, RIFleX, School of Textiles and Design, Netherdale, Galashiels TD1 3HF, United Kingdom. Tel: +44 1896 89 2135; Fax: +44 1896 75 8965; E-mail:
[email protected] Principal investigator(s): Prof. George K. Stylios Research staff: Liang Luo
Interactive wireless and smart fabrics for textiles and clothing
Research register
37
IJCST 20,6
38
Other Partners: Academic
Industrial
None None Project started: September 2002 Project end date: September 2009 Project budget: N/A Source of support: Worshipful Company of Weavers Keywords: Smart, Interactive, Textiles, Garment, Clothing, Sensors, Wireless The last few years have witnessed an increased interest in wearable technologies, smart fabrics and interactive garments. This has come about by certain technological innovations n the areas of sensor-based fabrics, micro devices, wire and wireless networks. In terms of textiles, most of current developments are towards the fashion markets and have resulted in glorifying garments as gimmicky gadgets. However, some efforts are also being directed in using the technology for improving the quality of life, or even for life saving purposes. Examples of such uses can be found in the military, healthcare, fire fighting, etc. This research project investigates new interdisciplinary technologies in fabrics, sensors and wireless computing, for the development of a prototype interactive garment for monitoring various functions of the wearer.
Project aims and objectives The general aim of the project is to develop technologies for use in interactive garments, which can provide monitoring functions for various applications such as the clinical or healthcare sector. More specifically, objectives are: .
Develop suitable wireless sensors for various measurements, including ECG, temperature, breathing, skin conductivity, mobility and movement, humidity, positioning, etc.
.
Develop a Personal Area Network and a Wireless Communication Centre. Optimise suitable wireless technologies such as Bluetooth to enable communication between sensors and a central processing unit. Conceptualise a smart multilayer fabric.
.
. .
Integrate technologies
Research deliverables (academic and industrial) .
Wireless sensors for physiological and other measurements.
.
Wireless communication centre for relaying information between sensors, wearers, central processing unit and Internet.
.
Conceptual multilayer fabric suitable for interactive garments
Publications Stylios, G.K. and Luo, L., “Investigating an interactive wireless textile system for SMART clothing”, 1st International Textile Design and Engineering Conference (INTEDEC 2003), Fibrous Assemblies at the Design and Engineering Interface, Edinburgh, UK, September 22-24, 2003.
Stylios, G.K. and Luo, L., “The concept of interactive, wireless, smart fabrics for textiles and clothing”, 4th International Conference, Innovation and Modelling of Clothing Engineering Processes – IMCEP 2003, Maribor, Slovenia, October 9-11, 2003. Stylios, G.K. and Luo, L., “A SMART wireless vest system for patient rehabilitation”, Wearable Electronic and Smart Textiles Seminar, Leeds, UK, June 11, 2004. Stylios, G.K., Luo, L., Chan, Y.Y.F. and Lam Po Tang, S., “The concept of smart textiles at the design/technology interface”, 5th International Istanbul Textile Conference, Recent Advances and Innovations in Textile and Clothing, Istanbul, Turkey, May 19-21, 2005.
Galashiels, Scotland Heriot Watt University, Scottish Borders Campus, Netherdale, Galashiels, TD1 3HF. Tel: 01896 892245; Fax: 01896 758965; E-mail:
[email protected] Principal investigator(s): Dr Alex. Fotheringham Research staff: Basel Younes
Optimisation of weaving process for biopolymers Other Partners: Academic
Industrial
None None Project started: October 2006 Project end date: September 2010 Project budget: Source of support: British Council Keywords: Fibre extrusion, Yarn production, Weaving This is part of a general research programme in the production and use of biopolymers. Current research investigates the optimisation of biopolymer fibre production using experimental design techniques to statistically map processing parameters to mechanical properties. Having produced a model of the fibre and yarn production, the relationship between fibre/yarn characteristics, weaving parameters and fabric properties can then be established. Related work is researching into the use of gas plasma for the pre-treatment of polylactic acid fabric, the printing of knitted biopolymer fabrics and dyeing of such materials in fibre/yarn form.
Project aims and objectives Aims: .
To establish the relationships which exist between processing and final properties.
.
To identify the value of gas plasma treatment on biopolymer fabrics for, e.g. dyeing, coating, etc.
.
To optimise printing on knitted biopolymer fabrics.
Research register
39
IJCST 20,6
Objectives: . .
40
To use experimental designs to create statistical models. To use statistical models as forecasting tools to establish the relationships between processing and properties.
Research deliverables (academic and industrial) A statistical model of fire, yarn and fabric (weaving) processing. To create similar models for fibre dyeing and printing. Publications Not available.
Hong Kong, China Institute of Textile and Clothing, The Hong Kong Polytechnic University, The Hong Kong Polytechnic University.Hung Hom, Kowloon, Hong Kong. Tel: 00852-27666470; Fax: 00852-29542521; E-mail:
[email protected] Principal investigator(s): Prof. Xiaoming TAO Research staff: Dr Guang-feng Wang, Dr Xiao-hong SUN, Dr Bo Zhu, Dr Yangyong Wang, Dr Jiang-ming Yu, Mr Zhi-feng Zhang, Mr Wei Zheng
Small sized fiber sensors Other Partners: Academic
Industrial
None None Project started: 1 March 2008 Project end date: 28 February 2010 Project budget: HK$5,264,000 Source of support: The Hong Kong Research Institute of Textiles and Apparel Limited; Innovation and Technology Commission, The Government of the Hong Kong Special Administrative Region;Hong Kong Tak Ying Trading Company; Pool Heng Company Limited; Best Technology Company Limited; Esquel Enterprises Ltd Keywords: Sensor; Small, Fibers In applications such as medical devices, industry, robotics and wearable electronics, the size of sensors is a very important parameter and sensors with smaller size are required. Fibers of a few microns in diameter are ideal candidates. Based on the previous research conducted at the laboratories of The Hong Kong Polytechnic University, the present project is aimed to further developing small sized fiber sensors with conductive fibers and assemblies, and polymeric photonic fiber. The sensors to be developed should have high sensitivity, long service life, reasonable material and production cost. The proposed project will investigate and develop design and fabrication technologies of the small
sized fiber sensors, as well as the package of the fiber sensors, and explore their applications in industries.
Research register
Project aims and objectives .
To select and fabricate materials for the small size fiber sensors.
.
To make appropriate fibers. To develop the structural design for the small size fiber sensors.
. . .
To develop fabrication technology and equipment for the fiber sensors. To explore package technologies for the fiber sensors.
Research deliverables (academic and industrial) .
The optimized design and materials for the small size fiber sensors in term of performance and cost.
.
Fabrication technology and equipment for the small size fiber sensors. Package method and production technology of the small size fiber sensors.
. .
A test protocol for performance and reliability of the small size fiber sensors.
.
The theoretical model for the single cell of the fiber sensors.
Publications 1. Tao, X.M., Yu, J.M. and Tam, H.Y., “Photosensitive polymeric optical fibres and gratings”, Transactions of the Institute of Measurement and Control, Vol. 29 No. 3, pp. 1-16, 2007. 2. Yu, J.M., Tao, X.M. and Tam, H.Y., “Trans-4-stilbenemethanol-doped photosensitive polymer optical fiber for grating fabrication”, Optics Letters, Vol. 29 No. 2, pp. 156-8, 2004. 3. Zhang, A.P., Guan, B.O., Tao, X.M. and Tam, H.Y., “Experimental and theoretical analysis of fiber bragg gratings under lateral compression”, Optics Communication, Vol. 206, pp. 81-7, 2002. 4. Zhang, H., Tao, X.M., Yu, T.X. and Wang, S.Y., “Conductive knitted fabric as large-strain gauge under high temperature”, Sensors and Actuators A, 126, pp. 129-40, 2006. 5. Zhang, H., Tao, X.M., Yu, T.X. and Wang, S.Y., “A novel sensate string for large-strain measurement under high temperature”, Measurement Science and Technology, Vol. 17, pp. 450-8, 2006. 6. Yang, B., Tao, X.M. and Yu, J.Y., “A study on the relationship between resistance and strain based on stainless steel fabric”, Rare Metal Materials and Engineering, Vol. 35 No. 1, pp. 96-99, 2006. 7. Xue, P. and Tao, X.M., “Morphological and electromechnical studies of fibers coated with electrically conductive polymers”, Applied Polymer Science, Vol. 98, pp. 1844-54, 2005. 8. Li, Y., Leung, Y.M., Tao, X.M., Chen, X.Y., Tsang, H.Y.J., Yuen, M.C.W., “Polypyrrole-coated fabrics as a candidate for strain sensors”, J. Materials Science, Vol. 40 No. 15, pp. 4093-95, 2005. 9. Li, Y., Chen, X.Y., Leung, Y.M., Tao, X.M., Tsang, H.Y.J., Yuen, M.C.W., “A flexible strain sensor from polypyrrole coated fabrics”, Synthetic Metals, Vol. 155 No. 1, pp. 89-94, 2005. 10. Zhang, H., Tao, X.M., Yu, T.X. and Wang, S.Y., “Electro-mechanical properties of stainless steel knitted fabric made from multi-filament conductive yarns under unaxial extension”, Text. Res. J., Vol. 75 No. 8, pp. 598-606, 2005. 11. Yang, B., Tao, X.M. and Yu, J.Y., “Fiber bragg grating sensor for simultaneous measurement of strain and temperature”, J. Industrial Textiles, Vol. 34 No. 2, pp. 97-116, 2004. 12. Yu, J.M., Tao, X.M. and Tam, H.Y., “Fabrication of UV sensitive single-mode polymeric optical fiber”, Optical Materials, Vol. 28 No. (3), pp. 181-8, 2006.
41
IJCST 20,6
42
13. Yang, B., Tao, X.M., Ho, H.L. and Yu, J.Y., “Tangential load measurement by optic smart cellular textile composite”, Text. Res. J., Vol. 74 No. 9, pp. 810-8, 2004. 14. Yang, B., Tao, X.M. and Yu, J.Y., “Measurement effectiveness of fiber bragg grating sensors in textile composites”, J. Text. Res., Vol. 25 No. 3, pp. 48-9, 2004. 15. Xue, P., Tao, X.M., Yu, T.X., Kwok, K. and Leung, S., “Electromechanical behavior and mechanistic analysis of fibers coated with electrically conductive polymer”, Text. Res. J., Vol. 74 No. 10, pp. 929-36, 2004.
Izmir, Turkey Dokuz Eylul University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90-232-4127211; Fax: +90-232-4127210; E-mail:
[email protected] Principal investigator(s): Prof. Dr Arif Kurbak Research staff: Dr Ozlem Kayacan
Investigations and improvements on the properties of medical textiles Other Partners: Academic
Industrial
None None Project started: 18 April 2006 Project end date: 18 April 2009 Project budget: e22,000 Source of support: Dokuz Eylul University Keywords: Microclimate cooling garment, Water cooling garment, Thermal manikin, Heat stress, Knitted garment, Medical textiles The body temperatures of individuals could increase when they are working in hot conditions, when they have special illness, etc. In order to decrease this temperature, special garments are needed. These garments are called microclimate cooling garments and can be classified as water cooling, air cooling and phase changing materials cooling systems. They are used in military clothes, space suits, protective clothes, in surgical clothes and in medical field to relieve the symptoms of special diseases like multiple sclerosis and ectodermal dysplasias. In this work, four types of water cooling garments, which are different from the other researches, are designed. In order to investigate the cooling effects of these garments, a test method is developed. For this aim a cooling device (chiller) which is pumping the water into these garments and a thermal manikin are designed and manufactured. The effects of water inlet temperature and flow rate on cooling effect and the differences of manikin temperature are investigated.
Project aims and objectives In this study cooling effects of different liquid cooling garments are investigated. For this aim four different types of liquid cooling garments are designed in order to alleviate the heat stress of patients in hospitals such as who have special illnesses like multiple sclerosis and hypohydrotic ectodermal dysplasia, etc. The effects of water inlet temperature and flow rate on cooling effect and the differences of manikin temperature are investigated. The most effective garment design is chosen according to the cooling temperatures of the manikin and the rate of cooling of the chiller.
Research deliverables (academic and industrial) Publications
Not available.
Izmir, Turkey Dokuz Eylul University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90 232 4127211; Fax: +90 232 4127210; E-mail:
[email protected],
[email protected] Principal investigator(s): Prof. Ays¸ e Okur, Sibel Kaplan Research staff: Prof. Ays¸ e Okur, Assoc.Prof. Serhan Ku¨cuka, Sibel Kaplan
Development of a method to determine garment thermal comfort Other Partners: Academic
Industrial
None None Project started: 1 July 2007 Project end date: 1 December 2008 Project budget: e25,000 Source of support: Dokuz Eylul University, The Scientific and Technological Research Council of Turkey (TUBITAK) Keywords: Clothing thermal comfort, Dynamic sweating hotplate system, Thermal manikin system, Subjective wear trials. Clothing comfort is one of the parameters affecting purchase decisions of people especially in recent years as it has a decisive influence on the daily life and work performance of people. Thermophysiological comfort is one of the important dimensions of comfort and affected by psychological and physiological state of the body and the physical mechanisms occurring in body-clothing-environment system. It affects the thermoregulation mechanism of the body, hence the physiological and psychological status of an individual. In this study, a method including objective and subjective thermal comfort measurements was developed to determine the thermal comfort performances of
Research register
43
IJCST 20,6
44
fabrics/garment systems. Knitted fabrics having different physical/constructional properties were tested by the developed dynamic sweating hotplate system to determine their thermal and water vapor resistance values. This device was used also for determination of water vapor permeability performance of garments under dynamic transfer conditions by a new calculation method put forward in this study. Thermal and water vapor resistance values of garments produced from some of these fabrics were determined by thermal manikin system and subjective wear trials were also conducted by using these garments with a group of professional sportsmen. Dynamic sweating hotplate and thermal manikin systems which are the devices used for objective thermal comfort measurements were developed and produced within this study. Some physiological parameters (skin temperature, microclimate temperature and relative humidity, heart rate, etc.) were measured and subjective coolness-dampness-comfort evaluations were gathered by psychological scales from the subjects during subjective wear trials carried out under determined environmental conditions and activity program. Forearm test, a method used for determination of coolness-dampness to touch sensations during skin-fabric contact was also carried out with the selected sportsmen as a preparation stage for the subjective wear trials. Relationships between objective comfort parameters, subjective and physiological data were investigated to find out how much the objectively measured parameters can simulate real wear conditions. Moreover, thermal comfort performances of different fabrics were determined in the light of their physical/constructional/permeability properties. Conclusions were put forward about comfort performances of different fabrics and relationships between objectively and subjectively measured parameters.
Project aims and objectives Throughout the history, clothing has changed from being a medium protecting human from unsuitable physical environments to a multi-dimensional concept affecting social life and psychology of a human. According to the consumer research studies, comfort has been identified as one of the key attributes taken into account in garment preferences. Comfort can be classified into three basic categories: sensorial comfort related to the psychological and physiological changes arising from skin-textile contact, body movement comfort and thermal comfort including heat and mass transfer mechanisms occuring in body-clothing-environment system. Thermal comfort performance of a garment has a decisive influence on thermoregulation mechanism of a body. As comfort is a sensation affected from the interrelations among body-clothing-environment system parameters, it is a nebulous and complex subject to define. Therefore, it is not possible to measure comfort property of a fabric/garment by using only objective measurement devices; it needs a more comprehensive study. In studies about determination of thermal comfort properties of garments, objective and subjective evaluation methods were developed and results were put forward about the relations between parameters measured with these methods and fabric/garment properties. Studies about objective comfort evaluation methods have generally focused on system developments for measuring thermal and water vapour resistance properties of fabrics/garments as these properties have a significant influence on thermal comfort. It is not possible to measure the sensations of a person about his/her garment objectively, so subjective wear trial method is used to evaluate comfort perceptions. It has been stated by many researchers
that the most realistic results for clothing comfort evaluations were obtained by using both objective and subjective methods. This study aims to put forward a method to determine thermal comfort performances of fabrics/garment systems including objective and subjective measurement systems or methods. To carry out this aim, two objective thermal comfort measurement systems, a dynamic sweating hotplate and a thermal manikin system were developed and produced to conduct thermal and water vapor resistance measurements on fabrics and garment systems. A group of knitted fabrics used for sportive garment production were investigated in the study. Subjective wear trials were also conducted with the garments measured by the thermal manikin system on a group of professional sportmen. There are two objectives for this study; one of them is to determine the relations between the objective measurement system results and the other is to put forward conclusions about the thermal comfort performances of the investigated fabric group by taking into account both objective and subjective results. The more the correlation between the objective comfort parameters and subjective comfort evaluations, the better the performance of a measurement system to simulate the real wear condidions. The relations between the resistance measurement results of fabrics and garment systems are also important. If a prediction model for garment can be developed by considering the thickness of microclimate air layer from the resistance values of fabrics, thermal comfort may be determined by a simpler and cheaper system. And if the thermal comfort performances of garments determined by objective and subjective measurements is significantly correlated, it can be concluded that the objective measurement systems developed can be a useful tool to determine the thermal comfort performances of garments.
Research deliverables (academic and industrial) Objective and subjective measurement results obtained were evaluated to investigate the interrelations between them and to put forward conclusions about the thermal comfort performances of fabrics/garment systems. According to the results, thermal and dampness sensations arise during skin-fabric contact are related to the surface and permeability properties of fabrics and the effect of fabric material is smaller. Among the physiological data obtained during subjective wear trials, temperature and relative humidity of the microclimate can reflect the differences between different fabrics better. Relative humidity of the microclimate is also significantly correlated with the subjective coolness and dampness sensations. According to the subjective evaluation and thermal resistance results, fabrics introduced as having higher water absorption (coolmax) and insulation (thermolite) capacities did not show their advantages about thermal comfort under the selected conditions. Significant relations could not be found out between methods (BS, ASTM, B and water vapor resistance values determined by the dynamic sweating hotplate system) used for determination of water vapor transmittance characteristics of fabrics. However, it was observed that water vapor permeability values determined by the ASTM Standard and resistance values determined by the dynamic sweating hotplate system differentiated fabric permeability performances better than the other methods. Insignificant differences between the thermal resistance values of different garments determined by the developed thermal manikin system may be attributed to the fact that selected environmental conditions was not sufficient to create the necessary temperature difference. It is thought that results obtained from this
Research register
45
IJCST 20,6
46
study can contribute to the literature about thermal comfort performances of different fabrics and relationships between objective-subjective thermal comfort parameters. Publications Kaplan, S. and Okur, A., “The meaning and importance of clothing comfort and sensory descriptors used to describe discomfort: a case study for Turkey”, Journal of Sensory Studies (in press). Kaplan. S. and Okur. A., “Investigating the relations between fabric properties and coolness to touch sensation with forearm test”, AUTEX 2008 World Textile Conference, 24-26 June, Biella, Italy, 2008. Kaplan, S. and Okur, A., “Determination of the product attributes and sensory descriptors related to clothing comfort: a case study for Turkey”, AUTEX 7th Annual Textile Conference, 26-28 June, 2007, Tampere, Finland. Kaplan, S. and Okur, A., “Effects of heat and mass transfer mechanisms in textile materials on clothing thermal comfort”, Tekstil ve Mu¨hendis, Vol 62-63, pp. 28-36, 2006 (in Turkish). Kaplan, S. and Okur, A., “Effects of permeability-conductivity properties of fabrics on clothing thermal comfort”, Tekstil Maraton, March-April, pp. 56-65, 2005 (in Turkish).
Izmir, Turkey Dokuz Eylul Uni. Dept. of Textile Engineering, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +902324127211; Fax: +902324127210; E-mail:
[email protected] Principal investigator(s): Prof. Dr Arif Kurbak Research staff: Tuba Alpyıldız
Studies on textile composites Other Partners: Academic
Industrial
None None Project started: December 2007 Project end date: December 2010 Project budget: e30,000 Source of support: Dokuz Eylul Univesity Keywords: Knit, Composite, Impact, Reinforcement Textile preforms are to be investigated and improved as reinforcements in the structural composite materials. Among the textile preforms, knitted fabrics can be formed, with considerably low costs, into almost every possible shape by making use of their extensional deformability offering the advantage that a more homogeneous fibre content is achieved over the entire surface of the part, and also at points of strong curvature. In the composite materials reinforced by knitted fabrics, fibre orientation distribution in the composite is determined by the knit structure and does not change significantly during the production of the composite. In this study different knitted structures will be investigated and the structure with sufiicient mechanical properties will be indicated.
Project aims and objectives The major aim of this project is to carry on the studies of the improvements of the textile preforms and to manufacture a kniitted reinforcement which is light in weight and has adequate resistance against impact.
Research register
Research deliverables (academic and industrial)
47
Publications Not available.
Izmir, Turkey Dokuz Eylu¨l University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 0090 232 4127211; Fax: 0090 232 4127210; E-mail:
[email protected] Principal investigator(s): Aysun Cireli Aks¸ it Research staff: Bengi Kutlu, Nurhan Onar, Mu¨ge Ko¨se
Usage of plasma technology for development of conductivity properties of textile materials Other Partners: Academic
Industrial
None None Project started: May 2008 Project end date: May 2010 Project budget: e40,000 Source of support: T.C. Dokuz Eylu¨l U¨niersity (BAP) Keywords: Plasma polymerization, Textile, Conductivity The surface structure of fibres is very important in processing and use, since friction, abrasion, wetting, adhesion, adsorption and penetration phenomena as well as antistatic behaviour are involved. In order to obtain textile materials with a desired property, the fibre surface is often modified with polymer, inorganic or hybrid organic/inorganic layers before use. The demand for electrically conductive fibres and textiles has increased in recent years because of applications as antistatic materials, sensors, materials for electromagnetic shielding and biomedical use. However, an ideal method for modification remains to be found for the preparation of stable conductive textiles. We want to give conductive feature on the textile materials with the use of plasma technology. In plasma processing technology, it is well established that exposure to plasmas generated in inert gases and/or reactive gases can clean the surface of materials and modify their characteristics, particularly their surface energy. Active species from the plasma bombard and/or react with monolayers on the surface of materials and change their surface properties either temporarily or permanently. Such work includes metals or polymers of interest in many industries and less commonly, textiles.
IJCST 20,6
48
Plasma technology applied to the treatment of textiles has developed markedly during the past decade, due to its potential environmental and energy conservation benefits, in developing high-performance materials for the world market and. In practice, the surface properties of natural and synthetic fibres or filaments can be modified using plasma treatment. This can lead to processes such as polymerisation, grafting, crosslinking, etc. with concomitant effects on wetting and wicking, dyeing, printing, surface adhesion, electrical conductivity and other characteristics of interest in the textile industry. Since adhesion is a surface-dependent property, mediated at a molecular scale, plasma technology can effectively achieve modification of this near-surface region without affecting the bulk properties of the materials of interest. Like polyaniline, polypyrrole conductive chemicals which compose a conductive thin film on textile’s surface is covered in this project. For furnising conductivity on textiles this project needs plasma machine which work with low pressure, have RF (radio frequency) plasma (10-100W and 30-360 s working power and time) with heat system (25-95 C).
