Springer Series in Synergetics
Editor: Hermann Haken
An ever increasing number ofscientific disciplines deal with complex systems. These are systems that are composed ofmany parts which interact with oneanother in a moreorless complicated manner. One of the most striking features of many such systems is their ability to spontaneously form spatial or temporal structures. A great variety 0 f these structures are found. in both the inanimate and the living world. In the inanimate world of physics and! chemistry, examples include the growth of crystals. coherent oscillations 0 flaser light. and the spiral structures formed in fluids and chemical reactions. In biology we encounter the growth ofplants and animals (morphogenesis) and the evolution ofspecies. In medicine we observe, for instance. the electromagnetic activity of the brain with its pronounced spatio-temporal structures. Psychology deals with characteristic features of human behavior ranging from simple pattern recognition tasks to complex patterns ofsocial behavior. Examples from sociology include the formation of public opinion and cooperation or competition between social groups. In recent decades, it has become increasingly evident that all these seemingly quite different kinds of structure formation have a numberofimportant features in common. The task ofstudying analogies as well as differences between structure formation in these different fields has proved to be an ambitious but highlyrewardingendeavor. The SpringerSeries in Synergetics provides a forum for interdisciplinary research and discussions on this fascinating newscientific challenge. It deals with both experimental and theoretical aspects. The scientific community and the interested layman are becoming ever more conscious 0 fconcepts such as self-organization. instabilities. deterministicchaos. nonlinearity, dynamical systems. stochastic processes. and complexity. All of these concepts are facets of a field that tackles complex systems. namely synergetics. Students, research workers. university teachers. and interested laymen can find the details and latest developments in the Springer Series in Synefgetics. which publishes textbooks. monographs and, occasionally, proceedings. As witnessed by the previously published volumes. this series has always been at the forefront of modern research in the above mentioned fields. It includes textbooks on all aspects ofthis rapidly growing field. books which provide a sound basis for the study ofcomplex systems. A selection of volumes in the Springer Series in Synergetics: Synergetics An Introduction 3rd Edition By H. Haken Handbook of Stochastic Methods for Physics. Chemistry, and the Natural Sciences 2nd Edition ByC. W. Gardiner Noise-Induced Transitions Theory and Applications in Physics. Chemistry. and Biology By W. Horsthemke. R. Lefever The Fokker-Planck Equation 2nd Edition By H. Risken Nonequilibrium Phase Transitions in Semiconductors Self-Organization Induced by Generation and Recombination Processes By E. Scholl Synerge,tics of Measurement. Prediction and Control By I. Grabec. W. Sachse Predictability of Complex Dynamical Systems By Yu. A. Kravtsov. J. B. Kadtke Interfacial Wave Theory ofPattem Formation Selection of Dentritic Growth and Viscous Fingerings in Hele-Shaw Flow By Jian-Jun Xu Cooperative Dynamics in Complex Physical Systems Editor: H. Takayama
lnforma tion and Self-Organization A Macroscopic Approach to Complex Systems By H. Haken Foundations of Synergetics I Distributed Active Systems 2nd Edition By A. S. Mikhailov Foundations ofSynergetics II Complex Patterns 2nd Edition By A. S. Mikhailov, A. Yo. Loskutov Synergetic Economics By W.-B. Zhang Quantum Signatures of Chaos By F. Haake Nonlinear Nonequilibrium Thermodynamics I Linear and Nonlinear Fluctuation-Dissipation Theorems By R. Stratonovich Nonlinear Nonequilibrium Thermodynamics II Advanced Theory By R. Stratonovich Modelling the Dynamics of Biological Systems Editors: E. Mosekilde. O. G. Mouritsen Self-Organization in Optical Systems and Applications in Information Technology 2nd Edition Editors: M. A. Vorontsov. W. B. Miller
Jan Awrejcewicz Igor V. Andrianov Leonid I. Manevitch
Asymptotic Approaches in Nonlinear Dynamics New Trends and Applications
With 58 Figures
Professor Jan Awrejcewicz Division of Control and Biomechanics (1-10), Technical University ofL6dz 1/15 Stefanowskiego St., PL-90-924 L6dz, Poland
Professor Igor V. Andrianov Pridneprovye State Academy ofCivil Engineering and Architecture 243 Chernyshevskogo St., Dnepropetrovsk 320005. Ukraine
Professor Leonid I. Manevitch Institute ofChemical Physics, Russian Academy of Sciences 4 Kosygin St., 117977 Moscow, Russia
Series Editor: Professor Dr. Dr. h.c.mult. Hermann Haken Institut fur Theoretische Physik und Synergetik der Universitiit Stuttgart 0-70550 Stuttgart, Germany and Center for Complex Systems, Florida Atlantic University Boca Raton, FL 33431, USA
ISSN 0172-7389 ISBN 3-540-63894-6 Springer-Verlag Berlin Heidelberg New York Library of Congress Cataloging-in-Publication Data Awrejcewicz, J. (Jan), 1938Asymptotic approaches in nonlinear dynamics: new trends and applications 1 Jan Awrejcewicz, Igor V. Andrianov, Leoni
©