Project aims and objectives Aim of this project is, composing conductive thin film on textile’s surface with conductive chemicals like polyaniline, polypyrrole by plasma technology.
Research deliverables (academic and industrial) To obtain conductivity on textile materials. Publications Not available.
Izmir, Turkey Dokuz Eylu¨l University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 0090 232 4127211; Fax: 090 232 4127210; E-mail:
[email protected] Principal investigator(s): Aysun Cireli Aks¸ it Research staff: Bengi Kutlu
Development of flame-retardant properties of synthetic and natural fibers using plasma polymerization technology Other Partners: Academic
Industrial
Mehmet Mutlu None Project started: May 2005 Project end date: November 2008 Project budget: e72000 Source of support: TUBITAK Keywords: Plasma polymerization, Textile, Flame retardancy, Ecological treatment, Fiber
Flame retardancy can be defined as retardation of ignition in textiles when exposed to flame, burning without flame and/or self-extinguishability after ignition. To make a textile material flame retardant, it is necessary to decrease the heat released by burning or to increase the heat required to burn the textile material. A flame retardant textile material is required to have a limiting oxygen index more than 27, a maximum char length of 18 cm in flammability tests, low release rate of smoke and toxic gases when burning, and not to change its physical properties when a flame retardant applied. The most common flame retardants are phosphorous and phosphorous-nitrogen based agents, halogen derivatives, aliphatic, aromatic and inorganic compounds.In conventional process, textile material is impregnated in a bath containing flame retardant agent at a pick-up of 70-80%. After that, it is dried at 110-150 C and cured at 150-175 C. Concentrations of chemicals used for effective flame retardant finishes are 300500 g/l. Using such high chemical concentrations is hazardous to environment and performance properties of textile material. For example, when using phosphate compounds, phosphoric acid releases in curing and it causes a decrease at about 20-40% in strength of cotton textile materials. Bromine compounds are extremely harmful for human health and safety, and environment. These compounds may cause dermatitis, asthma and cancer. It is known that some triazine substituted bromo-alkyl compounds show mutagenic activity. In addition to this, brominated compounds in water are taken by fishes and accumulate in their bodies. As a result of progressive accumulating, their amount may be fatal when they are taken by human. Studies shows that brominated diphenyl ether compounds accumulate in air, ground waters, soil, fishes, sea mammals, bird eggs and even in human milk. Toxic chlorine and chlorine-methyl ether gases occur when working with chlorinated compounds. To eliminate the hazards of flame retardants to human health and environment, researchers have tried to develop new chemicals but they cannot get the desired results. For instance, chlorine in THPC is substituted by a hydroxyl group to get THPOH, and by sulphur to get THPS. The way to reduce the hazards to a minimum is to use minimum amount of chemicals. Therefore, development of new methods instead of new chemicals is the most suitable solution for this problem. In this condition, plasma polymerization technology, in which water is not used and very small amount of chemical is used, would be the best alternative to the conventional flame retardant finishes.
Project aims and objectives One of the aims of the project is to prevent any wastewater and to save water by development of flame retardant properties of synthetic and natural fibers using – nonaqueous- plasma polymerization technology. The other aim is to eliminate the negative effects originated by chemicals used and to introduce an ecological production method which is harmless for environment and human health by decreasing the amount of chemicals used.
Research deliverables (academic and industrial) In the scope of this project, changing the process time, plasma gas, plasma discharge power and flow rate and using various monomers, textiles made from various synthetic and natural fibers will be made flame retardant by plasma polymerization. Commercial flame retardants and siloxane monomers will be used as flame retardants. After the
Research register
49
IJCST 20,6
50
treatments, properties of the textile material will be evaluated by burning performance tests such as limiting oxygen index, vertical flammability test and by fabric performance tests such as breaking strength. Chemical and physical changes at the surface will be determined by SEM, AFM and FTIR-ATR techniques. Publications 1. Cireli, A., Kutlu, B. and Mutlu, M. “Surface modification of polyester and polyamide fabrics by low frequency plasma polymerization of acrylic acid”, Journal of Applied Polymer Science, Vol. 104 No. 4, pp. 2318-22. 2. Aysun Cireli, Bengi Kilic¸ (Kutlu) and Mehmet Mutlu, “Surface modification of cotton fabrics using low pressure nitrogen plasma”, 6th International Conference TEXSCI 2007 – Book of Abstracts, 273-74. 3. Kılıc¸, B. and Cireli, A., (2007), “Plazma Teknolojisinin Tekstillerde Kullanımı Ve Tekstil Materyalleri U¨zerindeki Etkileri, II.” Tekstil Teknolojileri Ve Tekstil Makinalari Kongresi 19-20 Ekim 2007, Gaziantep-Tu¨rkiye. 4. B. Kilic, A. Aksit and M. Mutlu, “Surface modification and characterization of cotton and polyamide fabrics by plasma polymerization of hexamethyldisilane and hexamethyldisiloxane, III”, International Technical Textiles Congress 1-2 December 2007 Istanbul.
Izmir, Turkey Dokuz Eylu¨l University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 090 232 4127211; Fax: 0090 232 4127210; E-mail:
[email protected] ¨ zdemir Principal investigator(s): Ismail O Research staff: Aysun Cireli Aks¸ it, Nurhan Onar, Erdal C¸elik, Mehmet Mutlu, Ibrahim Avgin, Is¸ il Kayatekin, M. Faruk Ebeog˘lugilL
Production and development of radar-absorbing textiles using sol-gel processing – investigation of electrical, magnetic and microwave properties Other Partners: Academic
Industrial Ekoten Company
Dokuz Eylul University Textile Engineering Dept., Hacettepe University, Plasma Aided Bioengineering and Biotechnology Research Group (PABB), EGE University, Engineering Faculty, Department of Electrical and Electronic Engineering Project started: February 2007 Project end date: February 2009 Project budget: e82,000 Source of support: TUBITAK Keywords: Conductive polymers, Sol-gel, Organic-inorganic hybrid materials, Radar absorbing materials, Camouflage
Radar absorbing materials (RAM) constitute the part of stealthing defence systems for land, air and sea forces in military. While plane, rocket, helicopter, ship and tank as well soldier are illuminated with radar, target is established by radar cross section knowledge. It is important that stealthing techniques, which required for reducing radar cross section, are improved. Thus, it is obtained that the target is hardly determined. Electromagnetic energy could be partly transformed to heat energy by lossy materials obtained with stealthing technologies. The absorbing effect of such materials is very important. Thus, in this study, various textiles will be coated with absorbing and lossy materials prepared using sol-gel processing. It will be obtained that the textile material has the properties of light weight, flexible, easy handling by means of the covering of the textiles with materials. It may be achieved the stealthing technologies while the target was covering with coating textiles. It was determined that textile coating with radar absorbing materials prepared with sol-gel processing was not applied. The aims of the research project were to prepare and to characterize the RAM materials obtained with sol-jel methods and to determine the electrical, magnetic and microwave properties of coated textiles with the RAM materials. With this regard, RAM materials will be obtained from magnetic materials prepared with sol-gel methods and from conductive polymers prepared with chemical oxidative polymerization and will be coated on the cotton fabrics. Thus, such metalic based solution will be prepared, and then turbidimetry test, pH measurement for the solutions will be achieved. Chemical structure of products at film production in intermediate temperatures and the type of reaction will be determined and will be obtained proper process regime while using fourier transform infrared and differential thermal analysis-thermogravimetry apparatus. Structural analysis such phase, thin film and stress and surface morphology of producted RAM materials will be determined while using XRD and SEM-EDS, respectively. The electrical, magnetic and microwave properties of the coated textiles will be determined and compared with commercial RAM materials. Moreover, the durability for washing and tensile strength of the coated textiles will be tested.
Project aims and objectives Our goal is to produce the cotton fabric with microwave absorbing in 6-14 GHz frequency of microwave range.
Research deliverables (academic and industrial) Microwave absorbing materials has been intensively investigated since 1987. There are limited researchs about the RAM materials in Turkey. In this project, we will produce the cotton fabric with microwave absorbing in 6-14 GHz frequency of microwave range. It is important to develop RAM materials the for national defensive industry with regard to stealthing technology. Publications 1. Aysun Cireli Aks¸it, Nurhan Onar, M. Faruk Ebeoglugil, Isil Kayatekin, Erdal Celik and Ismail O¨zdemir, “Electromagnetic and electrical properties of coated cotton fabric with barium ferrite doped polyaniline film, APP-2008-06-1917, 2008, Journal of Applied Polymer Science (submitted).
Research register
51
IJCST 20,6
52
2. Onar, N., Aks ¸ it, A., Ebeoglugil, M.F., Birlik, I., C¸elik, E. and O¨zdemir, I., “Conductivity and magnetic properties of coated fabrics with barium ferrite doped aniline solution, III”, International Technical Textiles Congress, 1-2 December 2007, Istanbul Fair Center, Yesilkoy/Istanbul, pp. 198-206 (oral presentation). 3. Onar, N., Aksit, A., Avgin, I., Celik, E., Ebeoglugil, M.F., Kayatekin, I. and Ozdemir, I., “Magnetic properties of coated fabrics with barium ferrite doped silica sol”, 10th International Conference and Exhibition of the European Ceramic Society, June 17-21, 2007, Berlin (oral presentation).
Izmir, Turkey Dokuz Eylu¨l University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 0090 232 4127211; Fax: 0090 232 4127210; E-mail:
[email protected] Principal investigator(s): Aysun Cireli Aks¸ it Research staff: Nurhan Onar, Umit Halis Erdog˘an, Mu¨ge Ko¨se
Producing conductive fibers and developing their properties Other Partners: Academic
Industrial
None None Project started: July 2008 Project end date: July 2010 Project budget: e30,000 Source of support: T.C. Dokuz Eylul University (BAP, Scientific Research Project) Keywords: Fibre, Textile, Conductivity The multifunctional textiles for leisure clothing are required while progressing of technology. Increasing of functional properties of textile materials provided useable of textile materials in various areas. For example, intelligent textiles intensively attract the interest in the world. Conductivity textiles are the part of intelligent textiles. Conductive textiles can be used from data transfering to electromagnetic shielding, antistatic properties, heating element and sensors, so on. In the project, we will produce the powder of polyaniline and polypyrrole polymer as conductive polymer. Then we will add the powders to the melting during spinning. What’s more, we will coat the fabric with polyaniline and polypyrrole film.
Project aims and objectives In the project, we aimed to produce conductive textiles by coating with conductive polymers by chemical oxidative process and by melt spinning the textile fibers by doping of conductive polymer powders produced by chemical oxidative polymerization.
To the aim, we wil produce polyaniline, polypyrrole polymers. What’s more, we will characterized the samples by using FTIR, DTA-TG, XRD and SEM.
Research register
Research deliverables (academic and industrial) To occur conductivity on textile materials by chemical oxidative polymerisation method. Publications 1. Aysun Cireli Aks¸it, Nurhan Onar, M. Faruk Ebeoglugil, Isil Kayatekin, Erdal Celik and Ismail O¨zdemir, “Electromagnetic and electrical properties of coated cotton fabric with barium ferrite doped polyaniline film, APP-2008-06-1917, 2008, Journal of Applied Polymer Science (submitted). 2. Onar, N., Aks¸it, A., Ebeoglugil, M.F., Birlik, I., C¸elik, E. and O¨zdemir, I., “Conductivity and magnetic properties of coated fabrics with barium ferrite doped aniline solution, III”, International Technical Textiles Congress, 1-2 December 2007, Istanbul Fair Center, Yesilkoy/Istanbul, pp. 198-206 (oral presentation). 3. Onar, N., Aksit, A., Avgin, I., Celik, E., Ebeoglugil, M.F., Kayatekin, I. and Ozdemir, I., “Magnetic properties of coated fabrics with barium ferrite doped silica sol”, 10th International Conference and Exhibition of the European Ceramic Society, June 17-21, 2007, Berlin (oral presentation)
Izmir, Turkey Dokuz Eylu¨l University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: 0090 232 4127211; Fax: 0090 232 4127210; E-mail:
[email protected] Principal investigator(s): Aysun Cireli Aks¸ it Research staff: Bengi Kutlu
Development of flame-retardant and durable press properties of synthetic and natural fibers using plasma polymerization technology Other Partners: Academic
Industrial
Mehmet Mutlu Project started: June 2006 Project end date: December 2008 Project budget: e30,000 Source of support: T.C.Dokuz Eylu¨l U¨niersitesi (BAP) Keywords: Plasma polymerization, Textile, Flame retardancy, Durable-press, Fiber Development of performance properties of textiles is possible by application of chemical agents in an aqueous medium. These methods are known as finishing treatments.
53
IJCST 20,6
54
They include flame retardancy and durable-press.In spite of a wide variety of treatments and agents, to date, it is still very difficult to durable press and fire retard textiles in a durable manner, i.e. the protection should be resistant to the atmospheric condition for the first application, and for both to several cycles of washing. There are two different ways to confer these properties, either at the beginning of their manufacture by modifying the bulk (macromolecule or polymer) or by surface modification of the fibers/fabrics with active compounds. It is worth considering this later approach interms of durability only if the finishing agent is linked covalently to the fiber/fabric. Among all the different kinds of surface treatments (wet or dry), the cold plasma technique is one of the processes allowing to graft covalently small functional groups as well as macromolecular compounds. Plasma is the fourth phase of materials and has the highest energy. Plasma is occurred when energy which is higher than ionization energy is given. In this work, plasma polimerisation is applied on the synthetic and natural fibers for durable press and flame retardancy. Nitrogen methylol composite, silicone composite, phosphonium salts, composite with halogene and phosphorus nitrogen composites ar used for flame retardancy and durable-press.
Project aims and objectives Aims of the project is to prevent any wastewater and to save water by development of flame retardant properties of synthetic and natural fibers using – non-aqueous- plasma polymerization technology, to eliminate the negative effects originated by chemicals used and to introduce an ecological production method which is harmless for environment and human health by decreasing the amount of chemicals used, to decrease process time and the step of process, to balk the changes on handle properties of textile materials and to compose ecological product method.
Research deliverables (academic and industrial) Synthetic and natural fibers will have flame retardant and durable press properties at the end of this work. Wastle water, changes on handle properties of fabric, decline on endurance value and waste gases which is harmfull for human health won’t be inclueded in this project. Publications 1. Cireli, A., Kutlu, B. and Mutlu, M. “Surface Modification of Polyester and Polyamide Fabrics by Low Frequency Plasma Polymerization of Acrylic Acid” Journal of Applied Polymer Science, Vol.104, No. 4, 2318-22. 2. Aysun Cireli, Bengi Kilic¸ (Kutlu), Mehmet Mutlu, “Surface modification of cotton fabrics using low pressure nitrogen plasma”, 6th International Conference TEXSCI 2007 – Book of Abstracts, 273-74. 3 Kılıc¸ (Kutlu), B. & Cireli, A., (2007), “Plazma Teknolojisinin Tekstillerde Kullanımı Ve Tekstil Materyalleri U¨zerindeki Etkileri, II”. Tekstil Teknolojileri Ve Tekstil Makinalari Kongresi 19-20 Ekim 2007, Gaziantep-Tu¨rkiye. 4. B. Kilic (Kutlu), A. Aksit and M. Mutlu, “Surface Modification and Characterization of cotton and polyamide fabrics by plasma polymerization of hexamethyldisilane and hexamethyldisiloxane, III”. International Technical Textiles Congress 1-2 December 2007 Istanbul.
I˙zmir, Turkey ¨ niversity, Dokuz Eylu¨l University, Department of Dokuz Eylu¨l U Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90232 412 7211; Fax: +90232 412 7210; E-mail:
[email protected] Principal investigator(s): Merih Sarııs¸ ık Research staff: Go¨khan Erkan, Nurdan Pazarlıog˘lu
The microencapsulation of some antifungal agents and their applications to textiles Other Partners: Academic
Industrial
Ege University, Faculty of Scıence None Project started: April 2005 Project end date: September 2008 Project budget: e45,000 Source of support: The Scientific and Technological Research Council of Turkey Keywords: Microencapsulation, Textiles, Ketoconazole, Terbinafine, Melamineformaldehyde, Cyclodextrin
Finishing is the last wet applications of textiles. For that reason this step provides the consumers’ demands such as fire retardant, water repellant, fragrance, etc. It is important that these finishing effects must be saved in an extended period of enduses. Microencapsulation is an important tool for saving of functional finishing effects in an extended period. Especially to need peculiarities such as controlled release, etc. it seems unrivalled. Mikroencapsulation has application areas in other wet applications such as dyeing and printing fields of textiles. Washing and using conditions limit usage life of finishing agents. These agents are protected by a shell, which is achieved by microencapsulation. Thus, microencapsulation is important in applications of agents that are influenced by washing easily. Although microencapsulation can be achieved by using many methods, choosing of microencapsulation methods are influenced by properties of entrapped material and shell polymer. Antifungal pharmaceutical agents were microencapsulated by melamineformaldehyde resin and formed inclusion complexes with cyclodextrin derivatives. The preparations were characterized by differential scanning calorimeter, FT-IR (Fourier transformation-infrared spectroscopy), X-ray diffractometer, particle size analyze and scanning electron microscope. The preparations were applied to the 100% cotton fabric. The fabrics, which microcapsules were applied, were performed washing fastness test. The strength of fabrics to the washing fastness was analyzed by scanning electron microscope and CHNS elemental analyzer. Antifungal properties of washed and unwashed fabrics were evaluated by antifungal tests.
Research register
55
IJCST 20,6
Project aims and objectives . .
.
56
Microencapsulation of ketoconazole and terbinafine. Applications of microcapsules, which contain ketoconazole and terbinafine, to the textiles. To observe their strength to washing.
Research deliverables (academic and industrial) None Publications Erkan, G. and Sariisik, M., “The microencapsulation of some antifungal agents and their applications to textiles”, 6th Internatıonal Conference TEXSCI 2007, June 5-7, Liberec, Czech Republic (2007).
I˙zmir, Turkey Dokuz Eylul University, Dokuz Eylu¨l University, Department of Textile Engineering, Tınaztepe Campus, 35160 Buca/Izmir/Turkey. Tel: +90-232-4127211; Fax: +90-232-4127210; E-mail:
[email protected] Principal investigator(s): DrVildan Sular Research staff: Gonca Balci
Factors affecting yarn friction Other Partners: Academic
Industrial
None None Project started: May 2008 Project end date: May 2010 Project budget: e47,000 Source of support: Dokuz Eylul University Keywords: Yarn friction, Normal load, Contact area, Friction coefficient, Friction force Yarn friction properties is one of the important properties affecting production stages such as yarn production, fabric and garment formation and also the importance of friction properties continues till the end of the life of a textile product. Sometimes friction property is needed although it is undesirable for some processes.Yarn-to-different surfaces (metal, ceramic, etc.) friction, friction between and within yarns play a great role in winding, weaving, knitting and also sewing. Furthermore, the magnitude of friction of textile materials affects most of textile processing. For this reason to investigate yarn friction properties is very important. In this research, frictional properties of yarns produced having different structural properties (fibre composition, linear density, twist, production system) and different test parameters (normal load, contact area, etc.)
will be investigated. Friction coefficients and frictional forces of the yarns will be compared for different conditions and factors affecting these properties will be examined in a systematic way.
Research register
Project aims and objectives Friction is an important property affecting quality, productivity and performance of a product and it is also important for textiles from fiber to finished product. Although there are a lot of researches on frictional properties of textiles about the nature of friction, its impact on textile processing and its role in determining yarn and fabric properties, studies are still going on this topic because there are a lot of parameters affecting friction and to get under control these properties is difficult. This research aims to determine yarn frictional properties of the yarns having different characterictics and espeacially yarns made of new fibers are planned to examine. In the context of this research, also test parameters will be examined.
Research deliverables (academic and industrial) A large database containing different yarn characteristics and yarn frictional properties will be developed at the end of study. This database may be useful for researchers and also friction results may be useful for yarn producers, woven or knitted fabric producers.Relationships between yarn parameters and friction results will be examined and furthermore frictional properties of new fibers and comparison between different yarn types will be given for different test conditions Publications
The research has began on May 2008.
Jalandhar, India National Institute of Technology Jalandhar, Depratment of Textile Technology, National Institute of Technology, Jalandhar – 1440111, India. Tel: +91-181-2690301-2, 453; Fax: +91-181-2690320; +91-181-2690932; E-mail:
[email protected] Principal investigator(s): Dr Arunangshu Mukhopadhyay Research staff: Ms. Sunpreet Kaur
Designing non-woven fabric for pulse-jet filtration Other Partners: Academic
Industrial
None None Project started: 31 March 2005 Project end date: 30 September 2008 Project budget: Rs. 15 lacs Source of support: Ministry of Human Resource Development, Government of India Keywords: Dust particle, Nonwoven, Pulse-jet filtration
57
IJCST 20,6
58
Primary factors which determine the selection of a fabric for a particular application are state of aerosol medium (thermal and chemical condition, static charge on the particles, abrasive particles, moisture in the gas stream, etc.), filtration requirement (particle capturing efficiency, pressure drop and cleaning) and equipment consideration. The particle capturing mechanism by fabric filters based on reverse jet cleaning is well researched. A number of studies reveal the effect of fabric structural parameters on its filtration characteristics. Further a number of attempts have been made to improve filtration efficiency with reduced pressure drop and better cake release performance. However, it is worth mentioning that the above studies becoming rudimentary since pulse jet filter has become an attractive option of particulate collection utilities. Pulse jet cleaning is a technique whereby a short, periodic, high pressure burst of air is fired into the clean side of the fabric. The particles are dislodged and the pressure drop falls to an acceptable level. Pulse jet filtration can meet the stringent particulate emission limits regardless of variation in the operating conditions. The cleaning device is less expensive than other type of mechanism and requires considerable less space. Other merits of pulse jet fabric filters are high collection efficiency, on line cleaning application and outside collection which allows the bag maintenance in a clean and safe environment. The increasing adoption of pulse jet filters for control of process emissions and recovery of utility dusts has stimulated research on many aspects of their operation. Despite their wide applications, the functioning of fabric filters is poorly understood. The operating condition is usually specified by the manufacturer at the time of commission. In practice, however, the filtration process frequently undergoes changes in the operating condition caused by the disturbances due to the variation in the concentration of dirty air, variation in particle size, etc. Due to the complex characteristics inherent in the filtration and cleaning process of pulse jet bag houses, the proper understanding of the role of fabric construction on filtration is very important. Generation of such knowledge would represent an important first step towards designing fabric with improved barrier properties.
Project aims and objectives The separation of solids from fluids by textile filter media is an essential part of countless industrial process, contributing to purity of product, saving in energy, improvement in process efficiency, recovery of precious materials and general improvements in pollution control. The dust may create environmental pollution problems or other control difficulties caused by their toxicity, flammability and possible risk of explosion. The particles in question may simply require removal and be of no intrinsic value or alternatively may constitute part of a saleable product, for example, sugar or cement. Among several techniques, the most efficient and versatile is the fabric collector, especially when processing very fine particles. Fabric filtration under pulse jet situation becoming very common in industrial bag house filtration. However, there is lack of study on the role of fabric under the said circumstances. It is also important to develop the fabric suitable for pulse-jet filtration. Therefore, we are having the following objectives: . Understanding the mechanism of filtration under pulse-jet situation. .
Studing the effect of geometrical parameter of nonwoven fabric on filtration performance.
.
Design and development of suitable nonwoven fabric under pulse-jet situation.
Research deliverables (academic and industrial) Initially the work will be concentrated on studing the impact of structural parameter of nonwoven fabric on its filtration performance in case of cement dust. This will lead to the development of suitable nonwoven fabric under pulse-jet situation in cement industry. At a later stage the work will be extended for optimising fabric design for other dust particles.
59
Publications Not available.
Lodz, Poland Technical University of Lodz, Department of Clothing Technology, Zeromskiego 116 Str, 90-924 Lodz. Tel: 0048 042 631 33 21; Fax: 0048 042 631 33 20; E-mail:
[email protected] Principal investigator(s): Iwona Frydrych Research staff: Renata Krasowska
Influence of work conditions of the disc take-up on characteristics of lockstitch Other Partners: Academic
Research register
Industrial
None None Project started: 20 May 2005 Project end date: 19 May 2008 Project budget: 49,100 PLN Source of support: Ministry of Science and Higher Education Keywords: Lockstitch, Lockstitch machine formation zone (STS), Thread demand in STS, Thread feed by the take-up A subject of projekt presents an elaboration of original model simulating a thread movement in the zone of stitch creaction, on the basis of which a mechanizm of tke-up disc will be built. It will be evaluated in experimental research. The regulating point at this type of take-up will be presented as a novelty. According to this a possibility of reacting of the machine operator on the changes of sewing thread properties will and useful properties of lockstitch will be created. It requires elaborating the set of criteria assessing an action of this type of take-up disc.
Project aims and objectives The aim of the project is testing the working conditions of the disc take-up disc of the sewing thread taking into consideration the development of its construction of regulation points in the relation to existing solutions.
IJCST 20,6
60
Research deliverables (academic and industrial) New construction of mechanizm of take-up disc should create possibilities to adjust the parameters of its construction to thread sewing parameters and useful properties of the lockstitch. Publications 1. R. Krasowska and I. Frydrych, “Possibilities of modelling the control conditions of thread by the disc take-up in the lockstitch machine”, Fibres & Textiles in Eastern Europe 1/2006. 2. R. Krasowska and I. Frydrych, “Formation of the thread control curve by the disc take-up in the lockstitch machine”, Research Journal of Textile and Apparel, 2006
Manchester, UK University of Manchester, Textiles and Paper, School of Materials, The University of Manchester, Sackville Street, Manchester M60 1QD. Tel: 0161 306 4113; E-mail:
[email protected] Other Partners: Academic Industrial None Project started:
None Project end date: September 2008
The current September 2008 projects are: . Mathematical modelling of complex 2D/3D weaves. .
Mechnical modelling of 2D/3D woven composites.
.
3D compoasites against impact. Riot helmet research.
. . . .
Ballistic body armour. Riot body armours for female officers. 3D weaving technology on conventional looms.
Project aims and objectives Research deliverables (academic and industrial) Publications Not available.
Maribor, Slovenia University of Maribor, Faculty of Mechanical Engineering, Smetanova ulica 16, SI-2000 Maribor, Slovenia. Tel: +386 2 220 7960; Fax: +386 2 220 7990; E-mail:
[email protected]
Principal investigator(s): Prof. dr.sc. Jelka Gersˇak Research staff: Research Unit Clothing Engineering
Research register
Clothing engineering and textile materials Other Partners: Academic
Industrial
None None Project started: 1 January 2004 Project end date: 31 December 2008 Project budget: 79.295 ECU for 2008 Source of support: Slovenian Research Agency Keywords: Clothing, Fabric, Fabric mechanics, Behaviour, Comfort, Prediction The research programme was carried out in the frame of three interdependent and closely thematically connected complexes: . . .
basic investigations of mechanics of fabric as complex textile structures; modelling of the behaviour of complex textile structures; and characterisation of the parameters of thermophysiological comfort and design of a model for heat transfer from the human’s body to the environment.
In the frame of the first complex of activities we defined the qualitative characteristics of garment appearance. Here, we defined fabric behaviour from the point of view of mechanics during remodelling from a plane into a 3D form of a garment and define the importance of the parameters related to mechanical and physical properties of fabrics considering previously defined factors of garment appearance quality, defined with marks from 1 to 5. Taking into account the defined principles and factors related to the garment appearance quality we designed a system for predicting garment appearance quality. The developed system, entitled InSiNaKVO, represents an innovative and first of its kind tool for objective evaluation and prediction of a level of garment appearance quality. The essence of the system, which is based on defined relationships between the parameters of mechanical and physical properties of fabrics and level of the garment appearance quality, is prediction of garment appearance quality taking into account mechanical and physical properties of fabrics, built-in the garment, which enables: . .
. .
objective evaluation of the level of garment appearance quality; prediction of five important factors of the level of garment appearance quality: garment fall, 3D garment form, fit; quality of produced seams and garment appearance quality as a whole; planning of the required level of garment appearance quality;
prediction of the level of garment appearance quality for concrete fabrics in real production systems; and . planning of quality parameters related to mechanical and physical properties of fabrics for desired level of garment appearance quality. In the frame of the second complex of the research, i.e. modelling of the behaviour of complex textile structures, we continued with the researches related to complex .
61
IJCST 20,6
62
deformation of fabrics underlining the study of rheological model. The relaxation phenomena have been studied form the point of view of mechanical multi-component models as tools for describing deformational and relaxation phenomena in the fabrics. Based on extensive research work we have established that there are significant differences between the behaviour of conventional and elastic fabrics. Researches have shown that multi-component models are the most suitable for explaining the relaxation stress in fabrics containing elastane yarns because of greater number of parameters. Special attention was given to simulation of draping using the finite elements method, where the fabric model is based on rheological parameters and mass points method, which is derived form the linear rheological model. Important here is a study of the relationship between the computer based simulation of fabric draping used as material model and parameters of mechanical properties of fabrics. In the continuation we will focus above all on numerical simulation of fabric draping, based on mechanical multi-component models. Our expectations are to come closer to real properties of fabrics, which is a starting point for virtual representation of garment models. The third complex of the research is directed to the characterisation of thermophysiological comfort. For this purpose we have carried out extensive researches related to transmission and transition of heat, material properties of textile materials and thermophysiological comfort at garment wearing for business apparel and for an example of technical application (effect of the type of material for automotive seats and type and construction of a bed on thermophysiological comfort of a user). A special attention has been given within the frame of these researches to the study of material properties and specific requirements related to fabrics for men’s business suits (conventional and functional (phase-changing) fabrics). The results of research have shown that the business clothing system with built-in phase-changing material does not influence significantly the thermal regulation of a body. Clothing systems with built-in phase-changing material showed a short-time thermal effect, seen as small increase or decrease of skin temperature, depending on climatic conditions.
Project aims and objectives The main goals of the research programme Clothing engineering and textile materials, which is based on complex researches of fabric mechanics, material properties and relationship between the material properties of textile fabrics and parameters of thermophysiological comfort at garment wearing, are as follows: . to define the non-linear behaviour of flat textile products as complex structures at lower stresses; . . .
.
.
to design a model for prediction of garment appearance quality; to define the parameterisation of textile materials and products; to enable a virtual representation of different garment styles, based on real properties of textile materials; to set-up the characterisation of parameters related to thermophysiological comfort; and to develop the numerical simulation of heat transfer between the human’s body and garment or other textile product seen as a system and environment.
Research deliverables (academic and industrial) The research programme is systematically designed so as to result in the knowledge that could be used in establishing and engineering concept of garment and/or other textile product planning particularly the so-called “knowledge-based products”. The results have partially been made applicable through the development of the InSiNaKVO system, the main characteristics of which are as follows: (a) knowledgebased designing and predicting garment appearance quality degree, using known fabric mechanical and physical properties (b) planning fabric mechanical and physical properties so as to get the target garment quality level, and (c) tools to simulate the degree of garment quality. Publications Gersˇak, J. (2004), “Study of relationship between fabric elastic potential and garment appearance quality”, International Journal of Clothing Science and Technology, Vol. 16 Nos 1/2, pp. 238-51. Zavec Pavlinic´, D. and Gersˇak, J. (2004), “Vrednovanje kakvoc´e izgleda odjec´e”, Tekstil, Vol. 53 No. 10, pp. 497-509. Gersˇak, J., Sˇajn, D. and Bukosˇek, V. (2005), “A study of the relaxation phenomena in the fabrics containing elastane yarns”, International Journal of Clothing Science and Technology, Vol. 17 Nos 3/4, pp. 188-99. Tama´s, P., Gersˇak, J., Hala´sz, M. (2006), “Sylview 3D Drape Tester – New system for measuring fabric drape”, Tekstil, Vol. 55 No. 10, pp. 497-509. Zavec Pavlinic´, D., Gersˇak, J., Demsˇar, J. and Bratko, I. (2006), “Predicting seam appearance quality”, Textile Research Journal, Vol. 76 No. 3, pp. 235-42. Gersˇak, J. and Marcˇicˇ, M. (2007), “Development of a mathematical model for the heat transfer of the system man – clothing – environment”, International Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 234-41. Celcar, D., Meinander, H. and Gersˇak, J. (2008), “A study of the influence of different clothing materials on heat and moisture transmission through clothing materials, evaluated using a sweating cylinder”, International Journal of Clothing Science and Technology, Vol. 20 No. 2, pp. 119-30. Celcar, D., Meinander, H. and Gersˇak, J. (2008), “Heat and moisture transmission properties of clothing systems evaluated by using a sweating thermal manikin under different environmental conditions”, International Journal of Clothing Science and Technology, Vol. 20 No. 4, pp. 240-52. Gersˇak, J. and Zavec Pavlinicˇ, D. (2005), “System of predicting garment appearance quality”, Book of Proceedings of the 34th Textile Research Symposium at Mt. Fuji, Susono City, Japan, August 9-11, pp. 51-3. Gersˇak, J. and Marcˇicˇ, M. (2006), “Thermophysiological comfort”, Book of Proceedings of the 1st International Workshop Design – Innovation – Development, Iasi, 24-27 July, pp. 143-7. Gersˇak, J. and Zavec Pavlinic´, D. (2006), “Garment appearance quality as aesthetic function of clothes”, Annals of DAAAM for 2006 & Proceedings of the 17th International DAAAM symposium “Intelligent manufacturing & Automation: ‘“Focus on mechatronics and robotics’”, Vienna, Austria, 8-11 November. Gersˇak, J. (2006), “Thermophysiological comfort of the driver”, Book of papers of 2nd International material conference TEXCO, Ruzˇomberok, Slovakia, 17-18 August. Gersˇak, J. and Marcˇicˇ, M. (2006), “Modeling the heat transfer of the system man – clothing – environment”, Book of Proceedings of 3rd International Textile, Clothing & Design Conference ITC&DC, Dubrovnik, Croatia, 8-11 October. Zavec Pavlinic´, D. and Gersˇak, J. (2006), “Garment appearance quality: from subjective estimation to objective evaluation”, Book of Proceedings of 3rd International Textile, Clothing & Design Conference ITC&DC, Dubrovnik, Croatia, 8-11 October.
Research register
63
IJCST 20,6
64
Zavec Pavlinic´, D. and Gersˇak, J. (2007), “The advanced method for the evaluation of garment appearance quality grade”, Proceedings of 7th Annual Textile Conference by AUTEX, Tampere, Finland, 26-28 June. Gersˇak, J. and Marcˇicˇ, M. (2007), “A novel approach to improve the thermophysiological clothing comfort”, Book of Proceedings AGILTex Design – 2nd International Workshop Design – Innovation – Development, Ias¸ i, Romania, 31st August – 3rd September, pp. 102-12. Gersˇak, J. and Zavec Pavlinic´, D. (2007), “Objective evaluation of garment appearance”, Book of Proceedings of 5th International Conference Innovation and Modelling of Clothing Engineering Processes IMCEP 2007, Faculty of Mechanical Engineering, October 10-12, Maribor, Slovenia, pp. 34-9. Zavec Pavlinic´, D. and Gersˇak, J. (2007), “Application of the system for the prediction of garment appearance quality”, Book of proceedings of 5th International Conference Innovation and Modelling of Clothing Engineering Processes IMCEP 2007, Faculty of Mechanical Engineering, October 10-12, Maribor, Slovenia, pp. 81-8. Celcar, D., Gersˇak, J. and Meinander, H. (2007), “The influence of environmental conditions on thermophysiological wear comfort of business clothing”, Book of Proceedings of 5th International Conference Innovation and Modelling of Clothing Engineering Processes IMCEP 2007, Faculty of Mechanical Engineering, October 10-12, Maribor, Slovenia, pp. 141-9. Plazl, K. and Gersˇak, J. (2007), “The impact of fabric type and construction of bedding matresses on thermal physiological comfort of the user”, Book of proceedings of 5th International Conference Innovation and Modelling of Clothing Engineering Processes IMCEP 2007, Faculty of Mechanical Engineering, October 10-12, Maribor, Slovenia, pp. 220-7
Nottingham, UK University of Nottingham, School of Mechanical, Materials & Manufacturing Engineering, University Park, Nottingham, NG7 2RD. Tel: 0115 9513779; Fax: 0115 9513800; E-mail:
[email protected] Principal investigator(s): C.D. Rudd, A.C. Long, R. Brooks, I.A. Jones, S.J. Pickering, N.A. Warrior, M.J. Clifford, C.A. Scotchford, G.S. Walker Research staff: H. Lin, L. Harper, RRH Naqasha
Platform grant: processing and performance of textile composites Other Partners: Academic
Industrial
None None Project started: 1 February 05 Project end date: 31 January 09 Project budget: £445k Source of support: EPSRC Keywords: Textile composites, Unit cell analysis, TexGen Our current research portfolio is centred on the processing of polymer matrix composites with a growing emphasis on modelling and simulation. Given our high level of interest in textile-based composites and their growing importance in the field, we wish to introduce a common platform for our modelling studies based on our formalised textile
generator (TexGen). Textile modelling provides a launchpad for downstream simulation of processing, damage mechanics and environmental performance. The functionality of our existing TexGen software will be extended and coupled to materials models for simulation of each of the above aspects of physical behaviour. Simulation of each of the physical processes will be enhanced by a common, interchangeable geometric definition of the textile structure within the rigid composite. This will enable a rapid understanding of fabric architecture effects to be built and the approach has excellent potential for application to other physical problems which relate to rigid and flexible composites or technical textiles. The platform grant application seeks continuity of support for key workers during the period of this development. Further details are available at: www.textiles.nottingham.ac.uk
Project aims and objectives .
.
.
Implement an approach based on textile modelling throughout our research portfolio, integrating the multiple streams of processing, energy management, biomedical applications and textile modelling. Develop a series of downstream models relating to: three-dimensional permeability, formability (including shear compliance), static mechanical properties, damage mechanics and residual property estimation, diffusion and environmental degradation. Exploit the potential of the platform grant to raise our international profile, develop strategic links with other leading groups, and enhance our technology transfer activities.
Research deliverables (academic and industrial) Publications Not available.
Ohtsu, Japan SCI-TEX(Consultant), 12-15, Hanazono-cho, Ohtsu-city, 520-0222 Japan. Tel: 81-77-572-3332; Fax: 81-77-572-3332; E-mail:
[email protected] Principal investigator(s): Tatsuki Matsuo
Fiber assembly structures for technical textiles in terms of function and end-use Other Partners: Academic None Project started: October 2007 Project budget: 400,000yen
Industrial None Project end date: September 2008
Research register
65
IJCST 20,6
66
Source of support: Own Keywords: Fiber assembly structure, Function, End-use, Technical textiles, Nonwovens, Composites General criterion for selecting the structure of technical textiles is the ratio(performance/cost). Verasatility in the structure of technical textiles is much wider than apparel textiles. Objective of this project is to find guiding princiiples for designing the optimum structure of technical textile in terms of their required function and cost.Several fundamental analyses on the fiber assembly structure and function of technical textiles are investigated. Directive fiber assembly structures suitable for required functions are analysed based on the analytical results of the investigation. The results are tried to be applyied to nonwovens and composites.
Project aims and objectives Research deliverables (academic and industrial) Publications Not available.
Ohtsu-city, Japan SCI-TEX, 12-15, hanazono-cho, Ohtsu-city, 520-0222 Japan. Tel: 81-77-572-3332; Fax: 81-77-572-3332; E-mail:
[email protected] Principal investigator(s): Tatsuki Matsuo
Propagation of knowledge on new textile science and technology Other Partners: Academic
Industrial
None None Project started: Project end date: on going Keywords: Advanced technical textiles, Knowledge propagation The importance of advanced technical textiles has increased in the textile industry of developed countries. In addition, R&D on nano-technologies and electric-textiles are now intensively carried out. In this situation, propagation of knowledge on new textile science and technology must be meaningful. This project is being conducted individually by T. Matsuo through symposium lectures, journal articles and monographic books.
Project aims and objectives Research deliverables (academic and industrial) Publications Not available.
Pisa, Italy University of Pisa, Via Diotisalvi, 2, 56126 Pisa, Italy. Tel: +39 0502217053; Fax: +39 0502217051; E-mail:
[email protected] Principal investigator(s): Prof. Danilo De Rossi Research staff: Prof. Bruno Neri; Ing. Alessandro Tognetti; Ing. Enzo Pasquale Scilingo; Ing. Federico Carpi; Ing. Antonio Lanata`
PROETEX: Protection E-textiles: micro-nano structured fiber systems for emergency-disaster wear Other Partners: Academic Consiglio Nazionale delle Ricerche – INFM, Technical University of Lodz, Ghent University – Department of Textiles, University of Pisa, Dublin City University, Institut National des Sciences Applique´es de Lyon
Industrial Smartex srl, Milior, Sofileta SAS, Thuasne, Commissariat a` l’Energie Atomique – “CEA”, CSEM Centre Suisse d’Electronique et de Microtechnique SA, Sensor Technology and Devices Ltd, Steiger S.A., Philips GmbH, Zweigniederlassung Forschungslaboratorien, Ciba Spezialita¨tenchemie AG, Diadora Invicta SpA, iXscient Ltd, Zarlink Semiconductor Limited, Brunet-Lion SAS, Brigade de Sappeurs Pompiers de Paris, European Centre for Research and Training in Earthquake Engineering, Direction de la De´fense et de la Se´curite´ Civiles
Project started: February 2006 Project end date: January 2010 Project budget: University of Pisa: e780,443; Total: e12,792,242 (Requested: e8.100.000) Source of support: European Commission ProeTEX will develop integrated smart wearables for emergency disaster intervention personnel, improving their safety, coordination and efficiency and for injured civilians, optimising their survival management. This core application area, which is of significant societal importance in itself, will drive a wide range of key technology developments, building on current and past EU and national projects and the commercial activities of partners, to create micro-nano-engineering smart textile systems – integrated systems (fabrics, wearable garments) using specifically fibrebased micronano technologies. These are capable of being combined into diverse products addressing this core application area but also a wide range of other markets from extreme sports, through healthcare to transportation maintenance and building workers. The industrial partners can address these markets. Fiber systems can integrate sensors, actuators, conductors, power management, and the emergency disaster personnel smart garment will, within a wireless ambient planning and managing environment, progressively enhance and integrate fiber systems for:
Research register
67
IJCST 20,6
68
.
continuous monitoring of life signs (biopotentials, breathing movement, cardiac sounds);
.
continuous monitoring biosensors (sweat, dehydration, electrolytes, stress indicators);
.
pose and activity monitoring;
.
low power local wireless communications, including integrated fiber antennae;
.
active visibility enhancement, light emitting fibers;
.
internal temperature monitoring using fiber sensors;
.
external chemical detection, including toxic gases and vapours; and
.
power generation – photovoltaic and thermoelectric and power storage.
The technological base developed will concentrate on smart fibers/e-textiles, but the IP will combine these where appropriate with “conventional” microsystems (such as accelerometers, gyros, microcontrollers and wireless chips).
Project aims and objectives The central IP goal is to develop an integrated set of functional garments for emergency disaster personnel, such as firefighters and paramedics, plus systems for injured civilians. These will be produced using both enhanced and novel fibre based micronanosystems, whose development will extend the state of the art in this area. The project will roll out a sequence of progressively more capable integrated wearable systems for emergency disaster intervention personnel and injured civilians. Thus, overall the IP will: .
Progress the fundamentals of fibre-based sensor, processing, communications and power management systems.
.
Integrate these fibre-based capabilities into functional knitted or woven wearable garments.
.
Produce fully capable integrated communicating, sensor wearables, using additional ‘conventional’ systems where necessary.
.
Test their match of user needs and requirements in a lab-based setting.
.
Demonstrate their function in a real-world application in a number of field trials.
Scientific objectives: (1) Develop a multifunctional garment integrating an increasingly ambitious set of sensors and energy harvesting and storage which is reliable, robust, easy to wear and capable of manufacture. (2) Into this garment: Design, test and integrate a bioelectrical heart rate monitor into whole skin contact garment interface; Design, test and Integrate a cardiac sound monitor; Integrate sensor breathing monitor and ensure that signal conditioning and processing results in successful way. (3) Develop fibre and new textile based technological solutions, with reliable functionality, capable of integration into wearable garments covering the following set of technological area capabilities:
.
Monitor bioelectrical potential.
.
Sensing breath movements.
.
Sensing posture and movement.
.
Biochemical sensing, specifically determination of dehydration status.
.
Sensing core temperature.
.
Acting as local communications antennae.
.
Sensing external toxic gases/chemicals, including CO.
.
Generating local energy using thermoelectric generation.
.
Generating local energy using photovoltaic processes.
.
Storing energy using Li-Ion textile batteries.
Technical objectives: .
Develop and adapt textile manufacturing processes to these new active fibres and layers (weaving, knitting, coating, laminating) but also innovate in terms of clothes conception to optimise the assembly step regarding interconnection needs for e-textile garment.
.