The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typesetting; Data conversion by K. Mattes, Heidelberg Cover design: design & production GmbH, Heidelberg SPIN 10652100 55/3144 - 5 4 3 :2 I 0 - Printed on acid-free paper
Preface
How well is Nature simulated by the varied asymptotic models that imaginative scientists have invented? B. Birkhoff [52] This book deals with asymptotic methods in nonlinear dynamics. For the first time a detailed and systematic treatment of new asymptotic methods in combination with the Pade approximant method is presented. Most of the basic results included in this manuscript have not been treated but just mentioned in the literature. Providing a state-of-the-art review of asymptotic applications, this book will prove useful as an introduction to the field for novices as well a reference for specialists. Asymptotic methods of solving mechanical and physical problems have been developed by many authors. For example, we can refer to the excellent courses by A. Nayfeh [119--122]' M. Van Dyke [154]' E.J. Hinch [94] and many others [59,66,95, 109, 126, 155, 163, 50d, 59d]. The main features of the monograph presented are: 1) it is devoted to the basic principles of asymptotics and its applications, and 2) it deals with both traditional approaches (such as regular and singular perturbations, averaging and homogenization, perturbations of the domain and boundary shape) and less widely used, new approaches such as one- and two-point Pade approximants, the distributional approach, and the method of boundary perturbations. Many results are reported in English for the first time. The choice of topics reflects the authors' research experience and involvement in industrial applications. The authors hope that this book will introduce the reader to the field of asymptotic simplification of the problems of the theory of oscillations, and will be useful as a handbook of methods of asymptotic integration as well. The narration is commonly based on examples given by applied mechanics of structures (primarily, plates and shells) and fluid mechanics, but scarcely of quantum physics. Obviously, the methods in question are really versatile in application, covering applied mathematics, physics, mechanics and other basic sciences. The authors have paid special attention to examples and discussion of results rather than to burying the ideas in formalism, notation, and technical details. The aim is to introduce mathematicians - as well as
VI
Preface
physicists, engineers, and other consumers of asymptotic methods - to the world of ideas and methods in this burgeoning area. The effect of asymptotic methods (AM) on the theory of oscill~tions increases multifold. The vitality and prospect of AM becomes ObVIOUS from the fact that active interaction between numerical and analytical methods is accomplished via asymptotics. It is a pity that asymptotic mathematics does not occupy the decent place in education programmes of high schools. Certain tutorial aspects, useful for training mechanics, physicists, applied mathematicians and engineers, are presented here. Let us scan in detail the contents of the chapters. An introduction the depicting the principal ideas of asymptotic approaches through simple, ''transparent'' examples is given. The first part is devoted to discrete systems. First, an introduction to classical perturbation techniques are presented. The KBM methods and the equivalent linearization are described in some detail. Nonconservative nonautonomous systems are considered and nonresonance oscillations as well as oscillations in the neighbourhood of resonance are analysed. A general approach to the analysis of unstationary nonlinear systems is given. Particular attention is paid to consideration of. combined parametric and self-excited oscillations in a three-degree-of-freedom mechanical system. This example includes a derivation of the equation of motion and a determination of instability zones. The so-called modified Poincare approach is presented and illustrated on the basis of a one-degree-of-freedom system and then this approach ~ extended to the analysis of general nonlinear systems. Then, the Hopf bifurcatio~ is discussed from the viewpoint of the asymptotic approach. Finally, a method of controlling and improving the stability of periodic orbits of vibro-impact systems is proposed. This method is based on the feedback loop control with a time delay. This subchapter includes two parts of our investigation. In what follows a perturbation technique is applied to estimate delay loop coefficients for the improvement of stability of the vibro-impac;t motion for one-degree-of-fredom systems. Then, control of the periodic motion of the one-degree-of-freedom vibro-impact oscillator is analysed numerically, showing good agreement with the analytical prediction. Nonlinear normal vibrations are a generalization of normal (principal) vibrations of linear systems. In the normal mode all position coordinates can be defined from anyone of them. Using normal modes of nonlinear systems gives very interesting results, and in Sect. 2.10 we write about some aspects of the asymptotic construction of an object. Progress in the applications of AM in the theory of oscillations as well as in applied mathematics on the whole is closely linked with the introduction of new small parameters and, respectively, new asymptotic procedures. This is the field of Sect. 2.11. In Sect. 2.12 we deal with one- and two-point Pade approximants (PA). Usually PAs are used for the extension of the area of applicability of pertur-