Develop and test a multifunctional (inner and outer) garment integrating an increasingly ambitious set of sensors and energy harvesting and storage which is reliable, robust, easy to wear and capable of manufacture for both intervention people and injured civilians.
.
The inner and out garment will include an adapted set of functionalities based on the developed technologies. As example first inner garment could integrate bioelectrical heart rate monitor, cardiac sound monitor, strain sensor breathing monitor inner temperature measurement and ensure that signal conditioning and processing results in successful and robust physiological monitoring. Energy generated by the heat (thermoelectricity) and the movement (piezoelectricity) of the of the wearer. Outer garment will typically include toxic gas measurement, external temperature; motion and position monitoring, data transmission system, energy could be provided by photovoltaic external layer and textile Li-Ion batteries.
.
Realize field trial of the instrumented garment for technological validation.
Research deliverables (academic and industrial) The key deliverables will be: .
An inner garment for emergency disaster personnel, monitoring the health of the user through vital signs, biochemical parameters, activity and posture, generating and storing own power and communicating locally with other wearables and relaying through (D).
.
An outer garment for emergency disaster personnel, measuring potential environmental insults (temperature, CO, other toxic gases), sensing posture and movement of the wearer and offering improved visibility, generating and storing its own power communicating locally with other wearables and relaying through (D).
Research register
69
IJCST 20,6
70
.
An under-garment jerkin or chest band for injured civilians (closely related to (A)) monitoring their health, generating and storing its own power and communicating locally, relaying information via (E). Victims monitoring measures will include: body temperature; cardiac pulse; respiration rate; ECG; percutaneous CO saturation; percutaneous O2.
.
A portable unit for the emergency disaster personnel, communicating with A, B and C, but offering additional ‘conventional’ microsystem, providing both local and long range communication (acting as a relay for A, B and C), including some specific sensors not easily integrated into (B), plus accelerometers, gyros and GPS to enable high accuracy position and movement determination. This device should allow data entry and displays/alarms.
A portable unit for injured civilians, to include data relay capability and INS/GPS but no data entry or display. Some kind of integrated alarm or indicator to give the overall civilians health status. A simple user input, such as panic button, may be required. Publications Not available.
Pisa, Italy University of Pisa, Via Diotisalvi, 2, 56126 Pisa, Italy. Tel: +39 0502217053; Fax: +39 0502217051; E-mail:
[email protected] Principal investigator(s): Prof. Danilo De Rossi Research staff: Post doc researchers, PhD students, post graduate students.
FLEXIFUNBAR: Multifunctional barrier for flexible structure (textile, leather and paper) Other Partners: Academic Brunel University, Pisa University, Centro Di Cultura per l’ingegneria delle Materie Pastiche, Ghent University, Queen’s University of Belfast, DWI – Aachen, Institut Pasteur de Lille, Institute of Natural Fibres – Poznan
Industrial Alan, Amkey Management, Annebergs, Arjo Wiggins, Basilius, Calsta, CEI, Centexbel, Centro Tecnologico do Calcado, Clotefi, Clubtex, CREPIM, Curtumes Aveneda, Devan, DG Tec, Duflot, ECCO, Gleittechnik, Europlasma, IFTH, IMP Comfort, INCA, IQAP, Lauffenmu¨ hle, Linificio, Nabaltec, Nylstar, Patraiki, Procotex, Siamidis, Sinterama, Subrenat, Sveriges Provnings, Telice, Thrakika Ekkokistiria, Traitex, VTT, Wellman International
Project started: 1 October 2004 Project end date: 30 November 2008 Project budget: Total amount: e6,438,995 (University of Pisa: e343,000) Source of support: European Commission Keywords: Multifunctional, Barrier, Textile, Fiber, Flexible, Leather, Paper All citizens are permanently protected by flexible structures with barrier noise and thermal insulation, shield against electrostatic or electromagnetic phenomena, filtration of dust or insects. The application of flexible structure is very large thanks to their easy adapting properties and shape. Nevertheless, they will maximise the level of safety in building, transportation and to ensure the well-being of European citizens. The flexible structures, generally based on paper, leather or textile are usually treated to serve only one barrier effect.Ires a whole re-design of flexible structure functions that is the main purpose of FLEXIFUNBAR. For instance to prevent from all external aggressions in hazardous atmosphere, flexible structure must provide at least barrier effects. The ultimate goal of flexifunbar initiative is to develop innovative generation of hybrid multi barrier-effects materials, based on multi layer complex structures and funcionnalisation of micro and nanostructures.The development of such materials covers a large range of applications: . Transportation: filter, thermal and acoustic insulation panel, pollutant detectors. .
Home and building: wall covering, home furniture, carbon monoxide detectors, antibacterial mattress, electromagnetic insulation panels.
.
Health: protection of people against insects, protective clothing for military and defense, hygiene mask, operation area.
Project aims and objectives The innovation of Flexifunbar lies in the principle of associating in one same material several functionalities: Heat insulation, acoustic insulation, shielding against electromagnetic waves, anti odours, anti bacterial, flame retardancy. . . The objective of Flexifunbar is to develop and promote multi-functional flexible structure for use in many multisectorial industrial applications in the health field as well as in the building construction and transportation industrie.
Research deliverables (academic and industrial) New fiber, textile, leather and paper samples with multifunctional barrier properties. Publications Not available.
Research register
71
IJCST 20,6
72
Pisa, Italy “E. Piaggio” Centre – University of Pisa, Via Diotisalvi, 2, 56126 Pisa, Italy. Tel: +39 0502217053; Fax: +39 0502217051; E-mail:
[email protected] Principal investigator(s): Prof. Danilo De Rossi Research staff: Prof. Roger Fuoco; Prof. Arti Ahluwalia; Dr Fabio DI Francesco; Dr Thoas Schafer
BIOTEX: Bio-sensing textiles to support health management Other Partners: Academic Dublin City University(Ireland), University of Pisa (Italy)
Industrial CSEM Centre Suisse d’Electronique et de Microtechnique SA (Switzerland), CEA Commissariat a` l’Energie Atomique (France), Smartex s.r.l. (Italy), Thuasne (France), Penelope SpA (Italy), Sofileta (France)
Project started: 1 September 2005 Project end date: 29 February 2008 Project budget: Total amount: e3,108,029 (eligible cost) – Requested EC contribution: e1,900,000 University of Pisa quote: e255,750.00 Source of support: European commission Integration of health monitoring tools into textiles brings the benefits of safety and comfort to the users. Instrumented clothes will provide remote monitoring of vitals signs, diagnostics to improve early illness detection and metabolic disorder and benefits to the reduction on medical social costs to the citizen. Ambulatory healthcare, isolated people, convalescent people and patients with chronic diseases are addressed. To date, developments in that field are mainly focused on physiological measurements (body temperature, electro-cardiogram, electromyogram, breath rhythm, etc.) with first applications targeting sport monitoring and prevention of cardiovascular risk. Biochemical measurements on body fluids will be needed to tackle very important health and safety issues.
Project aims and objectives The BIOTEX project aims at developing dedicated biochemical-sensing techniques compatible with integration into textile. This goal represents a complete breakthrough, which allows for the first time the monitoring of body fluids via sensors distributed on a textile substrate and performing biochemical measurements. BIOTEX is addressing the sensing part and its electrical or optical connection to a signal processor. The approach aims at developing sensing patches, adapted to different targeted body fluids and
biological species to be monitored, where the textile itself is the sensor. The extension to whole garment and the integration with physiological monitors is part of the roadmap of the consortium. Textiles for applications in health monitoring are becoming a major theme for the citizen’s healthcare and safety since they allow: .
simultaneously comfort and monitoring (for safety and/or health);
.
non-invasive measurements, no laboratory sampling; continuous monitoring during daily activity of the person;
.
possible multi-parameter analysis and monitoring; and distributed sensing thanks to access to 90% of the body surface if integrated on clothing. Societal issues such as ambulatory healthcare, isolated elderly or disabled patient’s care may be tackled by these techniques. Moreover, non-invasive and continuous monitoring of people in a critical state is more and more needed, e.g. in emergency services, for heavily burnt patients and safety, e.g. exposed personnel like fire-fighters. At the present stage, health-monitoring systems using electronic textiles are mainly targeting applications based upon physiological parameter measurements, such as body movements or electro-cardiogram. To open a dramatically wider field of applications, biochemical measurements on body fluids (blood, sweat, urine) will be needed. At the present time, biochemical analysis systems compatible with integration into clothing are unfortunately lacking. This is a major drawback for instance in the case of sweat analysis which is potentially very rich in health related information. However, such analysis is hardly performed today because of the difficulty to sample sweat in sufficient quantity. Only a real textile sensor embedded in a garment through textile techniques will allow direct collection of sweat and a large body surface; moreover lower fabrication costs are expected. For blood analysis, the main interest will be to avoid invasive sampling and to allow continuous analysis. BIOTEX aims at the development of technologies to fulfil these needs. . .
Research deliverables (academic and industrial) New textile integrated systems for biochemical parameters detection. Publications Not available.
Shrewsbury, UK Rapra Technology Ltd, Shawbury, Shrewsbury, Shropshire SY4 4NR. Tel: +44 1939250383; Fax: +44 1939 251118; E-mail:
[email protected] Principal investigator(s): Chris Hare Research staff: R. Venables, S. Wallace
New classes of composite engineering materials from renewable sources
Research register
73
IJCST 20,6
Other Partners: Academic Upper Austria Research, Lulea University of Technology, Wroclaw University of Technology
74
Industrial Fraunhofer ICT, Risoe, Gaiker, Celabor, VTT, APC Composites, BAFA, Transfurans chemicals, PJH, Chemont, Tecnaro, MERL, Net composites, Tehnos, Griffner, MEDOP, Haidlmair, Fiedler, Ekotex, National Institute of wood Project end date: 1 October 2008
Project started: 1 April 2005 Project budget: e6.5 million Source of support: EU, Integrated Project, FP6
Future product design requires sustainable processes and eco-innovation in material development for engineering applications. The innovative approaches use new engineering materials – biocomposites and their development has to be knowledgebased, whereas predominant issues are resource saving, variability in properties and functionality, light weight, low costs and eco-efficiency in all stages of the product life cycle. The main objective of this project is to obtain a breakthrough for SMEs on the development and use of engineering thermoplastic and thermosetting materials mainly from natural resources, like lignin from the paper industry and from the High Presure Hydrothermolyses (HPH) process, other biopolymers (here referred as biopolymers:, e.g. Polylactide, Polyhydroxy-butyrate, Starch), furan resins, woven and non-woven cellulose fibres and fibre mats to final model products. The technical work programme will comprise the complete technical path from the input of natural raw materials (fibres, polymers and natural additives) to the output of final top quality engineering composite materials and model products (e.g. housings for electronic equipment, car front end interiors, glass frames, etc.) with an environmentally friendly life cycle. In parallel, there are activities concerning standardisation of characterisation and test procedures and quality control. Demonstration by the model products supporting the dissemination and the exploitation of results will exhibit the benefits of the materials and deliver a first input to material databases. An integrated concept of sustained skill and education of staff and students will provide routines and access to the material data. It includes the most interesting approaches of all current developments for engineering biocomposites. Innovative additives will provide flame retardancy and colouring.
Project aims and objectives To produce a range of ‘hi tech’ composite panels using a variety of natural fibres and natural resins.
Research deliverables (academic and industrial) Many deliverables in the project, some of which are: Reports on: Raw material characterization, data and tolerances; Compounding of materials; Test data; Safety and emissions; Environmental Benefits; Economic
Evaluation; Sample tools and parts; Demonstration Parts; Training, exploitation and dissemination. Publications Tama´s Pe´ter and Hala´sz Marianna, “3D body modelling in clothing design”, IMCEP 2003, 4th International Conference, 9-11. Oktober 2003, Maribor, Slovenia, ISBN 86-435-0575-7, pp. 64-8. L. Kokas Palicska; J. Gersak and M. Hala´sz, “The impact of fabric structure and finishing on the drape behavior of textiles”, AUTEX 2005, 5th World Textile Conference, Portorozˇ, Slovenia, 27-29 June 2005, ISBN 86-435-0709-1, pp. 891-97. P. Tama´s; M. Hala´sz and J. Gra¨ff, “3D dress design”, AUTEX 2005, 5th World Textile Conference, Portorozˇ, Slovenia, 27-29 June 2005, ISBN 86-435-0709-1, pp. 436-41. J. Kuzmina; P. Tama´s; M. Hala´sz and G. Gro´f, “Image-based cloth capture and cloth simulation used for estimation cloth draping parameters”, AUTEX 2005, 5th World Textile Conference, Portorozˇ, Slovenia, 27-29 June 2005, ISBN 86-435-0709-1, pp. 904-9. L. Szabo´ and M. Hala´sz, “Automatic determination of body surface data”, AUTEX 2005, 5th World Textile Conference, Portorozˇ, Slovenia, 27-29 June 2005, ISBN 86-435-0709-1, pp. 715-20. L. Kokas Palicska and M. Hala´sz, “Analysing of draping properties of textiles”, IN-TECH-ED’05, 5th International Conference, 8-9. September 2005, Budapest, ISBN 963 9397 06 7, pp. 133-38. O. Nagy Szabo´; P. Tama´s and M. Hala´sz, “Garment construction with a 3 dimension designing system”, IN-TECH-ED’05, 5th International Conference, 8-9. September 2005, Budapest, ISBN 963 9397 06 7, pp. 348-257. J. Kuzmina; P. Tama´s; M. Hala´sz and G. Gro´f, “Image-based cloth capture and cloth simulation used for estimation cloth draping parameters”, IN-TECH-ED’05, 5th International Conference, 8-9. September 2005, Budapest, ISBN 963 9397 06 7, pp. 358-65.
Zagreb, Croatia Faculty of Textile Technology, Prilaz baruna Filipovica 30, 10 000 Zagreb, Croatia. Tel: +385 1 37 12 577; Fax: +385 1 37 12 533; E-mail:
[email protected] Principal investigator(s): Prof. Zenun Skenderi, PhD Research staff: Prof. Miroslav Srdjak, PhD; Prof. Momir Nikolic´, PhD; Prof. Alka Mihelic´-Bogdanic´, PhD; Bozˇo Tomic´, MSc; Vesna Marija Potocˇic´ Matkovic´, MSc; Ivana Salopek, MSc; Dragana Kopitar, BSc.
Multifunctional technical non-woven and knitted textiles, composites and yarns Other Partners: Academic None
Project started: 1 January 2007
Industrial Cˇateks d.d. Cˇakovec, Croatia, Regeneracija non-woven and carpets j.s.c., Zabok, Croatia Project end date: 1 January 2010
Research register
75
IJCST 20,6
76
Source of support: Ministry of Science, Education and Sport, Republic of Croatia Keywords: Yarns, Knitted and nonwoven fabrics, Structures, Properties, Thermal and vater-vapour resistance properties Further dislocation of the textile production from deveoped countries into Asia is a basic characteristic for the world textile industry today. In the field of technical textiles profound resistance is felt against relocation. An increase in the production of technical textiles is recorded due to a permanent expansion of the application range. It is used in: transportation, industry, medicine, hygiene, household, garment industry, agriculture, fishing trade, civil engineering, sport, safety, ecology, etc. Nonwovens make the most significant contribution to the development of technical textiles. Over last decades the technology of nonwovens production has experienced a rapid development, and the production of late years has registered an annual increase of approx. 10 per cent. A significant application range for technical textiles or geotextiles is civil engineering, in particular road building. In addition to woven, knitted and similar structures, nonwovens play a predominant role with a share of approx. 75 per cent in 2005. The most important functions of geotextiles are: separation of weak soil, reinforcement of soil or elements of building structures, filtration and drainage. Geotextile properties are: stability, uniform structure, small thickness, high strength and stretching, porosity, small surface mass and water permeability. Various applications require a more or less marked particular structure and characteristic. The first part of the project will deal with various structures and properties of technical textiles based on nonwoven and knitted structures, in particular on geotextiles. Moreover, manufacturing technologies of technical textiles and knitted materials as well as their controlling parameters will be discussed. Conventional technologies such as: spinning, weaving, knitting and clothing technology will probably not withstand the competitiveness coming from Asia. Besides, relocation of the manufacture of man-made fibres into the Far East is taking place. It is undoubtedly the case that only those disposing of raw materials and enough knowledge to produce and sell high-quality products will have the chances of survival on the market. The investigation of possibilities of manufacturing from coarser sorts of wool which have similar fineness as domestic wool and the investigation of their possible use for products such as carpets and several articles of clothing will be within the scope of this project. The limit of fibre spinnability, typical stress-strain curves, yarn behavior in cyclic examinations of elongation properties, surface friction and yarn hairiness.
Project aims and objectives There are two dominant reasons why the field of technical textiles is dealt with in the project: .
Intensive development of technical textiles because of an increase in the application (technical textiles on average of approx. 5.5 per cent, nonwoven fabric approx. 10 per cent).
.
Resistance to the relocation of the manufacture of technical textiles to the Far East.
As a result the manufacture of technical textiles (quantity) in the developed countries accounts for more than 40 per cent of the total production of textiles. The state of the Croatian textile industry should be noted where the classic textile industry has almost vanished. Two companies Regeneracija and Cˇateks manufacture technical textiles which have their markets. Regeneracija and Cˇateks confirmed their participation in the project. Moreover, the use of domestic wool for various products of higher value is a chalange and obligation. However, today domestic wool is sold as a raw material, and in some regions it is not bought off which is an ecological problem. Based on the above mentioned facts, it is reasonable to deal with the subject matter of technical textiles and yarns and products, respectively, such as carpets within the scope of the project. In this way production of higher-quality products is promoted which is the purpose of this investigation. The aims of the investigation are as follows: Definition of the interdependence, process parameters and physical-mechanical as wel as other relevant properties. .
Technical nonwoven textile.
.
Technical textile based on nonwoven fabric coated with polyurethane (PUR).
.
Technical textile based on knitted fabric from PET and PA coated with polyurethane (PUR) and other technical textiles based on knitted fabric.
Definition of the interdependence of raw materials, process parameters and physicalmechanical yarn properties, primarily wool, and the behavior of carpets in dynamic investigations (new instrument required for the project). It will be attempted in case of interest of Regeneracija or other interested parties to investigate thermal properties of wool insulation materials.
Research deliverables (academic and industrial) Obtaining the new understandings of: .
Thermal and vapor resistance properties of the knitted and nonwoven fabrics.
.
Spun yarns, primarily coarser wool yarns: procedures of manufacturing and defined the controlling parameters of the processes.
.
Structure and properties of the coarser, industrially manufactured wool yarns should be emphasized that can be similar to the yarn spun from domestic wool as far as their properties.
.
Influence of fibres parameteres (finenness, length,. . .) on limit of spinnability and spun yarn properties.
.
Possibilities of manufacturing carpets from coarser wool fibres will be investigate as well as compressibility of carpets on the new instrument purchased from funds of the project.
.
Processes of manufacturing: technical nonwoven fabrics intended for use in civil engineering, technical nonwoven fabric coated with polyurethane (PUR), technical textiles based on knitted fabric coated with polyurethane (PUR), as well as their structures and properties.
Research register
77
IJCST 20,6
78
Publications 1. Salopek, Ivana; Skenderi, Zenun and Srdjak, Miroslav, “Stoffgriff – ein Aspekt des Tragekomforts von Strickware”, Melliand Textilberichte, Vol. 88, pp. 426-8, 2007. 2. Salopek, Ivana and Skenderi, Zenunm, “Thermophysiological comfort of knitted fabrics in moderate and hot environment”, Proceedings of the 3rd International Ergonomics Conference/Mijovic´, Budimir (ur.). Zagreb: Croatian Society of Ergonomics, 2007, pp. 287-93. 3. Potocˇic´ Matkovic´, Vesna Marija and Salopek, Ivana, Computer assisted study of knitted structures, Proceedings CE Computers in Education, Vol. IV, 2007, pp. 99-102. 4. Kopitar, Dragana and Skenderi, Zenun, “Prsteni i trkacˇi – glavni elementi prstenaste predilice”, Tekstil, Vol. 55, pp. 543-501, 2006.
Zagreb, Croatia Faculty of Textile Technology University of Zagreb, Prilaz baruna Filipovic´a 30, HR – 10 000 Zagreb, Croatia. Tel: +385 1 37 12 500; Fax: +385 1 37 12 595; E-mail:
[email protected] Principal investigator(s): Assoc. Prof. Zlatko Vrljicak, PhD Research staff: Kresimir Hajdarovic, PhD; Valent Strmecki, MSc; Tomislav Koren, MSc; Ivan Basnec, MSc
Design and manufacture of nets for the protection of fruit and vegetables against hail Other Partners: Academic
Industrial
None
Tvornica mreza i ambalaze, Biograd n/m, Croatia Project started: 1 January 2007 Project end date: 31 December 2009 Source of support: Ministry of Science, Education and Sport, Republic of Croatia Keywords: Fruit production, Nets, Hail, Viticulture, Plant material, Economic profit Approximately 7 per cent of Croatian fruit is present on the Croatian market. The production and sales of Croatian fruit can be quadrupled and sold at present prices on the Croatian market, but as first class fruit. Over the last ten years The Ministry of Finance received damage reports worth more than 100 million kuna which were caused by hail. Fruit, vegetables, plants, flowers, nursery-gardens, and animals, material resources: houses, agricultural machinery, automobiles and the like get damaged. Up to now rockets have been used to ensure hail protection of fruit and vegetables. On account of a rapid increase in the volume of air transport this technique is less used and is substituted by using protection nets. Several more developed and neighboring European countries have started using nets for the protection of fruit and vegetables against hail. Within the scope of this project systems of applying protection nets in European countries and their use in Croatia would be studied. The emphasis here would be on the safety of orchards, new plants or crops and how to pay compensation for damages by hail. Appropriate protection nets would be designed and manufactured for particular agricultural products and then installed on plantations. Net construction depends on the application
so that protection nets of various widths, shapes, colors and structures with special emphasis on the raw material for the production of nets and for shadowing the area to be covered. Across Croatia nets would be offered to the registered fruit growers for use. During the first year of the project a fruit grower would be offered 1000 m2 of nets without charge with the aim that he buys the same quantity (ratio 1:1) and that he should cover only one part of his plantation. By continuous monitoring orchards all changes under the nets would be analyzed and then compared to the results obtained outside the nets. Over the period of five years relevant conclusions about the use of nets for the protection of fruit and vegetables against hail can be made. It is to be emphasized that the nets, which protect agricultural products against hail, can protect against sun, birds, animals, etc. By adequate use of the above mentioned nets it is to expect that yield per hectar will be increased as well as fruit quality. Up to now our fruit growers have collected less than 50 per cent of first class fruit, but using nets it can be expected to increase this limit over 80 per cent. In this way, when Croatia enters into the European Union, we can sell our quality fruits on our market and compete on international markets.