Preface
VII
bation series. We propose to use PAs in connection with AM in many new cases, in particular: 1. Estimation of the convergence domain for perturbation series. In particular, such estimation may be obtained on the basis of the comparison of the perturbation series and PA. This result is justified by many interesting examples from nonlinear mechanics. 2. Elimination of nonuniformities of asymptotic expansions. The PA eliminates nonuniformities of asymptotic expansions in more important mechanical problems in a simpler way than, for instance, Lighthill's method. Up till now two-point PAs (TPPAs) have not been so widely used in mechanics. We represent new applications of the TPPAs for matching local expansions in nonlinear dynamics. The second part specifies the most important and useful forms and techniques of asymptotic thinking for the theory of oscillations for continual systems. Relations between the dynamics of discrete and continual systems are based on the procedures of discretization and continualization. The procedure of discretization is described well in many books, so we have paid some attention only to the continualization (the passage from discrete to continuous systems) in Sect. 3.1. Section 3.2 is devoted to the homogenization approach. The main problem in this field is in the solution of the so-called cell (or local) problem. This problem has usually been treated by a numerical method. We have used an asymptotic method for solving the cell problem and have constructed an approach in this book. The approach presented fills the substantial gap between numerical methods of the thin shell theory, which lack generality and the possibility of grasping the common features of the behaviour of the structures concerned, and approximate schemes, based on heuristic hypotheses. The methods proposed are wide-ranging in applications and lead to simple and clear design formulae, useful for practical analyses. The averaging approach is one of the most useful tools in the theory of nonlinear oscillations. Usually it is used with respect to time variables, but in Sect. 3.3 we show new perspectives for averaging with respect to spatial variables. V.V. Bolotin proposed an effective asymptotic method for the investigation of linear continuous elastic system oscillations with complicated boundary conditions. The main idea of this approach is in the separation of the continuous elastic system into two parts. Then the matching procedure permits us to obtain a complete solution of the dynamics problem in a relatively simple form. The idea of Bolotin's asymptotic method was generelized for the nonlinear case in Sect. 3.4. Regular and singular asymptotics in a wide range of forms are the old, but formidable weapons in the armoury of an asymptotic mathematician. In Sect. 3.5 a lot of interesting problems are solved on this basis, and some
VIII
Preface
interesting aspects of the application of these traditional approaches are notified. A new AM for solving mixed boundary value problems is considered in Sect. 3.6. The parameter e is introduced into the boundary conditions in such a way that the e = 0 case corresponds to the simple boundary problems and the case e = 1 corresponds to the general problem under consideration. Then, the e-expansion of the solution is obtained. As a rule, the expansion of the solution is divergent just at the point e = 1. The PAs are used to remove this divergence. The TPPA in application to nonlinear dynamic problems for a continuous system - a plate on a nonlinear foundation - is displayed in Sect. 3.7. The discovery of the soliton in 1965 by Kruscal and Zabusky has brought revolutionary changes in nonlinear science, and we describe some USes of the soliton technique in Sect. 3.8. In Sect. 3.9 a nonlinear analysis of spatial structures is described on the basis of the so-called modified envelope equation. The third part of the book includes an investigation of discrete-continuous systems. In Sect. 4.1, periodic oscillations of discrete-continuous systems with a time delay are analysed. In Sect. 4.2 a simple perturbation technique is described as it is used in the analysis of discrete-continuous systems with a time delay and with homogeneous boundary conditions. In Sect. 4.3 the nonlinear behaviour of an electromechanical system is investigated on the basis of an averaging technique supported by symbolic computation using the Mathematica package. Then the obtained averaging amplitude differential :. equations are analysed numerically. The book is mainly based on the authors' papers [6-24, 28-39, 115, 125, 156, 2d-27d, 55d, 56d, 62dJ. Finally, the first author (J.A.) wishes to acknowledge the financial support by the Polish National Scientific Research Committe Grants No 7T07AOl710 and No 7T07A00210. Mr K. Tomczak and 11r G. Wasilewski are thanked for their time and consideration paid to the preparation of this book. L6di Dnepropetrovsk Moscow April 1998
J. A wrejcewicz 1. V. A ndrianov £.1. Manevitch
Contents
1.