Project aims and objectives The purpose of the investigation is to provide assistance to fruit growers to protect their orchards, both trees and fruits against hail. The aim of the project is to design and make nets that will be helpful and commercially acceptable in the protection of fruits and vegetables against hail. Manufactured nets would be offered to fruit growers for use. During the duration of the project several types of nets for the protection of fruit and vegetables would be developed. The application aim of these protection nets is primarily to protect trees and fruits. In this way fruit growers would increase the quality of collected fruit and yield quantity per unit area. In the long term, when Croatia enters into the European Union, our fruit growers will be able to sell their quality fruits on the Croatian market and compete on the international markets. If we do not do it, the European fruit producers will sell their fruits on the Croatian market without competition. The Croatian fruit growers will not be able to enter into this market because of minor fruit quality. This action will prevent the breakthrough of export fruit on the Croatian market.
Research deliverables (academic and industrial) The whole plan and protocol is mentioned in section 9.2. Therefore, only basic characteristics are mentioned here. After 1st year more than 20000 m2 of nets will be installed on different fruit plantations across Croatia. We would collect data on necessary properties of nets for the application in a particular sector or for a particular orchard. Based on the gathered data we would design and make new nets with different physical-mechanical properties. We would come to basic data on the influence of sun beams and weathering on the changes in net characteristics. We would publish approximately 5 papers. During previous three years we would keep records of all important influences on the fruit quantity and quality. We should come to data on the influence of hail, sun, wind, thrash, birds, animals and the like on the fruit quality and quantity. We would publish a newsletter about the application of protection nets in the cultivation of fruit and vegetables. Publications Not available.
Research register
79
IJCST 20,6
80
Zagreb, Croatia Faculty of Textile Technology University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 37 12 552; Fax: +385 1 37 12 599; E-mail:
[email protected] Principal investigator(s): Assist. Prof. Zˇeljko Sˇomoi, PhD Research staff: Assist. Prof. Ana Kunsˇtek, PhD; Slavica Bogovic´, MSc; Anica Hursa, MSc, Igor Petrunic´, MSc; Assist. Prof. Simona Jevsˇnik, PhD; Daniela Zavec-Pavlinic´, PhD
Computational modelling in engineering analysis of textiles and garments Other Partners: Academic
Industrial
Kamensko d.d., Zagreb University of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Clothing technology, Numerical methods, Optimisation, Reinforcements on clothing The intention of this project is to give a contribution to advanced application of the methods of engineering analysis in the field of textiles and garment. This goal will be achieved by introduction, adaptation, elaboration and application of up-todate computational methods in the analyses of problems relevant for the field of textile and clothing engineering. Considering the existing experience and an overview of questions and problems actual for the engineering science in the field, the research is to be undertaken in a number of areas, such as: optimal design of structural reinforcements in garment based on the finite element analysis; three parameter model of tensile nonlinearity of textiles; computational evaluation of post-buckling stable state in prediction and simulation of fabric drape; general numerical solution of thin plate bending with application to optimal grip geometry in automated work piece manipulation. Depending on the timing and realization of these researches, there is a possibility of opening further research areas from the field of computational modelling in mechanics of textiles and garment, including the spatial modelling and design of clothing items. The methods of research to be applied primarily consist of derivation and elaboration of numerical models suitable for application in the problems under consideration, and the development and application of computer programmes based on these models. At the same time, the plan is to acquire and apply some of the existing software applicable in the problems to be considered, as well as to prepare and conduct experimental verification of results obtained by computations.
Project aims and objectives Aim and scope of the proposed research is to give a contribution in the improvement of the level of engineering and technological know-how in the field of textile and garment. The research is expected to result in computer programmes or engineering data collected in tables, diagrams, etc. that will be useful for the problem solution in the area of expertise covered by the research. The knowledge and methods developed by the research will be on offer for the interested subjects, from Croatia or elsewhere, primarily from the branch of textiles and garment. It can also be expected that after some time the collected knowledge and methods will be included in the teaching process at the Faculty of Textile Technology, primarily as parts of the subjects at the doctoral or diploma levels.
Research deliverables (academic and industrial) The principal user of the results will be Faculty of Textile Technology, as the institution for production and transfer of knowledge in the field of textile and garment. Further users will be the firms, institutions and individuals from among the designers and manufacturers of textile and garment, who already have the co-operation with the faculty, or shall have that co-operation in the future. The results and findings of the research shall be offered to these subjects by means of the Centre for development and transfer of textile and clothing technology and fashion design, as a unit in the structure of the faculty. The specific applications are expected in the expert analyses related to engineering in preparation of production processes in which the problems from the field of research appear. Publications 1. Hursa, Anica; Rogale, Dubravko and Sˇomoi, Zˇeljko, “Application of numerical methods in the textile and clothing technology, Tekstil, Vol. 55 No. 12, pp. 613-23, 2006. 2. Akrap-Kotevski, Visˇnja and Kunsˇtek, Ana, “Writing ability after brain damage, Proceedings of 3rd International Ergonomics Conference, Ergonomics 2007, Mijovic´, Budimir (Ed.), Croatian Society of Ergonomics, Zagreb, 2007, 279-85. 3. Sˇomodi, Zˇeljko; Hursa, Anica; Rolich, Tomislav and Rogale, Dubravko, “Numerical analysis and optimisation of mechanical reinforcement on clothing”, Book of Proceedings of the 1st meeting of Croatian Society of Mechanics, Cˇanaija, Marko (Ed.), Rijeka, Croatian Society of Mechanics, 2007, pp. 173-8. 4. Sˇomoi, Zˇeljko; Hursa, Anica and Rogale, Dubravko, “A minimisation algorithm with application to optimal design of reinforcements in textiles and garments”, Internationl Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 159-66, 2007.
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 3712575; Fax: +385 1 3712599; E-mail:
[email protected] Principal investigator(s): Assoc. Prof. Stana Kovacˇevic´, PhD Research staff: Assist. Prof. Zˇeljko Penava, PhD; Josip Haina, MSc; Ivana Schwarz, BSc Assist. Prof. Andrea Pavetic´, Nikol Margetic´, B.Sc, Irena Sˇabaric´,
Research register
81
IJCST 20,6
B.Sc, Valent Strmecˇki, MSc; Dubravka Gordosˇ, MSc; Biserka Vuljanic´, MSc; Prof. Vladimir Oresˇkovic´, PhD, in retirement; Assoc. Prof. Krste Dimitrovski, PhD, Blago Brkic´, PhD in retirement, Diana Franulic´ Sˇaric´, MSc
Advanced technical woven fabrics and processes
82
Other Partners: Academic
Industrial
Varteks d.d., Varazˇdin; Cˇ ateks d.d., Cˇakovec; Kelteks d.o.o., Karlovac; Lola Ribar d.d., Karlovac; Uriho, Zagreb; Peng d.o.o., Zagreb Project started: 1 January 2007 Project end date: 31 December 2009 Source of support: Croatian Ministry of Science, Education and Sport, Republic of Croatia Keywords: Technical woven fabrics, Composites, Wool, Flax, Tapestry, Ethno heritage
University of Ljubljana, Slovenia
The subject of this project is advanced technical woven fabrics and processes. They are intended for the use in interior decoration, transportation, industrial and medical purposes, tapestry and the like. These fabrics contain raw materials in common, and domestic wool and linen yarn as well as glass and carbon yarn will be preferred, but other natural raw materials will be used too, yarns of chemical fibers from synthetic polymers and “smart” yarns. The aim of this research is to find the most optimal raw material and fabric construction and to make a commercially acceptable, qualitative, healthy, comfortable and smart technical fabric. Basic investigations will include: physical-mechanical, thermal, relaxation and elongation properties, dimensional stability, abrasion, effect of sun rays, inflammability, air permeability, water repellency, degradation and the investigation of these properties depending on fabric application. The scope of investigation will include technical fabrics intended for use in civil engineering, transportation and household (3D fabrics for composites, fabrics for seat covers, furnishing fabrics, etc.) on which high requirements are set, such as: safety, resistance, comfort and aesthetics. Technical fabrics for industrial purposes such as filter fabrics and fabrics for composites which are of great importance for better utilization and productivity, and still more important in terms of ecological protection of environment, will be investigated. Healthy fabrics in medical terms from natural raw, and generally fabrics with various properties and applications subjected to additional treatments according to health standards. Part of the project will be directed at the investigation and revival of Croatian eco and ethno heritage, including the manufacture of tapestry, blankets and mats interwoven with art and skill of weaving, using domestic raw materials. The aim of this research is that tapestry authenticity and originality of work of art represent a unique value. The significance of this project is to revive the processing of domestic wool and flax in parallel with the investigation of new constructions and forms of glass technical fabrics, and new materials processed by new technologies. Several technical fabrics replicated in this project will serve as an encouragement for processing domestic wool and flax in smaller batches in karts regions of Croatia.
Project aims and objectives .
To design and manufacture technical fabrics from natural and new materials, and with their combination, with new methods and manufacturing procedures.
.
To investigate qualitative properties of technical fabrics from natural textile raw materials and to promote the use of primarily domestic wool and linen.
.
Within the basic aim it is necessary to define a numerical model of the manufacturing process and development of new composites so that chemical and physical processes affecting the quality and applicability of composites are investigated and analyzed experimentally. Thereby the influence of particular processes (manufacturing processes, temperatures, outside influences) on the changes of composite properties as well as the interrelationships will be determined.
.
The knowledge gained will be used to define a computer model for the construction of new types of technical fabrics and accordingly new composite materials. To systematize utility and quality values of wool yarn from domestic sheep breeds and to investigate the possibility of the industrial process in blending with other raw materials. To optimize the production of applicable products of domestic raw materials and to promote the revival of the textile ethno heritage in Croatia. To investigate the most cost-effective process and the most optimal shares of individual raw materials for the target product with new properties. To include other raw materials with domestic materials with the aim of higherquality, more valuable and attractive products. To manufacture healthy, high-quality, “smart”, comfortable, unique and commercially acceptable textile materials by integrating “smart” materials and new materials dyed with natural dyes.
.
.
.
.
.
Research deliverables (academic and industrial) Manufacture of technical fabrics from different composites with improved properties such as: more healthy (natural fibers), more comfortable (better thermal properties), and more attractive (uneven surface structure made by weaving various densities, fineness, colors and raw materials with different weaving techniques and different finishing treatments). It may be expected that the results obtained will make a considerable contribution to further development of manufacturing technical fabrics and composites based on them, and that their application range will be expanded. Manufacture of medical fabrics based on natural raw materials and new composites which will be used as part of orthopedic and surgical aids that will dispose of improved properties such as: strength, thermal properties, liquid absorption, elasticity, stability and other properties. Manufacture of tapestries, blankets and mats with natural and new materials dyed with natural dyes whereby modern, high-quality and unique products can be produced. Moreover, by mastering these skills the possibility of a quality restoration of the textile ethno heritage will be provided.
Research register
83
IJCST 20,6
84
By applying the results obtained in various economic branches the development of new construction materials as a substitution for classic materials will be possible which will make a direct contribution environmental protection and entire ecological development of the country. By realizing such results the project will make a further contribution of the development of science in the field of developing new materials which according to long-term strategic directions of research in Croatia. They emphasize the need for the investigation of new materials, constructions and manufacturing processes. At the same time, the project is on the track of short-term strategic directions of scientific researches which direct investigations at natural, glass and organic-inorganic hybrids, making intelligent materials and polymer research. Publications 1. Kovacˇevic´, S. and Schwarz, I., “Hand weaving – tradition of the future”, 7th Annual Textile Conference by Autex: From Emerging Innovations to Global Business, 26-28 June 2007, Tampere, Finland. 2. Schwarz, I., Flincˇec Grgac, S., Kovacˇevic´, S., Katovic´, D. and Bischof Vukusˇic´, S., “The efect of drying methods on sized yarn characteristics”, The 18th International DAAAM Symposium – Intelligent Manufacturing & Automation: Focus on Creativity, Responsibility, and Ethics of Engineers, 24-27 October 2007, Zadar, Croatia, B. Katalinic (Ed.), Vienna: DAAAM International, 2007 (in press). 3. Ujevic´, D., Kovacˇevic´, S,; Schwarz, I. and Brlobasˇic´ Sˇajatovic´, B., “Novi visˇeslojni tekstilni plosˇni proizvodi.”, 6th International Scientific Conference on Production Engineering: Development and modernization of production, October 24-26 2007, I.Karabegovic, V. Dolecek, M. Jurkovic (Eds), RIM 2007 (in press). 4. Kovacˇevic´, S., Ujevic´, D., Schwarz, I., Brlobasˇic´ Sˇajatovic´, B. and Brnada, S., “Analysis of motor vehicle fabrics”, Fibres & Textiles in Eastern Europe, 2007 (in press)
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipivic´a 30, 10 000 Zagreb, Croatia. Tel: +385 1 48 77 357; Fax: +385 1 48 77 357; E-mail:
[email protected] Principal investigator(s): Assoc. Prof. Sandra Bischof Vukusˇic´, PhD Research staff: Prof. Drago Katovic´, PhD; Assoc. Prof. Tanja Pusˇic´, PhD; Prof. Emeritus Ivo Soljacˇic´, PhD; Sandra Flincˇec Grgac, BSc
Antimicrobial finishing of textiles Other Partners: Academic
Industrial
Jadran, Hoisery Factory d.d., www.jadranUniversity of Maribor, Zagreb Public carapa.hr, Pamucˇna Industrija Duga Resa, Health Institute, Croatian d.d. www.pamucna-industrija.com National Institute of Public Health Project started: 1 January 2005 Project end date: 1 January 2008 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Antimicrobial finishing, Antimicrobial agents, Testing methods for antibacterial activity, Testing methods for antifungal activity, Legislative regulations
Protection of the environment and human health is now perceived as more important then the profitability and eficiency of a business. Antimicrobial finishing is used on textile fabrics to control bacteria, fungi, mold, mildew and algae present in everyday life, to resolve problems of deterioration, staining, odors and health concerns they cause. It has been long ago recognized that microorganisms, particularly bacteria can negatively textile influence fabrics, especially cotton ones. Most textiles used as cloths, underwear and bedclothes in hospitals and hotels are conductive to cross infection or transmission of diseases caused by microorganisms. During the project time alternative agents for the purpose of antimicrobial treatment will be developed. Two parallel investigations will be performed. Croatian partners will be developing system with polycarboxylic acids. Slovenian partners will be working on grafting of cyclodextrine on textile substrates. Their bacteriostatic activity will be determined using standard methods: JIS L 1902: 2002 and AATCC 147, before and after the washing process. The antimicrobial activity will be tested againts gram positive (Staphylococcus aureus) and gram negative (Klebsiella pneumoniae) bacteria. Additional tests would include Escherichia coli, which might cause urinary infection that is of particular interest for underwear knitted fabrics. In order to prevent this negative microbial influence knitted material will be treated with several agents of different producers in order to choose the best one and recommend it to the industrial partners. For efficiency control of the treated products, strict conditions of the microbiological laboratory and qualified staff are necessary. For that reason cooperation with Zagreb Public Health Institute and Croatian National Institute of Public Health will be conducted within the project. Further experiments include development of alternative method for antimicrobial treatment, such as microwave treatment. Microwave effectiveness for the purpose of antimicrobial treatment will be determined.
Project aims and objectives Project aim: .
implementation of optimal antimicrobial treatments to Croatian and Slovenian market.
Project objectives: . antibacterial and antifungal efficiency control of textiles treated with several antimicrobial agents, present at the market. . development of novel antimicrobial systems based on polycarboxylic acids. .
investigation of microwave effectiveness for the purpose of antimicrobial functionalisation.
Research deliverables (academic and industrial) .
Report ordered from SME “Pamucˇna industrija Duga Resa, d.d.”: Antimicrobial hygiene finishes, Usage guidelines.
.
Report ordered from SME “Jadran Hoisery Factory d.d.”.: Antimicrobial functionalisation of socks.
Research register
85
IJCST 20,6
86
Publications 1. Bischof Vukusˇic´, S., Pusˇic´, T., Katovic´, D. and Soljacˇic´, I. (2005), “Antimicrobial finishing of Socks”, 36. Simpozij o novostih v tekstilstvu, Tekstilije za sˇport in prosti cˇas, Ljubljana, Slovenia, ISBN 961-604530-X, pp. 163-77. 2. Flincˇec Grgac, S., Bischof Vukusˇic´, S., Katovic´, D., Matica, B. and Dragosˇa, M. (2006a), “Antimicrobial functionalisation of knitted fabrics”, Rewiew 2006, Published papers of Zagreb Public Health Institute, ISBN 953-6998-30-0, p. 147. 3. Flincˇec Grgac, S., Bischof Vukusˇic´, S., Katovic´, D., Matica, B. and Dragosˇa, M. (2006b), “Antimicrobial functionalisation of knitted fabrics”, 3rd International Textile, Clothing & Design Conference, Dubrovnik, Croatia, ISBN 953-7105-12-1, pp. 270-5. 4. Bischof Vukusˇic´, S., Flincˇec Grgac, S. and Katovic´, D. (2006c), “Antibacterial & antifungal protection of military socks”, 3rd International Textile, Clothing & Design Conference, Dubrovnik, Croatia, ISBN 953-7105-12-1, pp. 241-6. 5. Bischof Vukusˇic´, S., Flinecˇ Grgac, S. and Katovic´, D. (2007), “Antimicrobial textile treatment and problems of testing methods”, Tekstil., Vol. 56, accepted for publication.
Zagreb, Croatia University of Zagreb, Faculty of Textile Technology, Prilaz baruna Filipovica 30, HR-10000 Zagreb, Croatia. Tel: +385 1 37 12 566; Fax: +385 1 37 12 599; E-mail:
[email protected] Principal investigator(s): Prof. Maja Andrassy, PhD Research staff: Prof. Zvonko Dragcevic, PhD; Assoc.Prof. Emira Pezelj, PhD; Prof. Dubravka Raffaelli, PhD; Assist. Prof. Edita Vujasinovic, PhD; Zvonko Orehovec, PhD; Vera Friscic, MSc; Ruzica Surina, BSc, Prof. Majda Sfiligoj Smole, PhD
High performance textile materials and added-value fibers Other Partners: Academic
Industrial
None University of Maribor Faculty of Mechanical Engineering, Maribor, Slovenia Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Ministry of Science, Education and Sport, Republic of Croatia Keywords: HP materials, Textile fibers, Textile design Contemporary global trends of development in the field of textile fibres and fibrous materials have led to their increased use in various fields of industry and technique. Increase in consupmtion of these types of materials has constantly been recorded and by the beginning of the 21st century technical fibres account for half of all the fibres manufactured. The requirements imposed on fibres and materials in particular fields of application and extraordinary high and specific. These requirements have been met
through fibre engineering, i.e. development of new generic types of fibres. It can be assumed that innovative manufacturing and finishing processes, applied to conventional fibres, will result in their added value, so that they can be used to design new fabrics of pre-determined end-use properties. Such improvements in fibre properties and their use in fabric manufacture of added market value and broader scope of application are completely in accordance with the intentions of the European technological platform for future textile and garment, but it can also strongly stimulate the development of Croatian textile industry and its comptetitiveness in the global market. This is supported by the fact that there are considerable research, industrial and raw-material potentials in Croatia, necessary to accomplish the goals. Although domestic production is mostly based on imported fibres, clearly defined modifications of fibre structure and propertties, even for domestic fibres, such as wool, flax, textile regenerates and fibres made from recycled PET, that have been used in Croatian textile industry insufficiently until now, could be used as a starting raw material for the manufacture of high-performance textiles. In this manner, domestic fibrous raw materials would cease being waste material and would become strategic Croatian raw material, as well as a basis of future rational management of natural resources and a step in approaching sustainable development trends, recommended by the European Union and United Nations. The investigations proposed aim at establishing the possibilities of modifying conventional fibres, as well as developing the methods and procedures of objective measurement and evaluation of unconventional textile materials, in accordance with specific rules and requirements for individual types of high-performance textiles, including composites reinforced with fibres of modified properties. The results obtained will offer the construction of high-performance materials based on conventional fibres of added value, as well as the design and optimisation in accordance with the properties of the fibres used and pre-determined high end-use properties.
Project aims and objectives The main purpose of the project proposed is to determine possible interventions and modifications of conventional fibres and textiles, so as to obtain added value and to broaden the scope of their application. This immrovement of fibres and their application in the manufacture of new, knowledge-based innovative textiles of added market value is in accordance with the short term (energy and materials) and long-term (nano-science, new materials, constructions and production processes) strategic trends of research in the Republic of Croatia, strategies of the European technological platform for the future of textiles and garment in the XXI century, as well as with the trends of rational management of raw material resources and the concept of sustainable development, as proposed by the EU and UN recommendations. Valuable results and new knowledge are expected, especially regarding ecologically friendly and economically feasible production of high-performance textiles through the usage of domestic raw materials in their manufacture. There is a broad diversity of fibres and constructions present in the area of non-conventional textile materials and structures, which makes objective characterisation of their quality a difficult task, we propose to develop new methods, procedures and equipment for testing, so as to enable higher degree of objectivity in quality evaluation. The investigations are planned to initiate the development and optimising of the manufacture of high-performance textiles, matched with their
Research register
87
IJCST 20,6
88
increased and more and more specified areas of application. The results expected to be obtained will enhance the scope of knowledge in the field of textile fibres, materials and textile testing, also valuable in the education of young researchers, knowledge transfer and preparation of future textile engineers for the European labour market.