Introduction: Some General Principles of Asymptotology . 1 1.1 An Illustrative Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2 Reducing the Dimensionality of a System. . . . . . . . . . . . . . . . . . 4 1.3 Continualization........................................ 5 1.4 Averaging.,........................................... 6 1.5 Renormalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.6 Localization........................................... 8 8 1.7 Linearization........................................... 1.8 Pade Approximants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.9 Modern Computers and Asymptotic Methods. . . . . . .. . . . . .. 10 1.10 Asymptotic Methods and Teaching Physics , 11 1.11 Problems and Perspectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 11
2.
Discrete Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2.1 The Classical Perturbation Technique: an Introduction. . . . .. 2.2 Krylov-Bogolubov-Mitropolskij Method. . . . . . . . . . . . . . . . . .. 2.3 Equivalent Linearization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2.4 Analysis of Nonconservative Nonautonomous Systems. . . . . .. 2.4.1 Introduction..................................... 2.4.2 NonresonanCe Oscillations . . . . . . . . . . . . . . . . . . . . . . . .. 2.4.3 Oscillations in the Neighbourhood of Resonance. . . . .. 2.5 Nonstationary Nonlinear Systems. . . . . . . . . . . . . . . . . . . . . . . .. 2.6 Parametric and Self-Excited Oscillation in a Three-Degree-of- Freedom Mechanical System 2.6.1 Analysed System and Equation of Motion. . . . . . . . . .. 2.6.2 Transformation of the Equations of Motion to the Main Coordinates , 2.6.3 Zones of Instability of the First Order. . . . . . . . . . . . . .. 2.6.4 Calculation Examples. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2.7 Modified Poincare Method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2.7.1 One-Degree-of-Freedom System. . . . . . . . . . . . . . . . . . .. 2.7.2 General Nonlinear Systems. . . . . . . . . . . . . . . . . . . . . . .. 2.8 Hopf Bifurcation "
13 13 19 24 26 26 27 31 42 55 55 59 62 74 81 81 86 93
X
Contents Stability Control of Vibro-Impact Periodic Orbit _ 2.9.1 Introduction 2.9.2 Control of Vibro-Impa.ct Periodic Orbits 2.9.3 Stability Control 2.9.4 Simulation Results 2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom , 2.10.1 Definition 2.10.2 Free Oscillations and Close Natural Frequencies 2.11 Nontraditional Asymptotic Approaches , 2.11.1 Choice of Asymptotic Expansion Parameters , 2.11.2 b-Expansions in Nonlinear Mechanics 2.11.3 Asymptotic Solutions for Nonlinear Systems with High Degrees of Nonlinearity , 2.11.4 Square-Well Problem of Quantum Theory 2.12 Pade Approximants 2.12.1 One-Point Pade Approximants: General Definitions and Properties 2.12.2 Using One-Point Parle Approximants in Dynamics 2.12.3 Matching Limit Expansions 2.12.4 Matching Local Expansions in Nonlinear Dynamics .. , 2.12.5 Generalizations and Problems
2.9
3.
100 100 101 103 105 106 106 108 118 118 121 128 130 132 132 134 139 143 148
Continuous Systems , 151 3.1 Continuous Approximation for:a Nonlinear Chain 151 3.2 Homogenization Procedure in the Nonlinear Dynamics of Thin-Walled Structures , 155 3.2.1 Nonhomogeneous Rod 155 3.2.2 Stringer Plate 158 3.2.3 Perforated Membrane 162 166 3.2.4 Perforated Plate 3.3 Averaging Proced'ure in the Nonlinear Dynamics of Thin-Walled Structures 171 3.3.1 Berger and Berger-Like Equations for Plates and Shells 171 3.3.2 "Method of Freezing" in the Nonlinear Theory of Viscoelasticity 176 3.4 Bolotin-Like Approach for Nonlinear Dynamics 177 3.4.1 Straightforward Bolotin Approach 177 3.4.2 Modified Bolotin Approach 185 3.5 Regular and Singular Asymptotics in the Nonlinear Dynamics '" . " , ., 191 of Thin-Walled Structures 3.5.1 Circular Rings and Axisymmetric Cylindrical Shells ., 191 3.5.2 Reinforced and Isotropic Cylindrical Shells 197 3.5.3 Nonlinear Oscillations of a Cylindrical Panel 214
Contents 3.5.4
3.6 3.7 3.8 3.9
4.
Stability of Thin Spherical Shells Under Dynamic Loading 3.5.5 Asymptotic Investigation of the Nonlinear Dynamic Boundary Value Problem for a Rod One-Point Pade Approximants Using the Method of Boundary Condition Perturbation ..... Two-Point Pade Approximants: A Plate on Nonlinear Support Solitons and Soliton-Like Approaches in the Case of Strong Nonlinearity . . . . . . . . . . . . . . . . . . . . . . . . Nonlinear Analysis of Spatial Structures 3.9.1 Introduction 3.9.2 Modified Envelope Equation
XI
Discrete-Continuous Systems 4.1 Periodic Oscillations of Discrete-Continuous Systems with a Time Delay 4.1.1 The KBM Method 4.2 Simple Perturbation Technique 4.3 Nonlinear Behaviour of Electromechanical Systems 4.3.1 Introduction 4.3.2 Dynamics Equations 4.3.3 Averaging 4.3.4 Numerical Results -
General References
,
228 229 236 240 247 247 248 253 253 253 267 280 280 281 282 285 291
Detailed References (d) Index
219
299
,
303
1. Introduction: Some General Principles of Asymptotologyl
Almost any physical theory formulated in mathematical terms in a general way is extremely complicated. Therefore, both in creating a theory and in its further development, the simplest limiting cases that admit analytical solutions are of paramount importance. It is quite common that in the limiting case there are fewer equations or the (differential) equation has a lower order or the nonlinear equation is replaced by a linear one or the original system is subjected to a kind of averaging and so on and so forth. Behind the above-mentioned idealizations, however diverse they may seem, lies a high degree of symmetry inherent in a mathematical model of the phenomenon at issue in its limiting situation. An asymptotic approach to a complex and perhaps "insoluble" problem consists basically in treating an original - insufficiently symmetric - system as an approximation to a given symmetric one. It is basically important that the determination of corrections allows one to study deviations from the limiting case in a way which is much simpler than a direct study of the original system. At first sight, the potentialities of such an approach are limited by a narrow range of variations in the parameters of the system. The experience gained in the study of various physical systems has shown, however, that in the case of system parameters varying considerably and the system itself departing from one limiting symmetric pattern, in general another limiting system, often with a less pronounced symmetry, exists and a perturbed solution can now be formed for the latter one. This enables the system behaviour to be defined over the entire range of the parameter using a finite number of limiting cases. Such an approach makes the most of one's physical intuition and contributes to its further enrichment and also leads to the formation of new physical concepts. Thus the boundary layer - an important concept in fluid mechanics - is of a pronounced asymptotic nature and is related to the localization at the boundaries of a streamlined body in the zone where the viscosity of the fluid cannot be neglected (see Fig. 1.1 and also "Album of boundary layers" [3]). In the mechanics of a deformable rigid body and in the theory of electricity, similar phenomena are known as the edge effect and the skin effect, respectively.
1
See also [40, 45, 80, 105, 109, 142, 28d, 38d].
1. Introduction: Some General Principles of Asymptotology
2
Fig. 1.1. Boundary layer near a streamlined sphere
That the asymptotic method assists in relating different physical theories with one another is of little consequence. Albert Einstein would point out that "the happiest lot of physical theory is to serve as a basis for more general theory while remaining a limiting case thereof" . The above-mentioned problems will be clarified in this chapter.
1.1 An Illustrative Example t As an example of the technical aspect of the method, consider a simple algebraic example. The biquadratic equation
x4
-
2x 2
-
8= 0
(1.1.1 )
is reduced to a quadratic -equation and readily solved by setting z we have Xl,2
= ±2,
X3,4
= ±h i,
i=
= x 2 • Then
vCT.