Research deliverables (academic and industrial) The results of the investigation will be directly applicable in Croatian textile industry, since new solutions for designing high-performance materials, based on conventional fibres of added value, will be proposed as based on the results of the investigations proposed. As there are some processing capacities still in Croatia (Regeneracija, Kelteks, Vrbenka, Konoplja, LIO, Feniks), working with imported fibres, the system of objective description and evaluation of domestic fibrous raw materials (especially flax and wool), as well as some instructions regarding their environmentally acceptable manufacture, processing and modifications, with the aim of enhancing end-use properties, are expected to create adequate conditions for econbomically feasible manufacture of textiles. Higher content of domestic fibres and raw materials in manufacture of textile would be a sound basis for new development and growth of the Croatian textile industry and its competitiveness in the European and global markets, as well as for realising the principles of sustainable development and rational raw material resource management. Developments and innovations in testing metodology and objective evaluation of relenat properties of the modified fibres and new high-performance textiles will enhance objectivity of testing and evaluation of high-performance technical textiles in general, which is, not only in Croatia but globally as well, a problem with no acceptable solution on the horizon. Some testing methods are expected to be used in production monitoring and control, which could contribute to more reliable and stable manufacture and realising pre-planned levels of quality. The investigations proposed open the way to scientific and professional collaboration with other institutions and with the industry. Publications 1. Ruzˇica Sˇurina i Maja Somogyi, “Biodegradable polymers for biomedical purpose”,Tekstil. Vol. 55 No. 12, pp. 642-5, 2006. 2. Ruzˇica Sˇurina i Maja Andrassy, “Resistance of lignocellulosic fibers to microorganisms, hrvatski skup kemicˇara i kemijskih inzˇenjera, knjiga sazˇetaka, posvec´en Lavoslavu Ruzˇicˇki i Vladimiru Prelogu, hrvatskim nobelovcima u kemiji”, Zagreb, 26. veljacˇa – 01. ozˇujka 2007., 286. 3. Cindric´, Jasna, “Improvements properties of flax fibers, diploma work”, Zagreb, Tekstilno-tehnolosˇki fakultet, 25.04. 2007., 56 str. Voditelj: Andrassy, Maja 4. Klasic´, Sanja, “Usable properties of modified flax fabric, diploma work”, Zagreb, Tekstilno-tehnolosˇki fakultet, 25.04. 2007., 53 str. Voditelj: Andrassy, Maja 5. Sˇurina Ruzˇica i Andrassy Maja, “Quality of modified flax fibers”, The 18th International DAAAM Symposium, “Intelligent Manufacturing & Automation: Focus on Creativity, Responsibility and Ethics of Engineers”, 24-27 October 2007 (in press). 6. E. Vujasinovic, Z. Jankovic, Z. Dragcevic, I. Petrunic and D. Rogale, “Investigation of the strength of ultrasonically welded sails”, International Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, ISSN: 0955-6222, pp. 204-214, 2007. 7. E. Vujasinovic, Z. Dragcevic and Z. Bezic, “Descriptors for the objective evaluation of sailcloth weather resistance”, Proceedings of 7th Autex Conference 2007, Tampere, Finland, ISBN: 978-952-151794-5, 26-28 June 2007.
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipivic´a 30, HR-10 000 Zagreb, Croatia. Tel: +38514877351; Fax: +38514877357; E-mail:
[email protected] Principal investigator(s): Prof. emeritus, Ivo Soljacˇic´, PhD Research staff: Asoc. Prof. Tanja Pusˇic´, PhD; Prof. Ljerka Bokic´, PhD; Asst. Prof. Branka Vojnovic´, PhD; Iva Rezic´, PhD; Prof. Jelena Macan, PhD; Asoc. Prof. Barbara Simoncˇic´, PhD, Prof. Sonja Sˇostar-Turk, PhD., Asist.Prof. Sabina Fijan, PhD; Mila Nuber, MSc; Ivan Sˇimic´, MSc; Dinko Pezelj, PhD, Versˇec Josip, MSc
Ethics and ecology in textile finishing and care Other Partners: Academic
Industrial
Labud, d.d. Zagreb and Vodovod, Zagreb University of Maribor and University of Ljubljana, Slovenia Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Wellness finishing of textiles, Determination of harmful substances on textiles, Toxicological and alergenic properties, Environmental protection, Hygiene and effects of textile care, Textile material sample preparation Modern textile finishing processes have to fulfill high demands due to the expectations of new textile materials properties and their persistence during care. Especially interesting in this respect are the new production processes of socks which include implementation of microcapsules that can release active materials for skin moisturizing. Their primal role is prevention of dryness, dandruff and allergenic reactions of the skin. The most suitable analytical methods for determination of durability to washing, friction and sweat will be tested. Durability to washing of products with special properties will be tested with different amounts of anionic and cationic surfactants in liquid detergents. The mechanism of adsorption and desorption, their influence on primary effect of the treatment, and the influence of the pH value and the mechanical way of treatment will be tested. On the ground of the obtained results, analytical methods for determination of micro components in the macro components of textile materials should be proposed, without regards to the specifications of the materials or the method of the treatment. The testing will involve a review of the analytical method of each individual analytical procedure as well as its impact on the obtained information. The parameters of the analytical procedure will be worked out with the purpose of restoration of historical textile by destructive and non-destructive methods for the preservation of national heritage. European controlling methods of new materials have ethical demands involving the human population health which demands an environmental friendly process. For this purpose the processes of textile finishing and care will be optimized. The possibility of obtaining new preventive properties, which were not previously present on the textile
Research register
89
IJCST 20,6
90
material or improvement of present protection, will be tested. The impact of washing cycles with detergent and UV absorber on pastel colored textile materials made of cotton, polyester and their mixtures on UPF and the shade change will be investigated. The quality control of water and effluents will be based on the determination of micro quantities of potential allergens, heavy metals, pesticides, dyes, and surfactants. The traces of solvents will be controlled on the clothing material and in the air during the chemical cleaning and further treatment processing.
Project aims and objectives The main goal of this investigation is to stimulate ethic ecological demands on the production processes, care processes, and thereby on the utilization properties of the textile materials wherewith it would be possible to get the optimal properties of materials regarding their functional properties by avoiding all possible harmful allergenic and toxicological influences of textile materials to consumers. Elaboration of production and textile finishing processes for the optimal effects (wellness finishing, protection from unwanted changes of utilization properties in texcare, elaboration of pastel dyed textiles laundering in detergent with UV absorber, additional laundering quality – UV protection), formulation of new compositions for laundering for the purpose of avoiding secondary harmful effects in modern conditions with maximal saving of water and energy, more safe treatment with solvents during dry cleaning. By monitoring of harmful inorganic and organic substances that are present in micro quantities on the textile materials, textile accessories, textile wastewaters and finished textile products, new analytical methods would be determined. Sampling procedures, sampling preparation steps, selection of appropriate analytical method and the processing of the obtained result will be optimized. In this investigation the mathematical modes for guiding of analytical procedure will be applied, what is economically justified because the time spend for investigation is much shorter, and the consumption of chemical reagents, energy and emission of harmful substances to the environment reduced. Special contribution will be in development of analytical methods for determination of components present on the historical textile, for the purpose of avoiding the damaging of the textile material during restoration conservation treatments.
Research deliverables (academic and industrial) The project is scheduled over three years. Eco problems and human ecology, especially presence of heavy metal traces in textile processes and fibres will be investigated and some results will be published. Analytical methods for qualitative and quantitative determination will be developed. The influence of sweat on the heavy metal emission will be tested from colored textile materials. Possibility and durability of wellness finishing effects particularly on PA pantyhose’s as well as methods will be established. UPF and change in shade of white and pastel colored textiles made from cotton, PET, PA and their blend with cotton during laundering with addition fluorescent compounds in detergent will be researched, too. Hygienic laundering with chemothermic and chemical treatments in order to destroy micro-organisms in compliance with existing recommendations will be done. Potentially irritations of the skin caused by textiles, finishing agents and inadequate rinsing during laundering will be studied. Investigation of anionic, cationic and nonionic
surfactant adsorption and desorption influenced by different composition of textile fibres, pH and temperature will be performed. The adsorption and desorption will be studied in order to establish a correlation between zeta potential and swelling capacity of textile fibres. Publications 1. Rezic´, Iva and Steffan, Ilse, “ICP-OES determination of metals present in textile materials”, Microchemical Journal, Vol. 85 No. 1, pp. 46-51, scientific paper, 2007. 2. Fijan, Sabina; Pusˇic´, Tanja; Sˇostar-Turk, Sonja and Neral, Branko, “The influence of industrial laundering of hospital textiles on the properties of cotton fabrics”, Textile Research Journal, 2007 (in publishing). 3. Pusˇic´, Tanja; Jelicˇic´, Jasenka; Nuber, Mila and Soljacˇic´, Ivo, “Istrazˇivanje sredstava za kemijsko bijeljenje u pranju”, Tekstil., 2007 (in publishing). 4. Pusˇic´, Tanja and Soljacˇic´, Ivo, “Changes in shade of cotton fabrics during laundering with detergents containing fluorescent brightening agent and UV absorber”, AATCC Review, 2007 (in publishing). 5. Vojnovic´, Branka; Bokic´, Ljerka; Kozina, Maja and Kozina, Ana, “Optimization of analytical procedure for phosphate determination in detergent powders and in loundry wastewater”, Tekstil, 2007 (in publishing)
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10 000 Zagreb, Croatia. Tel: +385 1 4877 360; Fax: +385 1 4877 355; E-mail:
[email protected] Principal investigator(s): Prof. Ana Marija Grancaric´, PhD Research staff: Assoc. Prof. Tanja Pusˇic´, PhD; Assist. Prof. Zˇeljko Penava, PhD; Anita Tarbuk, M. Sc., Lea Markovic´, BSc., Assist. Prof. Jasenka Bisˇc´an, PhD; Sonja Besˇenski, M. Sc., Ivancˇica Kovacˇek, PhD, D. Med., Prof. Djamal Akbarov, PhD, Prof. Emil Chibowski, PhD, Prof. Rybicki Edward, PhD, Prof. Eckhard Schollmeyer, PhD, Prof. M.M.C.G. Warmoeskerken, PhD
Interface phenomena of active multifunctional textile materials Other Partners: Academic Croatian National Institute of Public Health, Zagreb; Tashkent Institute of Textile and Light Industry, Uzbekistan; Maria Curie-Skłodowska University, Lublin, Poland; Technical University of Lodz, Poland; Deutsches Textilforschungsinstitut Nord-West eV; Institut der Universitat Duisburg Essen; University of Twente, Netherlands
Industrial Pamucˇna industrija Duga Resa, Duga Resa
Research register
91
IJCST 20,6
92
Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Textile material, Interface phenomena, Surface modification and finishing, Multifunctionality The goal of the project is synergistic effects of some compounds on modified textile surfaces for achieving multifunctionality of textiles. Interface phenomena of textile surfaces with special accent on surface free energy, zeta potential, electroconductivity, adsorption and desorption of surfactants and other compounds usually used in textile finishing will give a great contribution to multifunctionality of textile. The mechanism of adsorption and desorption of surfactants and other finishing agents on modified textile surfaces is expected to be clarified in the present project. Different surface modifications, pretreatment and finishing of textile, especially cotton and polyester, will be performed according to European Technology Platform for the future of textile and clothing. For such purpose advance processes like mercerization, cationization, alkali, EDTA, other compounds and enzymes for surface hydrolysis of PET fabric, optical bleaching, implementation of nano antimicrobial active silver ions and mineral delivery mechanism, zeolite and others will be performed. Aminofunctional and other compounds will be added to azalides for the synergistic high antimicrobial effects. In cotton pretreatment enzymatic scouring will be applied using enzymes pectinase and the newest cutinase, for removal of pectins and bioplymers from cotton impurities with lipophylic character, instead of ecologically unfavorable alkali scouring. The goal of the project is synergistic effects of some compounds on modified textile surface. Interface phenomena of the new textile materials produced from electroconductive, low electro resistance fibers will be investigated for the purpose of static electricity and electromagnetic protection and for its implementation as sensors or other electronic devices in intelligent textiles. Traditional protection and aesthetic role of textile will be spread in active textile multifunctionality. Project will deal with elektrokinetic phenomena (zeta potential, isoelectric point, IEP, point of zero charge, PZC, surface electrical charge, surface free energy), hydrophility and hydrophobilicity, whiteness, fluorescence and phosphorescence, friction, fabric cover factor, elasticity, air and water vapor permeability of textile materials and their protection on UV radiation, microbes and fungi, coldness, heat and flame, static electricity and electromagnetic field.
Project aims and objectives Project will continue researching on assignments from previous project (0117012). Purpose of these investigations is based on lightening of interface phenomena on textile which effect directly to its adsorption and interaction intensity between textile fibers and chemical compounds. Almost all possibilities in modification during manufacturing high performance synthetic fibers are used, therefore nowadays attention and research is on textile surface modification. Procedures and compounds for that modification varies, as their effect varies, but the purpose and aim are directed to synergism of two or more components for accomplishing hydrophob or hydrophil textile, textile highly resistant to atmospheric condition, bacteria, microbe and fungi, UV radiation and
open flame. Furthermore, important aim of the project is cotton high level of purity by unconventional agents and material pretreatment procedures for mercerization and cationization. Pectinase in previous project investigation showed good elimination of pectine from primary cotton layer, but hydrophility was not so high like alkali scoured cotton. Chemical composition of cotton cuticula has lypophilic polymers, biopolyesters, which can be degraded by cutinase, new enzymes for degradation of waxes for better hydrophility. Cotton cationization during mercerization is the most important innovation of previous project and the patent for it was asked. Electronegative cotton surface charge, of which anionic substances adsorption depends, is lower after cationization in harsh mercerization conditions. The aim of this project is antibacterial, UV and flame protection by nanoparticle implementation (Ag) using mineral delivery compound (zeolite and others) as well. Electroconductive fibers implementation in yarns of textile materials should result in static electricity removal, and hopefully other effects. The aim of polyester surface modification, optical bleaching, other compounds treatments is well-known aesthetic, as well as high UV protection, high material elasticity as a result of changes in fiber microstructure. Interface phenomena research on wide range possible fabric knitted and woven construction will givethe solution of problems of fabric construction influence to high effect in this project.
Research deliverables (academic and industrial) Interface phenomena of textile materials surface in wet medium results in textile electric surface charge cognition and surface free energy as well on which adsorption depends. Important application of this project results is in ecological enzymatic scouring with pectinases.Enzymatic scouring with new enzymes, cutinase, will remove biopolyester cuticula and improve cotton hydrophility, and therefore replace harsh conventional alkali scouring entirely. Important application will have, patent requested cotton cationization during mercerization. By this pretreatment electropositive cotton is achieved, with great anion adsorption on its surface in all textile finishing processes. These anions enclose all low and high molecular compounds for textile finishing and all pricondensates. Implementation of nanoparticles (Ag and others) is predicted during mercerization and cationization processes, therefore it is important to emphasize rational component of these procedures which gives cotton multifunctionality in all textile usage. The next important application is antibacterial textile accomplished with azalide treatment especially in synergism with aminofunctional and other compounds and systems. It is well-known that fluorescence of optically bleached increases whitening of textiles. Optical brighteners and other compounds researching will be of great importance in UV protection with textile material. Heavy metals are toxic and their research is of great importance in human health protection. Furthermore, in nowadays growing demands on life safety from external influences especially UV radiation, research of differently structured textile material interface phenomena will find application in textile for summer clothing. Publications 1. Grancaric´, Anamarija; Pusˇic´, Tanja and Tarbuk, Anita, “Enzymatic scouring for better textile properties of knitted cotton fabrics, Biotechnology in Textile Processing, Guebitz, Georg; CavacoPaulo, Artur; Kozlowski, Rysard (Ed.). New York, NY, The Haworth Press, Inc., 2006. 2. Grancaric´, Ana Marija; Tarbuk, Anita; Dumitrescu, Iuliana and Bisˇc´an, Jasenka, “UV protection of pretreated cotton – influence of FWA’s fluorescence”, AATCC Review, Vol. 6, No. 4, pp. 2-6, 2006.
Research register
93
IJCST 20,6
94
3. Anita Tarbuk, Ana Marija Grancaric´ and Volker Ribitsch, “Electrokinetic phenomena of textile fibers”. Book of abstracts XX.Croatian Meeting of Chemists and Chemical Engineers 2007, p. 301. 4. Ana Marija Grancaric´, Lea Markovic´, Anita Tarbuk and Eckhard Schollmeyer, “Properties of multifunctional cotton in accordance with international standards”, Conferece of Textile days, Zagreb 2007. 5. Ana Marija Grancaric´, Anita Tarbuk and Ivancˇica Kovacˇek, “Micro and nanoparticles of zeolite for the protective textiles”, Book of Proceedings of 7th Annual AUTEX Conference, AUTEX 2007, p. 1123, Tampere, Finland. 6. Anita Tarbuk, Ana Marija Grancaric´ and Mirela Leskovac, “Surface free energy of pretreated and modified cotton woven fabric”, Book of Proceedings of 7th Annual AUTEX Conference, AUTEX 2007, p. 1104, Tampere, Finland.
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10 000 Zagreb, Croatia. Tel: +385 1 48 77 352; Fax: +385 1 48 77 352; E-mail:
[email protected] Principal investigator(s): Prof. Drago Katovic´, PhD Research staff: Asoc. Prof. Sandra Bischof Vukusˇic´, PhD; Prof. emeritus Ivo Soljacˇic´, PhD; Dubravka Dosˇen Sˇver, PhD., Sandra Flincˇec Grgac, BSc, Asoc. Prof. Radovan Despot, PhD; Asist. Prof. Jelena Trajkovic´, PhD; Asist. Prof. Branka Lozo, PhD; Luka Cˇavara, MSc; Bozˇo Tomic´, M.c., Prof. Charles Yang, PhD; Prof. Christian Schram, PhD
Alternative eco-friendly processing and methods of cellulose chemical modification Other Partners: Academic
Industrial
Cˇateks, d.d., www.cateks.hr Faculty of Forestry, Croatia; Faculty of Graphic Art, Croatia; University of Georgia, USA; University of Innsbruck, Austria Project started: 1 January 2007. Project end date: 31 December 2011. Source of support: Ministy of Science, Education and Sports, Republic of Croatia Keywords: Multifunctional eco-friendly textile finishing, Polycarboxylic acids, Protective functionalities, Chemical modification of cellulose, Microvawe treatment of cellulose materials One of the requests of European Union for higher competiteveness of european market is rebuilding and reconstruction of traditional industrial sectors, especialy textile and wood industry. According to the strategical goals of the Republic of Croatia the project
emphasizes the use of highly sofisticated production processes and treatments of cellulose materials, i.e. obtaining additional and improved characteristics of wooden and paper matherials which can be acchieved by using high-tech processes and by introduction of nano- micro- and bio-technologies. One of the alternative methods for replaciong the conventional reactants containing formaldehyde which were used in textile and wood treatments so far, would be the modification with eco-friendly agents such as polycarboxylic acids. Efficiency of these treatments will be determined quantitatively by ester crosslinking analytical methods or by means of isocratic HPLC and spectrophotometric FTIR method. Standard methods of textile, wood and paper material testing would be used for examining their protective performance and resistance to weathering conditions. Part of the proposed project will be development of optional multifunctional treatment that would provide better protection of cellulose materials against microorganisms, UV, electromagnetic rays, flame, oil or water. Therefore, a particular attention will be payed to development and application of the agents which will not only improve the characteristics of textile matherials but also give it permanent freshness and provide additional care and protection, i.e. medical characteristics. Optimisation of alternative processing and methods will provide ecologically and economicaly favorable characteristics of treated matherials. Further process optimisation in order to improve processing quality could be obtained with new alternative method using microwave energy. Improved characteristics obtained with this method in our previous research confirm its usability in textile finishing processes as well as in chemical modification of wood. Previous research in this field represent worlwide novelty which should be by all means continued.
Project aims and objectives The purpose and aim of the proposed project is to obtain highly valuable and multifunctional treated textile materials that will acquire analogous price on the demanding market. This is the basic condition for the survival of Croatian textile, wood and paper industry on EU market. In textile area experiments will be conducted to obtain multifunctional environmentally friendly textile material which will simultaneously offer dimensional stability, flame retardancy, crease and antimicrobial resistance and will have no effects on human health. Further goal is to obtain chemicaly modified wood that will have reduced shrinking and water absorption as well as to obtain flame retardancy on wood and paper products. One of the equally important goals is construction of a semi industrial microwave device for continuous planar treatment of cellulose materials. The results obtained would be presented in the world best known papers in the relevant field. The most important goal of the project is affirmation of Croatian science in Europe and rest of the World, by presenting the results in international papers so as on International Conferences. It is important to stress that established cooperation with EU and USA experts, so as with their scientific institutions will be continued and expanded. In this project, where will scientists from abroad have an active contribution with their work, further contribution to development of high quality products will be added. We certainly hope it will affect development of Croatian industry and economy.
Research register
95
IJCST 20,6
96
Research deliverables (academic and industrial) Publications 1. Katovic´, D., Bischof Vukusˇic´, S. and Flincˇec Grgac, S. (2007), “Crosslinking cotton with citric acid and organophosphorus agent for the purpose of flame retardant finishing”, 85th Textile Institute Conference, Colombo, Sri Lanka, pp. 820-4. 2. Bischof Vukusˇic´, S., Flincˇec Grgac, S. and Katovic´, D. (2007), “Catalyst influence in low formaldehyde flame retardant finishing system”, 7th AUTEX Conference, Tampere, pp. 60-1. 3. Flincˇec Grgac, S., Katovic´, D. and Bischof Vukusˇic´, S. (2007), Combination of organophosphorus Agent and Citric acid in Durable Press Finishing of Cellulose Fabrics, Croatian Society of Chemical Engineers, Zagreb, Croatia, p. 281. 4. Bischof Vukusˇic´, S., Flinecˇ Grgac, S. and Katovic´, D. (2007), “Antimicrobial textile treatment and problems of testing methods”, Tekstil, Vol. 56, accepted for publication.
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovica 30, HR-10000 Zagreb, Croatia. Tel: +385 1 37 12 557; Fax: +385 1 37 12 591; E-mail:
[email protected] Principal investigator(s): Prof. Budimir Mijovic, PhD Research staff: Prof. Miroslav Skoko PhD, in retirement; Prof. Dragutin Taborsak PhD, profesor emerituss; Prof. Salah-Eldien Omer, PhD; Prof. Jovan Vucinic, Ph. D., NenadMustapic, Mr Sc., Jasenka Pivac, Mr Sc., Zlatko Jurac, Mr Sc.
Ergonomic design of the worker-furniture-environment system Other Partners: Academic
Industrial
Faculty of Forestry, University of Zagreb Tvin, Virovitica, Croatia Project started: 1 January 2007 Project end date: 31 December 2009 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Sitting, Furniture, 3D model of workplace, Work environment, Virtual reality Sitting furniture should enable the worker to take an optimal bodily sitting posture, ensuring active and dynamic sitting. A long-lasting and non-ergonomic bodily posture in this position causes uncomfortable sitting. Defining optimal working postures and strains makes a contribution to the reduction of necessary energy and facilitates working and circulation functions. The total work space should be designed in compliance with all criteria of the working posture and technical requirements. It is necessary to know the worker well, his working capabilities, work place and work methods to ensure an optimal working environment. Furniture dimensions and
workplace, surrounding the furniture, regarding its optimal utilization, should be harmonized with the worker’s anthropometric sizes. Research methods are experimental, theoretical and numerical. Functional dependences of the workerfurniture-environment system will be investigated, based on ergonomic postulates in order to find optimal conditions between work humanization and productivity. Investigations are determined by measuring and recording typical working postures as well as conditions of excessive workload. Using digitally scanned 3D anthropometric characteristics of the human body, a digital 3D biomechanical model is obtained, taking account of the appropriate kinematic-dynamic motion rules and the construction of the inner skeleton. 3D program applications with advanced automated defined anthropometric and ergonomic features of biomechanical models and digital figures will be recorded. A 3D visualization of the workplace by using a computer-based 3D model of furniture and computer-based character animations of workers will be performed. By using computer 3D program solutions, the prototype is substituted by 3D models on which all necessary designs and changes in real time have been carried out interactively. Computer visualization will be used to perform a biomechanical analysis of movements based on the real correlation within the space of the interaction of workers and belonging working environment on the obtained 3D models of workers and workspace. It is necessary to analyze the workspace and time studies of motion accurately. The 3D virtual model will enable a detailed biomechanical analysis of motion, speed and acceleration and more designer’s solutions of furniture with biomechanical and ergonomic parameters. Detrimental impact of too a high noise on workers as well as efficient procedures of noise reduction will be investigated. Special attention will be focused on detrimental action of microclimatic conditions regarding technological requirements of the industry. The optimization of work energy during the performance of the work by the worker will be performed to lessen fatigue and to remove excessive workload and to reduce sick-leaves. The performance of these investigations will result in ergonomic technical-economic design of the interactive work-furnitureenvironment system which is of great importance for the development of the Republic of Croatia and elsewhere in the world.
Project aims and objectives The purpose of these investigations is to achieve the reduction of work energy when workers work in sitting position and to enable the reduction of fatigue and falling ill. When performing work, a better economic performance with a lower energy consumption and operator’s fatigue should be obtained. In testing the level of detrimental sound effects their efficacious reduction could be achieved. The action of noise and vibrations could be prevented by investigating efficient noise and vibration dampers, damping the transfer of vibrations on workers by using pads and personal protective equipment. The aim of these investigations is to optimize work movements in sitting position with less operator’s fatige and reduction of sick-leaves in the industry. Thereby, working conditions and environment as well as safety in technological processes are to be especially stressed. They should be used to improve economic characteristics substantially. Special purposes of the investigation are the optimization of working postures and the confirmation of the knowledge about defining new criteria for the right ergonomic design of working pieces of furniture.
Research register
97
IJCST 20,6
98
Research deliverables (academic and industrial) The application of investigations means work simplification, investigation and determination of manufacture time from the point of view of ergonomic starting points. When performing work, a better economic effect is to be achieved with a less energy consumption and operator’s fatigue. Based on the investigated causes of the increase of sound pressure level of machines, it is necessary to find out qualitative solutions to reduce the level into permitted and tolerable limits. To obtain satisfactory microclimatic conditions, work spaces, machinery and air-conditioning instruments with better economic and other characteristics will be defined, whereby current energy consumption will be reduced. Furniture dimensions are determined by computer based visualization of virtual 3D character in interaction with digitized and really designed furniture and environment. Virtual simulation offers the opportunity to create optimal constructive solutions of furniture which enables active and dynamic sitting. It is a relaxation bodily posture that is thereby obtained when sitting. Ergonomic, anthropometric, functional and technical requirements of workers are thereby satisfied. A 3D simulation model of comfortable and safe bodily posture in sitting position is created, together with all biomechanical characteristics, and interactive effects with existing and new materials used for making sitting furniture are attained. Thereby, design and construction of sitting furniture is supplemented, which, because of its designer and constructive solutions and design observance, reduces the possibilities of low-quality designs and problems of body illness and causing inability to work. Publications 1. A. Agic, V. Nikolic and B. Mijovic, “Foot anthropometry and morphology phenomena”, Collegium Antropologicum, Vol. 30 No. 4, pp. 815-21, 2006. 2. A. Agic and B. Mijovic, “Planar model of the deformation behaviour of electrospun fibrous nanocomposites”, Tekstil, Vol. 55, pp. 606-12, 2006. 3. A. Agic, B. Mijovic and T. Nikolic, “Blood flow multiscale phenomena”, Collegium Antropologicum, Vol. 31 No. 2, pp. 523-29, 2007. 4. U. Reischl, V. Nandikolla, C. Colby, B. Mijovic and H.C. Wei, “Ergonomics consequence of chinese footbinding: a case study”, Ergonomics 2007, Budimir Mijovic (Ed.) Croatian Society of Ergonomics, 2007, pp. 1-6. 5. D. Novak and B. Mijovic, “Applying cognitive rrgonomics to teaching math”, Ergonomics 2007, Budimir Mijovic (Ed.) Croatian Society of Ergonomics, pp. 15-20. 6. N. Mustapic and B. Mijovic, “Assessing the slip resistence of flooring”, Ergonomics 2007, Budimir Mijovic (Ed.) Croatian Society of Ergonomics, 2007, pp. 155-63
Zagreb, Croatia Faculty of Textile Technology University of Zagreb, Prilaz baruna Filipovic´a 30, HR-10 000 Zagreb, Croatia. Tel: +38513712500; Fax: +38513712599; E-mail:
[email protected] Principal investigator(s): Prof. Darko Ujevic´, PhD Research staff: Jadranka Akalovic´, BSc, Prof. Jadranka Bacˇic´; Prof. Zoltan Baracˇkai, PhD; Vinko Barisˇic´, BSc, Ing. Iva Berket; Bajro Bolic´, BSc, Blazˇenka
Brlobasˇic´ Sˇajatovic´, BSc, Ksenija Dolezˇal, BSc, Mirko Drenovac, PhD; Prof. Milan Galovic´, PhD; Marijan Hrastinski, BSc, Renata Hrzˇenjak, BSc, M.D. Natasˇa Kaleboti; Prof. Isak Karabegovic´, PhD; Ivan Klanac, BSc, M.D. Irena KosTopic´; Prof. Tonc´i Lazibat, PhD; Nikol Margetic´, BSc, Prof. Zlatka Mencl-Bajs; M.D. Zˇeljko Mimica; Prof. Gojko Nikolic´, PhD; Alem Orlic´, BSc, PhD.M.D. Vedrana Petrovecˇki; BSc, M.E. Zˇeljko Petrovic´; Prof. Dubravko Rogale, PhD; Prof. Andrea Russo; Igor Sutlovic´, PhD; Prof. Vlasta Szirovicza, PhD; Irena Sˇabaric´, BSc, MSc.M.D. Nadica Sˇkreb-Rakijasˇic´, Marija Sˇutina, BSc, Prof. Larry C. Wadsworth, PhD
Anthropometric measurements and adaptation of garment size system Other Partners: Academic
Industrial
None None Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Anthropometric measurements, Garment size system Systematic anthropometric surveys have been conducted since 1901 with the aim of developing and improving systems for clothing and footwear sizes. The measurement results show how a national population changes over a period of several decades in physical build and size due to a series of factors (food habits, sports development, genetic predispositions, population migrations, climatic conditions, etc.). Based on the results of anthropometric measurements in the Republic of Croatia (2004/05) on the sample of 30,866 test persons aged between 1 and 82 a statistical analysis of body measurements was performed, a database including 5 basic studies of sex and age as well as a new standard for clothing and footwear was built. These results enable a significant and stimulating continuation of scientific research and a comparison to other national standards and their contributions to the creation of systems for clothing and footwear sizes. Elements common for national standards of garment sizing by an exact approach will be investigated and analyzed, in particular because the presumptions of national systems and starting elements, respectively, are not universally founded like intersize intervals which differ in sizes since the conformity of individual starting places is missing. Data will be provided for a common base with methods of body measuring and size designation of clothes according to the recommendations of the Technical Committee TC133 within ISO and EN standards as well as the design and development of a sophisticated computer system (DOV-KO) for unifying all body measurements and basic garment construction based on one or all other sizes. Within the scope of this project and based on experience, a very important cycle of anthropometric measurements of the sporting population in football, water polo, rowing, basketball and handball will be performed. 4,000 test persons from
Research register
99
IJCST 20,6
100
Zagreb, Osijek, Rijeka, Split and Dubrovnik will be measured, whereby specific body differences and deformations of muscles caused by longstanding training will be analyzed. A comparative analysis of the representative sample of the anthropometric measurements of sportsmen and other population as well as the investigation of other trends of body measurements will be performed. This will enable an exceptional insight into the anthropometric dimensions which reflect body shape, proportionality, composition and elements of success in sports, respectively. Stadiometar or a new measuring instrument for continuous measuring body height, foot length and width will be designed too.
Project aims and objectives Problems of garment sizing and fit affect the market globally, and a consequence of bad predictions of the quantity of necessary stocks for manufacturers and dealers poses a risk of high costs. In the case of domestic manufacturers samples may be additionally divided. Particular solutions may be considered more efficiently by interpreting the data from the anthropometric database connecting 5 studies according to sex and age and the system of sizing. Therefore, using the results of the anthropometric measurements taken and the basic projection of the new standard for clothing and footwear, one of the directives of the project is to investigate other national standards of Europe and the world and to create size intervals and a new Croatian standard. The study of body differences and specific deformities of the body muscles during the longstanding practice of athletes such as leg circumference, chest circumference, torso, shoulder width, arm and leg length, body height, palm length is an additional aim of this project which will show the body elements affecting success level in sport. Besides a greater adaptation of clothing and footwear to the home market, it would be advisable to ensure the continuity of investigating the national size standard by creating a sustainable Croatian system of clothing and footwear sizes in conformity with anthropometric surveys that are conducted periodically and systematically in developed countries in which a change in the morphology of the human body occurred over the last decades. By way of proof, systematic anthropometric measurements and sizing in France showed a tendency of average height growth. In Great Britain in female population a growth of bust circumference was recognized, whereas in USA studies point to the tendency of an increase in obesity (besides an aesthetic also a health problem). The interest of the scientific and professional public, manufacturers, tradespeople and consumers in sizing will continue to grow, since a faster change in established body proportions may be expected thanks to changes in living and food habits of the population, an unavoidable mingling of ethnic groups, increase in the number of older consumers of clothing and footwear which will be doubled in the next two decades, etc. Thus, it is additionally stressed how much it essential at the moment to ensure a valid starting point or a Croatian standard to pursue next movements of body dimensions in order to avoid a discrepancy and imposition of the specificities of domicile consumers to home and foreign manufacturers.
Research deliverables (academic and industrial) Anthropometry is the study of the measurement of the human body, but Pheasant has expanded it as “applied anthropometry” including quantitative data of size, forms and other physical characteristics of people that can be used in garment design. Since the form of the human body changed through time, the problem of ageing proved to make a contribution to perceived changes in body shape and size more than any other individual factor, such as for example improved nutrition and prolonged life, in particular the knowledge that the number of older consumers will be doubled by 2003. Therefore, the systems of sizing shall be updated periodically to ensure a correct fit of ready-made clothing. On the other hand, the home industry of clothing, fashion wear and footwear disposes of modest and aged data based on the out-of-date anthropometric measurements from 1962. It was therefore necessary to conduct a new cycle of anthropometric measurements and to use the obtained results. World fashion industry shows a special interest in measuring anthropometric characteristics of the population so that it gathers such data permanently, motivated by the wish for designing articles of clothing for all population groups, including the persons with pronounced specificities (higher stature, higher body weight, etc.). The use of investigations will contribute to creating a new and modern Croatian standard for clothing and footwear harmonized with ISO and EN standards. Besides the clothing and footwear industry, pediatricians, specialists of occupational and sports medicine, experts in wood processing industry, automotive industry, in the army and police will benefit from the investigation results. Teachers and students in undergraduate and graduate studies as well as teachers and pupils at technical schools will benefit from the development results of the computer system based on the selection of garment sizes. By using the investigations of the sporting population, one can get an insight into tendencies of diversities of body measurements and changes in muscles as a result of longstanding practice. Various specialists of sports medicine, orthopedists, garment and footwear designers will benefit form the results of this investigation because based on previous experience it is evident that mass customization is necessary for athletes. Knowing dimensional characteristics, this method would be considerably promoted and improved. Publications 1. Ujevic´, D., Rogale, D., Hrastinski, M., Drenovac, M., Szirovicza, L., Lazibat, T., Bacˇic´, J., Prebeg, Zˇ., Mencl-Bajs, Z., Mujkic´, A., Sˇutina, M., Klanac, I., Brlobasˇic´ Sˇajatovic´, B., Dolezˇal, K. and Hrzˇenjak, R. (2006), “Normizacija, antropometrijski pregledi i Hrvatski antropometrijski sustav”, Tekstil, Vol. 55 No. 10, pp. 516-26. 2. Ujevic´, D., Firsˇt-Rogale, S., Nikolic´, G. and Rogale, D. (2006), “Pregled razvojnih dostignuc´a u tehnologiji sˇivanja – IMB 2006”, Tekstil, Vol. 55 No. 12, pp. 624-31. 3. Ujevic´, D., Dolezˇal, K. and Lesˇina, M. (2007), A”naliza antropometrijskih izmjera za obuc´arsku industriju”, Poslovna izvrsnost, Vol. 1 No. 1, pp. 171-183. 4. Ujevic´, D., Hrzˇenjak, R., Dolezˇal, K. and Brlobasˇic´ Sˇajatovic´, B. (2007), “Hrvatski antropometrijski sustav – jucˇer, danas, sutra”, HZN Glasilo, Vol. 3 No. 1, pp. 5-10. 5. Hrzˇenjak, R., Ujevic´, D., Dolezˇal, K. and Brlobasˇic´ Sˇajatovic´, B. (2007), “Investigation of anthropometric characteristics and body proportions in the Republic of Croatia”, Proceedings of 7th Annual Textile Conference by Autex, Tampere, Finland, 25-28 June, pp. 1191-8.
Research register
101
IJCST 20,6
6. Ujevic´, D., Brlobasˇic´ Sˇajatovic´, B., Dolezˇal, K., Hrzˇenjak, R., Mujkic´, A. (2007), “Rezultati prvog antropometrijskog mjerenja stanovnisˇtva Republike Hrvatske”, Drugi kongres hrvatskih znanstvenika iz domovine i inozemstva, Split, Croatia, 5-10 May. 7. Nikolic´, G., Ujevic´, D. (2007), “Protractor for measuring shoulder slope”, Patent. 8. Ujevic´, D. (2007), “One-arm and/or two-arm anthropometer”, Patent.
102
Zagreb, Croatia University of Zagreb Faculty of Textile Technology, Prilaz bazuna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 48 77 359; Fax: +385 1 48 77 355; E-mail:
[email protected] Principal investigator(s): Prof. Ðurdica Parac-Osterman, PhD Research staff: Martinia Ira Glogar, PhD; Assist.Prof. Darko Golob, PhD; Assoc. Prof. Marija Gorensˇek, PhD; Assoc. Prof. Darko Grundler, PhD; Assoc. Prof. Nina Knesˇaurek, PhD; Prof. Nina Rezˇek-Wilson; Assist.Prof.Tomislav Rolich, PhD; Ana Sutlovic´, MSc; Vedran urasˇevic´ BSc.
Colour and dyestuff in processes of ecologically acceptable sustainable development Other Partners: Academic
Industrial
Jadran Stockings Factory University of Ljubljana and Maribor, Slovenia Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Dyestuff selection, Nano-technology, Optimizing dyeing process, Purifying and decolouring wastewaters, Colour management, Fuzzy logic Scientific contribution to sustainable development relies on unlimited support of basic, developing and employable research. Therefore, selection of multi-functional dyes (UV protection, antibacterial protection, micro capsules of multi-functional performance), applying nano-technology in the dyeing processes with the aim of preventing water contamination, development of new methods as well as purifying dyed wastewaters contribute to sustainable development. Both input and output parameters of water will be controlled throughout the entire dyeing process: amount of residual dye in dye-bath using Lamber-Beer absorption model; X, Y, Z standard spectral characteristics of colour defined by specific absorption coefficient SAC and water quality defined by BOD5, COD, TOC, AOX, electrical conductivity and other defining values. System of control comprising advance models of control such as fuzzy logic (model based on rules) and model based on physical and chemical processes will be developed and applied. Capital area of research will involve models of dyeing processes, colour control and its correlation to dye as well as the interactive system of dye control.
Models should describe and predict kinetics, reactivity, affinity, exhaustion, fixation and interaction of solutions containing various dyestuffs. Prediction of output process result as well as definition of both physical and chemical parameters crucial for controlling the process will be conducted based on afore-mentioned models. These models encompass kinetic models (according to Nernst and Langmuir) modified for interactions between dyestuff on fibre and in the solution. Interdisciplinarity of dye within the system of sustainable development is based on spectral characteristics of colour as the fundament model dependent of the employment conditions. Instrumental measurement of colour is involved in all industrial production processes: textile technology, design, graphic industry and elsewhere which enables implementing control and colour harmonization. Application of evolutionary algorithms for modeling computer aided design of textiles based on principals of examinee’s subjective evaluation. Methods of descriptive statistics as well as methods of statistic reasoning will be applied within the frame of statistic analysis. Scientific affirmation of research results will be computer simulation as well as in vivo confirmation.
Project aims and objectives Contribution to sustainable development relies on unlimited support of basic, developing and employable research. Aim of the project is to contribute to humane ecology (regarding UV, antibacterial and other protective properties), through use of multifunctional dyes and selection of appropriate waste water discoloration and purifying methods, all in order of obtaining biological quality of water (free of toxic, aromatic components which may form in the process of dye degradation). Cognition of structure and use of thermo sensible nano sized dyes will enable their use on fibres for special purposes. Advance dyeing technologies, with the overview on pretreatment of textile substrates (enzymatic, plasmatic and other) in order of preserving environment and saving energy, will be applied. Base of the project is application of dyeing process control, including advance models of control; fuzzy logic and model based on physico-chemical processes. Colour used as constant value will be applied for formation of fuzzy logic model, used for complex colour designing, automatic dye selection, direct transfer of colour coordinates data into the dyeing recipe setup system, advance recipe correction, as well as control and colour matching. Project research will enable use of new instrumental methods and development of researcher’s creativity, while graduate students and potential PhD students will be given a chance to get acquainted with scientific methodology, development of experimental skills and writing scientific papers.
Research deliverables (academic and industrial) Influence of dye’s chemical constitution and mode of dye-fibre bond onto antibacterial (e.g. Staphiloccoci, Escherichia), UVA and UVB protection properties. Influence of additives (electrolytes and surfactants) on dyeing process and degree of water pollution. Control of, in dye-bath and wastewaters, present electrolytes – elaboration of mathematical model. Further results considering influence of dye onto protection properties are expected. Application of thermo sensible dyes on children clothing. Influence of textile substrate’s pre-treatment (enzymatic, plasmatic pre-treatment, etc.) on dyeing kinetics and
Research register
103
IJCST 20,6
104
energy saving. Results of wastewater purifying and discoloration methods, with the emphasis on salt removal using physico-chemical methods, nano filtration and reverse osmosis. Colour as constant value of monitoring process, dye properties and colour matching in design applying evolutionary algorithms. Application of nano size particles. Influence of surfactants onto reactivity, affinity, exhaustion and fixation degree of reactive dyes. Advantages and disadvantages of physico-chemical decolouring methods. Dye degradation products and their toxicity (considering aromatic components) in wastewaters. Selection of dyestuff and its interaction with in the dye-bath present additives. Mathematical model based on measured values will be elaborated, while control system including a model based on physico-chemical processes will be applied. Fuzzy logic model, based on colour as constant value within control system will be worked out. Application of capsulated dyestuff and nano particles of zink and silver for special use (medical textiles). Pre-treatment of hydrophobic, synthetic fibres in the aim of increasing hydrophility and applicability of, in water soluble, dyes. From the economical and ecological aspect, a more acceptable system of purifying and decolouring wastewaters is expected. Mathematical model based on the analyses of input and output measured values, considering coloured waters, will be elaborated, while a control system using advance models, such as fuzzy logic (model based on rules) and based on physicochemical processes model. Evolutionary algorithms for modelling computer design of fabrics based on principals of subjective asessment. Model must comply with standards, flexible, stabile, precise, and easily applicable. It includes complete process modelling: dyeing, colour control and its relation to dye. These models involve kinetics models (Nernst, Langmuir) modified for dye – dissolved dye. Applying CCM (computer colour matching) methods based on Kubelka-Munk theory, spectral characteristics and colour parameters according to CIEL*a*b* system, a model of fuzzy logic for complex design by colour, automatic dye selection, direct transfer of colour coordinates data into the dyeing recipe setup system, advance recipe correction, as well as control and colour matching, will be elaborated. Evolutionary algorithms for modelling computer design of fabrics based on principles of subjective asessment. Publications 1. Ðurdica Parac-Osterman, Ana Sutlovic´, Vedran Ðurasˇevic´ and Tjasa Griessler Bulc “Use of wetland for dye-house waste waters purifying purposes”, Asian Journal of Water, Environment and Pollution, Vol. 4 No. 1, pp. 101-6. 2. Ðurdica Parac-Osterman, Vedran Ðurasˇevic´, Ana Sutlovic´ “Comparison of some chemical and and physical-chemical waste water discoloring methods”, Chemistry in Industry, in publication. 3. Martinia Ira Glogar, Darko Grundler, Ðurdica Parac-Osterman, Tomislav Rolich “Fuzzy logic based approach to textile surface structure influence in colour matching”, AATCC Rewiev, in publication. 4. Vesna Tralic´-Kulenovic´, Livio Racane, Ana Sutlovic´, Vedran Ðurasˇevic´ “Dyeing properties of new benzothiazol disperse dyestuff”, Croatian Meeting of Chemists and Chemical Engineers, February 26 – March 1, 2007, Zagreb, Croatia. 5. Ðurdica Parac-Osterman, Nevenka Tkalec Makovec, Ana Sutlovic´, Ljerka Dugan “Staphylococcus aureus and escherichia coli behavior on undyed and dyed wool”, Croatian Meeting of Chemists and Chemical Engineers, February 26-March 1, 2007, Zagreb, Croatia. 6. Ðurdica Parac-Osterman, Ana Sutlovic´, Vedran Ðurasˇevic´ “Application of wetland system”, Textile Dyes Zagreb 2007, March 9th 2007, Zagreb, Croatia.
7. Ðurdica Parac-Osterman, Ana Sutlovic´, Martinia Ira Glogar, “Dyeing wool with natural dyes in light of the technological heritage”, 7th annual Textile Conference by Autex: “From Emerging Innovations to Global Business”, 26-28 June 2007, Tampere, Finland. 8. Ðurdica Parac-Osterman, Ana Sutlovic´, Vedran Ðurasˇevic´, “Application of wool, CA and PP fibers as filters in wetland pretreatment media formation”, University of Zagreb, Faculty of Textile technology Intrenational Conference on Multi Functions of Wetland Systems, 26-29 June, Legnaro (Padova), Italy.
Research register
105
Zagreb, Croatia University of Zagreb, Faculty of Textile Technology, Prilaz baruna Filipovic´a 30, HR-10000 Zagreb, Croatia. Tel: +385 1 37 12 521; Fax: +385 1 37 12 599; E-mail:
[email protected] Principal investigator(s): Assoc. Prof. Emira Pezelj, PhD Research staff: Prof. Ruzˇica Cˇunko, PhD; Prof. Maja Andrassy, PhD; Assist. Prof. Edita Vujasinovic, Prof. Vili Bukosˇek, PhD; Antoneta Tomljenovic´, Ph. D., Sanja Ercegovic´, M. Sc., Maja Somogyi, BSc; Dubravka Gordosˇ, M. Sc.