Such a simplification is due to the symmetry of the equation: substituting (-x) for x does not change it. Let us assume that the original equation describes a given physical system with its parameters undergoing small changes and, as a consequence, the equation takes the form y4
_
cy3 _ 2y2 - 8 = O.
(1.1.2)
In this case the system is said to have received a small perturbation; the expression (- cy3) is referred to as the "perturbation term" and c as the "small parameter". The system becomes asymmetric, and the solution of the new equation can no longer be written in a simple form. The roots of the new equation, Yi (i = 1, ... ,4), however, should not differ significantly from
1.1 An Illustrative Example
3
hence set Yi = Xi' The error of such a substitution is determined by the value of the discarded term (-cy 3 ). To make the solution more accurate, let us represent it as a series Xi,
Yi
=
Xi
+ cY?) + c2 y?> + ... ;
i
= 1, ... ,4.
(1.1.3)
Substituting this expansion into the perturbation equation and equating the coefficients of the same power of c we find Yi
= O.25x~ 2 1; xi -
i
= 1, ... ,4.
(
1.1.4
)
Evaluation of corrections could be continued without any difficulty, but the deviation from the exact solution will inevitably increase with the increase in the value of c. Consider now the opposite case of large perturbations. Then the reciprocal c- 1 will be small. Then, the roots of equation (1.1.2) can be divided into two groups. As c- 1 tends to zero, three roots tend to zero and the fourth increases indefinitely. The two groups may be found using expansions in the small parameter c- 1 . Yl=c + ... , - 2c -1/3 Y2--
(1.1.5 )
+ ... ,
Y3,4=ac- 1/ 3 (1 ±
v'3 i).
(1.1.6)
There exists, however, a region where the asymptotic approximation produces unsatisfactory results. This is the region where "small" values of care already "large" and "large" values of c are still "small" (see Fig. 1.2). The problem of forming a solution within such a region on the basis of available limiting values is one of the most difficult when employing asymptotic methods as in the problem of deciding as to what is to be considered "small" or "large" . This will be considered later. Besides, it should be noted that perturbation solutions of a problem represented as expansions in series in the power of a small parameter of type (1.1.3) do not necessarily converge to the solution which is being sought. The expansions are often asymptotic. The ratio of each term of the series to the preceding one tends to zero when the expansion parameter approaches its limiting value, say, zero; and the deviation of the sum of the first N terms of such a series from tl:Ie function represented by the complete series is of the (N + l)th order. (In examining a series for convergence, the parameter is regarded as fixed and the limit of the sum of N terms of the series is taken as N tends to infinity.) In particular cases a divergent asymptotic series (with infinite limit) is sometimes more useful than a convergent one as only a few of the initial terms give a fair approximation. Let us consider some typical situations where the asymptotic approach is effective.
1. Introduction: Some General Principles of Asymptotology
4
6.----------r--------,.------:l
4
~-~--~-+I
--- Fig. 1.2. Comparision of exact (nu-
2
4
6
merical) and asymptotic solutions for the algebraic equation of fourth order
1.2 Reducing the Dimensionality of a System A high order of an algebraic or a differential equation or a large. number of such equations are all manifestations of one of the principal difficulties that arise in solving physical problems. This difficulty is sometimes called "the imprecation of dimensionality". In order to get over it, two antithetical approaches have been developed. The first one proves to be effective if individual elements of a system under consideration differ markedly from each other in one or another characteristic. Then, by introducing characteristics of different elements, one is able to carry out an asymptotic reduction of dimensionality, or in other words, a reduction in the degrees of freedom, and then one can try to improve the solution obtained by using the asymptotic approximation. A typical example of such a situation is a three-body problem in elastic mechanics. The masses of celestial bodies (say, those of the Sun, the planet Jupiter and the Earth), ~ a rule, differ markedly, and a small parameter the mass ratio - enables an asymptotic reduction of the dimensionality to be achieved. The classical methods of celestial mechanics are based on this, the limiting (high symmetry) case being the exactly solvable two-body problem. Celestial mechanics is the first branch of science where the asymptotic method (the perturbation theory) has played a dominant role, and moreover, this method was originally developed in response to the pressing necessity of solving problems in celestial mechanics. It should be noted that asymptotic methods are often used without being specifically regarded as such and even without being fully understood. Thus, one-degree-of-freedom models are employed extensively in engineering. Clearly, employing such models always involves an asymptotic reduction in the dimensionality and the possibility, at any rate in principle, of finding the corresponding corrections, but a clear indication that this is the case is rare. Let us now consider the second way of getting out of the difficulty.