Multifunctional human protective textile materials Other Partners: Academic
Industrial
None None Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Croatian Ministry of Science, Education and Sport, Republic of Croatia Keywords: Protective textiles, Multifunctionality, Smart textiles, Ceramic coatings, Sol-gel process The investigations proposed have been motivated by the fact that people are more and more exposed to various influences from the environment, which can be harmful to their health. Such harmful influences are, for example, UV irradiation, electromagnetic smog, high temperature, fire, etc. Contemporary textile materials for personal protection are required to offer high efficiency, in most cases multifunctionality, as well as a necessary level of comfort. The fabrics used are high-performance ones and interdisciplinary approach is necessary in research dealing with their development and manufacture. The thesis we propose is that the application of contemporary research results in the field of materials can be used to offer a new contribution to the development of multifunctional protective textile materials. The accent will be given to a purposeful surface modification of fabrics, using environmentally friendly agents and processes, which is in accordance with contemporary European trends of research in the field of materials. Special attention will be paid to investigating modifications using the new sol/gel process, combined with preceding ultrasound, laser and plasma treatment of textile surfaces.
IJCST 20,6
106
New possibilities of manufacturing efficient protective layers will be investigated, using various inorganic substances, including functional layers of nano-dimension made of hybrid inorganic-organic polymers. The aim is to optimise modification parameters of achieving efficient protection from UV and EM irradiation, as well as to increase resistance to abrasion, cutting and heat in particular materials, establishing antimicrobial properties at the same time. Adequate testing procedures will be established to evaluate the newly created materials. New levels of knowledge is expected to be achieved regarding correlation of protective properties and textile fabric composition, as well as the development of practical processes of obtaining aimed fabric modifications and the development of the methods of new material evaluation. New knowledge will contribute to the quality of education in the field of textile materials. Transfer of knowledge into actual industrial production is also expected. The results will be presented on international conferences and will be published in relevant international publications. The obtained results to be obtained could be used to stimulate manufacture of new high-performance textile materials for special purposes in Croatia.
Project aims and objectives The purpose of the investigations is to obtain new knowledge in the field of material development, especially regarding the new composites with textiles as a basic component. The knowledge should be directly applicable in practice, and simultaneously used to improve the quality of education, of both students, young researchers and experts from the industry. The new knowledge is expected to further the development of the Department of textile materials, where the investigations are organised. Based on the knowledge of high-performance materials, that has resulted in the development of the composites, and the role of textile component in them, the possibility will be investigated of obtaining highperformance composites for protection, in which textiles are the basic component. These are new textile materials to be used as protection from harmful influences of the general and working environment in high-risk industrial processes and other activities where people are exposed to risks of mechanical, thermal or chemical injuries, of infection by micro-organisms and even fatal risks from the causes. This is why protective materials are expected to offer high efficiency under various conditions, while the best solutions are aimed at obtaining multi-functional protection by a single material. The purpose of the research is to investigate the solutions that could be applied in textile industry, which could stimulate the introduction of knowledge-based and new-technology-based production in the industry, through adapting the industry to manufacture high-performance composite materials for special purposes. The aim of the investigation is to determine the procedures of obtaining multi-functional textiles for personal protection, simple to manufacture and use. The protective properties will be obtained by modifying the surfaces of the fabrics of various constructions, with the aim to establish optimal modification procedures and processing parameters which could offer efficient protection from individual influences, or, otherwise, protection from more influences. The investigations are supposed to result in solutions for objective evaluation of the effect achieved and the durability of protection as well, but also in the evaluation of the adequacy of the materials for a particular purpose. Adequate testing methods
and procedures will be developed, appropriate indicators defined and the correlation of the modification parameters and properties achieved established.
Research register
Research deliverables (academic and industrial) The purpose of the investigations is to obtain new knowledge in the field of material development, especially regarding the new composites with textiles as a basic component. The knowledge should be directly applicable in practice, and simultaneously used to improve the quality of education, of both students, young researchers and experts from the industry. The new knowledge is expected to further the development of the Department of textile materials, where the investigations are organised. Based on the knowledge of high-performance materials, that has resulted in the development of the composites, and the role of textile component in them, the possibility will be investigated of obtaining high-performance composites for protection, in which textiles are the basic component. These are new textile materials to be used as protection from harmful influences of the general and working environment in high-risk industrial processes and other activities where people are exposed to risks of mechanical, thermal or chemical injuries, of infection by micro-organisms and even fatal risks from the causes. This is why protective materials are expected to offer high efficiency under various conditions, while the best solutions are aimed at obtaining multi-functional protection by a single material. The purpose of the research is to investigate the solutions that could be applied in textile industry, which could stimulate the introduction of knowledge-based and new-technology-based production in the industry, through adapting the industry to manufacture high-performance composite materials for special purposes. The aim of the investigation is to determine the procedures of obtaining multi-functional textiles for personal protection, simple to manufacture and use. The protective properties will be obtained by modifying the surfaces of the fabrics of various constructions, with the aim to establish optimal modification procedures and processing parameters which could offer efficient protection from individual influences, or, otherwise, protection from more influences. The investigations are supposed to result in solutions for objective evaluation of the effect achieved and the durability of protection as well, but also in the evaluation of the adequacy of the materials for a particular purpose. Adequate testing methods and procedures will be developed, appropriate indicators defined and the correlation of the modification parameters and properties achieved established. Publications 1. R. Cˇunko, S. Ercegovic´, D. Gordosˇ, E. Pezelj, “Influence of Ultrasound on Phisical Properties of Wool Fibres”, Tekstil, Vol. 55, pp. 1-9 (2006). 2. A. Tomljenovic´, E. Pezelj, F. Sluga, “Application of TiO2 nanoparticles for UV protective shade textile materials”, Proceedings of 38th symposium of textile novelity, Ljubljana, Slovenija, 21st June 2007. 3. E. Vujasinovic, Z. Jankovic, Z. Dragcevic, I. Petrunic, D. Rogale, “Investigation of the strength of ultrasonically welded sails”, International Journal of Clothing Science and Technology, Vol. 19 Nos 3/4, pp. 204-14, ISSN: 0955-6222 (2007). 4. E. Vujasinovic, Z. Dragcevic and Z. Bezic, “Descriptors for the objective evaluation of sailcloth weather resistance”, Proceedings of 7th Autex Conference 2007, Tampere 26-28 th June 2007, Finland, ISBN: 978-952-15-1794-5. 5. “Ruzˇica Sˇurina i Maja Somogyi: Biodegradable polymers for biomedical purpose”, Tekstil, Vol. 55, No. 12, pp. 642-45, (2006).
107
IJCST 20,6
6. Ruzˇica Sˇurina i Maja Andrassy: Resistance of lignocellulosic fibers to microorganisms, XX. hrvatski skup kemicˇara i kemijskih inzˇenjera, knjiga sazˇetaka, posvec´en Lavoslavu Ruzˇicˇki i Vladimiru Prelogu, hrvatskim nobelovcima u kemiji, Zagreb, 26. veljacˇa – 01. ozˇujka 2007., 286. 7. Sˇurina Ruzˇica i Andrassy Maja: Quality of Modified Flax Fibers, The 18th International DAAAM Symposium, “Intelligent Manufacturing & Automation: Focus on Creativity, Responsibility and Ethics of Engineers”, 24-27th October 2007
108
Zagreb, Croatia Faculty of Textile Technology, University of Zagreb, Prilaz baruna Filipovica, HR-10 000 Zagreb, Croatia. Tel: +385 1 3712 540; Fax: +385 1 3712 599; E-mail:
[email protected] Principal investigator(s): Prof. Dubravko Rogale, PhD Research staff: Prof. Zvonko Drgacˇevic´, PhD; Prof. Gojko Nikolic´, PhD; Prof. Maja Vinkovic´; Snjezˇana Firsˇt Rogale, PhD; Slavenka Petrak, PhD; Goran Cˇubric´, BSc; Bojan Mauser, BSc.
Intelligent garment and environment Other Partners: Academic
Industrial
None None Project started: 1 January 2007 Project end date: 31 December 2011 Source of support: Ministry of Science, Education and Sports, Republic of Croatia Keywords: Intelligent garment, Active thermal protection, Micropneumatic system, Intelligent sick bed, Adaptable ironing machine, Multiaxial testing of textiles and their joints Investigations, construction and development of intelligent article of clothing related to its direct environment by developing an adaptable bed, adaptable ironing machine and measuring instrument for multiaxial testing physical-mechanical properties of technical textile and joined parts. The purpose of the project is that a research team makes researches resulting in a construction and realization of the first intelligent garment whose basic function is active thermal protection. It contains a sensor system for monitoring the values of air temperature inside and outside of the garment, data bus for data transfer, microcomputer and micro controller, and execution devices for the automatic regulation of thermal protection value. Controlling conduction and convection of the heat of the human body regulates thermal protection in such a way that based on anthropometric measurements several types of various air thermo insulation elastic chambers are constructed which are integrated into the construction of the garment between the outer shell and lining. Thermoinsulation chambers consist of several segments and have a twofold function so that by inflating sealing properties are assumed, and the heat loss of the human body by convection can be regulated and the
thickness of the air chambers can be changed by program, whereby the heat loss of the human body by conduction can to be regulated. Micropneumatic elements and the chambers would be equipped with sensors of air pressure integrated into them, because depending on air pressure values in the chambers there will be defined chamber forms, their sealing properties and thickness on which thermal resistance depends. Investigations would prove that the integration and efficient joint operation of the integrated sensors, microcomputers with associated algorithms of intelligent behavior and actuators so that an independent action of the garment is realized with the aim of thermal protection whereby the garment would have the attribute of active, adaptable and intelligent behavior in variable temperature conditions. Communication possibilities of intelligent garment with the environment would be examined and an intelligent sick bed, adaptable ironing-machine and an instrument for testing load would be developed. They would practically use the same or very similar sensor, computer and micropneumatic actuator systems, connection techniques, constructions and design as well as intelligent garment.
Project aims and objectives The basic aim of the proposed research project is to investigate the possible construction and practical realization of an intelligent article of clothing with thermal protection, adaptable bed, ironing machine for the technological manufacturing process and necessary measuring instruments. The purpose of the investigation is to investigate characteristics of all elements of the system and behavior of the system as a whole and the communication between intelligent garment and environment. In addition to the basic aim of all investigations it is necessary to point out other aims too emerging as the result of the said investigation. Establishment of the leading European and world scientific role in investigations and development of intelligent garment. Writing 4 doctoral dissertations in the mentioned field (two dissertations in the mentioned field have been registered and approved by the Senate of the University of Zagreb. To prove that the Croatian clothing industry possesses a strong scientific research basis that guaranties it a technological excellence on demanding foreign markets.
Research deliverables (academic and industrial) After 1st Year: Selection and investigation of properties of installation materials, sensors, microcontroller systems and actuators as well as materials for making thermoinsulation chambers. Shapes and constructions of thermoinsulation chambers adjusted to dynamic anthropometric properties and necessary degree of freedom of movements, parameters of ultrasonic welding of chambers and influence of geometry, arrangement of chamber elements on thermoinsulation characteristics. Manufacture of microcontroller systems, sensor amplifiers, actuator drive and power supply system. Collection of technical literature and familiarizing with problems of beds for patients, ethimology and decubitus. Collection of technical documentation on the achievements of the technique of the intelligent ironing machine. Investigations of optimal ironing parameters: specific pressure, humidity, temperature, duration of humidification and drying.
Research register
109
IJCST 20,6
110
After 2nd Year: Computer digitalization of cutting parts of the garment, modeling, seam allowance and cutting pattern production with NC units. Installation of shell elements and thermoinsulation chambers together with integration of buses of the microcontroller system. Investigations and check of the characteristics of the operation of the sensor and actuator system and the operation of the power supply system. Actuation and investigation of the operation of the program languages and tools for the thermoinsulaiton chambers and 3D construction of all installation elements. Implementation of the microcontroller program. Manufacture of a functional prototype. Experimentation with various forms of the bladder for beds for patients on segments, various materials, procedure of bladder elongation without stretching, method of joining several segments. The process of the controlled generation of wet air with specific humidity and the accurate control of the intelligent ironing machine will be implemented. After 3rd Year: Design and examination of the complex program for measuring by means of integrated sensors, data interpretation and making decisions on the optimal combination of thermal protection, actuation of thermoinsulation chambers. Creation of an algorithm of intelligent behavior. The application of the PWM technology and communication protocol with display. Functional examination of the whole system of intelligent garment. Measurements and final check of the characteristics of the 3D construction, thermoinsulation chambers and all integrated components. Design of the sensor system for measuring the pressure in the bladders of the bed for patients and segments, respectively. Determination of the optimum air pressure. Devising a proportional pneumatic system that can independently and quickly change the pressure level in each segment and bladder. System design of measuring the ironing surface of the adaptable ironing machine by tactile sensors or camera system with computer program. After 4th Year: Investigations of autonomous operation and characteristics of an intelligent article of clothing in variable climatic conditions in the air conditioning equipment and in real atmospheric conditions. Improvement and conformation of the algorithm of intelligent behavior. Investigations and improvement of the communication channel between intelligent article of clothing and its wearer. Devising and experimentation with operation algorithms and control program of the system of the bed for patients, and the use of the touch screen monitors. Examination of the sensor system for accurate measuring fabric moisture, and a system for data acquisition about material type and its thickness by means of the integrated chip or bar code in the intelligent ironing machine. After 5th Year: Investigations of the communication channels between the intelligent garment and its environment by using suitable wireless communication. Unifying all systems of the bed for patients and designing a prototype with examination in real conditions. Unifying all partial systems into the operation system of the adaptable ironing machine that, depending on sensor data, makes own decisions on the most suitable operation parameters using fuzzy logic for making decisions and integrated operation algorithms. The system should be examined on prototype. Publications 1. Rogale, Dubravko; Firsˇt Rogale, Snjezˇana; Dragcˇevic´, Zvonko; Nikolic´, Gojko. “Intelligent article of clothing with an active thermal protection”, PK20030727 (patent).
2. Nikolic´, Gojko; Ujevic´, Darko. “Protractor for Measurement of Shoulder Slope (patent claim)”. 3. Dragcˇevic´, Zvonko; Rogale, Dubravko. “Pneumatski ulozˇak za sprecˇavanje deformacija perive obuc´e (patent claim)”. 4. Zˇeljko Sˇomodi, Anica Hursa, Dubravko Rogale. “A minimisation algorithm with application to optimal design of reinforcements in textiles and garments”, International Journal of Clothing Science and Technology, 2007, Vol. 19 Nos. 3/4, pp. 159-66 (original scientific paper). 5. Nikolic´, Gojko; Cˇubric´, Goran. “Investigating the positioning edge accuracy of sensors in textile and clothing manufacture”, International Journal of Clothing Science and Technology, 2007, 19, 3/4, 178185 (original scientific paper). 6. Edita Vujasinovic, Zeljka Jankovic, Zvonko Dragcevic, Igor Petrunic, Dubravko Rogale. “Investigation of the strength of ultrasonically welded sails”, International Journal of Clothing Science and Technology, 2007, 19, 3/4, 204-214 (original scientific paper). 7. Snjezˇana Firsˇt Rogale, Dubravko Rogale, Zvonko Dragcevic, Gojko Nikolic, Milivoj Bartosˇ. “Technical systems in intelligent clothing with active thermal protection”, International Journal of Clothing Science and Technology, 2007, 19, 3/4, 222-233 (original scientific paper). 8. Petrak, Slavenka. The method of 3D garment construction and cutting pattern transformation models, Doctoral thesis. Zagreb: Tekstilno-tehnolosˇki fakultet, 17.05. 2007., 220 str. Voditelj: Rogale, Dubravko. 9. Snjezˇana Firsˇt Rogale. Intelligent clothing with active thermal protection, Doctoral thesis. Zagreb: Tekstilno-tehnolosˇki fakultet, 19.03 2007., 224 str. Voditelj: Dragcˇevic´, Zvonko. 10. Hursa, Anica; Rogale, Dubravko; Sˇomoi, Zˇeljko. “Application of numerical methods in the textile and clothing technology.” Tekstil. 55 (2006), 12; 613-623 (review paper). 11. Ujevic´, Darko; Firsˇt-Rogale, Snjezˇana; Nikolic´, Gojko; Rogale, Dubravko. “Survey of development achievements in the sewing technology – IMB 2006.” Tekstil, cˇasopis za tekstilnu tehnologiju i konfekciju. 55 (2006.), 12; 624-631 (presentation on symposium). 12. Ujevic´, Darko; Rogale, Dubravko; Hrastinski, Marijan; Drenovac, Mirko; Szirovicza, Lajos; Lazibat, Tonc´i; Bacˇic´, Jadranka; Prebeg, Zˇivka; Mencl-Bajs, Zlatka; Mujkic´, Aida; Sˇutina, Marija; Klanac, Ivan; Brlobasˇic´ Sˇajatovic´, Blazˇenka; Dolezˇal, Ksenija; Hrzˇenjak, Renata. “Standardization, anthropometric surveys and croatian anthropometric system. Tekstil. 55 (2006), 10; 516-526 (review paper). 13. Sˇomoi, Zˇeljko; Hursa, Anica; Rolich, Tomislav; Rogale, Dubravko. “Numerical analysis and optimisation of mechanical reinforcements on clothes.” Zbornik radova Prvoga susreta Hrvatskog drusˇtva za mehaniku/Cˇanaija, Marko (ur.). Rijeka: Hrvatsko drusˇtvo za mehaniku, 2007. 173-178 (professional paper).
Research register
111
Research index by institution
IJCST 20,6
112
Institution
Page
Budapest University of Technology and Economics
19-25
Dokuz Eylul University
13-19, 42-57
Ege University
10-13
Heriot-Watt University
25-40
Hong Kong Polytechnic University
40-42
National Institute of Technology Jalandhar
57-59
National Technical University of Athens
7
Rapra Technology Ltd
73-75
Sci-Tex, Japan
65-66
Technical Univerity of Lodz
59-60
Technological Education Institute of Piraeus
6
University of Bolton
8-9
University of Manchester
60
University of Maribor
60-64
University of Nottingham
64-65
University of Pisa
67-73
University of Zagreb
75-111
Research index by country Country
Page
Croatia
75
England
8, 60, 73
Greece
6
Hong Kong
40
Hungary
19
India
57
Italy
67
Japan
65
Poland
59
Scotland
25
Slovenia
60
Turkey
10, 13, 42
Index by country
113
Research index by subject
IJCST 20,6
114
Anthropometric measurement garment sizing
98
Antihail nets
78
Antimicrobial finishing, textiles
24, 84
Barrier flexible structures; textiles, leather, paper
70
Bio-controlled delivery, nanoporous
36
Cellulose chemical modification
94
Clothing, Active thermal protection, 33, 38, 59, Anthropometric measurement, 60, 61, 80, Ergonometric, intelligent garment, 96, 98, 108 Ironing Sewing Work conditions Colour, Standard depth of textiles
26
Composites, Composite reinforcement, Strength properties
19, 73
Computer fluid dynamics
31
Durable press
53
Dyestuff selection, dyeing
102
Ecology, Ecofriendly processing renewable sources, Fibre spinning Finishing, Cellulosic enzyme, Ecology, Flame retardance, Plasma polymerisation Synthetic and natural fibres,
Friction, yarn
16
Health management, bio-sensing
33
Inkjet printing
27
Medical textiles Anti-microbial, Biopolymers, Compression therapy, Controlled delivery, Microencapsulation, antifungal agents
8, 24, 36, 39, 46
55
Modelling, 3D garment simulation, Composites, Garments, High performance
17, 19, 28, 60, 80
Nanotechnology, Electrospinning, Nano alignment Nano fibre production, Nanoporous
32, 34, 36, 102
Non-woven, knitted textiles, composites, yarns
75
Photovoltaic thin film
25
Plasma technology,
47
Processing and performance, composites
64
Protective clothing
60
Pulse-jet filtration
57
12, 94
Sewing, work conditions
59
48, 53, 52
Smart, Active thermal protection Bio-sensing Conductive fibres, Electro textiles,
15, 33, 38, 40, 47, 50, 52, 67, 105, 108
73, 89, 94, 102 14
Textile materials, fabric properties, 7, 61 structures
Emergency disaster wear, Fibre sensors, Micro-nano structures Radar-absorbing textiles Wireless clothing Sol-gel, Sol-gel processing, ceramic coating, Technical textiles, Composites Filtration High performance textiles, Multifunctional textiles, Protective textiles, Technical textiles,
50, 105
19, 57, 64, 65, 81, 86, 91, 105
Textile science, knowledge
66
Thermal, water vapour properties
75
Ultrasound, environment, finishing
10
University and industry interaction
6
Weaving, biopolymers, 3D weaving Wireless, Yarn hairiness, cellulose, Tencel, friction
39, 60 38 16, 56
Index by subject
115
Research index by principal investigator
IJCST 20,6
116
Principal investigator Aksit Aysun Cireli
Page 47, 48, 52 54
Okur Ayse
16
Osterman Urica Parac
102
Ozdemir Ismail
50
Andrassy Maja
86
Pezelj Emira
105
Borsa, J.
24
Pickering, S.J.
64
Bulgun Ender Yazgan
15
Provatidis, C.
7
Chen X.
60
Rajendran S.
Christie, R.M.
27
Rogale Dubravko
108
Clifford, M.
64
Rudd, C.
64
De Rossi, D.
67, 70, 72
Sariisik Merih
55
Duran Kerim
10
Scotchford C.A.
64
Erdem Nilufer
13
Skenderi Zenum
75
Fotheringham A.
39
Soljacic Ivo
89
Frydrych, I.
59
Somoi Zeljko
80
Gersˇak, J.
61
Stylios, G.K.
Grancaric Marija
91
28, 31, 32, 35, 36, 38
Hare, C.
73
Sular Vildan
56
Jones, I.A.
64
Tao Xiaoming
40
Kaplan Sibel
43
Ujevic Drago
98
Katovic Drago
94
Vas Laszlo M.
19
Korlu Aysegul
12
Vassiliadis, S.
6
Kovacevic Stana
81
Vrljicak Zlatko
78
Kurbak Arif
46
Vukusic Sandra Bischof
84
Long, A.
64
Walker, G.S.
64
Mather, R.R.
25
Wardman, R.H.
8
26, 27
65, 66
Warrior, N.A.
64
Mijovic Budimir
96
Wilson, J.I.B.
25
Mukhopadhyay, A.
57
Yesilpinar M Seril
17
Matsuo T.