1.3 Continualization
5
1.3 Continualization If a system under consideration consists of a set of homogeneous elements, then the asymptotic approach can be used not only for the reduction of dimensionality but also for increasing dimensionality. Thus, we approach a highly important class of physical models where discrete systems are replaced by continuous ones. As an example, let us consider the longitudinal oscillations of an infinite chain of the similar masses connected by springs of equal length L and rigidity C (Fig. 1.3a). With the smooth oscillation form characterized by the displacement Uk at each point kL (k = 0, ±1, ±2, ... ), the chain can be replaced by a continuous rod, thus enabling us to change from the infinite system of ordinary differential equations mUktt
= C(UIc+l -
2Uk
+ Uk-I)
to the single partial differential equation mUtt =
cL 2 u:r,x'
The degrees of freedom have grown in number (the continuum replacing the countable set), and the relative simplicity of this limiting case of longwave oscillations is due to the symmetry of the partial differential equation not varying under an arbitrary displacement along the rod. As the period of oscillations and their wavelengths will decrease, the error of the approximate solution obtained in this way will increase. Another limiting case for the same system is for the minimum possible wavelength oscillations (Fig. 1.3b). Their form can be readily calculated and employed as the first approximation in the study of the short-wave oscillations of the system. In this case the desired solution should have the form of the product of the solutions of the limiting case on the smooth function which is deduced from the partial differential equation (Fig. 1.3c). The method of transition from discrete models to continuous models has found extensive applications in physics, and the entire mechanics of the continuum is based essentially on this method. This is not always so, however, as in th~ case under consideration. Fluids, say, do not lend themselves to the purpose of defining a periodic equilibrium structure in reference to which oscillations are executed. Nevertheless, at a macroscopic level we perceive the fluid flow as a continuum flow which can be simulated by a continuous fluid model. It is true that continuity is provided by the averaging of small-scale (microscopic) movements. The consequences of such an averaging will be discussed below. This will show now that the transition to the differential equations of hydrodynamics becomes possible. In conclusion let us quote from Erwin Schrodinger who figuratively explains the efficiency of the method: "Let's assume we would tell an ancient Greek that the individual particle path in a fluid could be traced. The ancient Greek would not believe that man's limited intellect could solve such
6
1. Introduction: Some General Principles of Asymptotology
a)
j-2
j-I
j
j+1 j+2
b) y
c)
y
Ol---~--+---r---f-----lt---------i""
x
Fig. 1.3. (a) Infinite chain; (b) minimum possible wavelength oscillations; (c) short-wave oscillations
an intricate problem. The point is that we have learned to master the whole of the process using but a single differential equation" .
1.4 Averaging In many physical problems, some variables vary very slowly, others more rapidly., It is natural to bring the question of whether it is appropriate to study first the global structure under consideration, digressing from its local distinctive features, and then to investigate the system locally. It is the averaging method that is aimed at the division of the fast and slow components of the solution. Without going into the details of the method - the more so because it has at present many modifications - it will be noted only that it involves the introduction of the "slow" (macroscopic) and the "rapid"
1.5 Renormalization
7
,,, ," , , , , ", r\ / , , , , , ", ,, ,, ,, ,," ,,, ,,, ,,, ,,, , , ,, ,, ," ,, " "" " ...-< ~
/
/
e-+O
-+-----iI----+---4 ----..
//
\
)'
V
Fig. 1.4. Homogenization procedure: from the periodic inhomogeneity to the homogeneous material with new properties (microscopic) variables whose equations are separated and can be solved independently, or sequentially (see Fig. 1.4). This method was developed for and gained wide use in solving problems in celestial mechanics and nonlinear oscillation theory that are defined by common differential equations. At present, the method is used to great advantage for solving variable-coefficient partial differential equations in such disciplines as the theory of composites, or the design of reinforced, corrugated, perforated, etc., shells. An original nonhomogeneous medium or structure is reduced to a homogeneous one (generally speaking, to an isotropic one) with some effective characteristics. The averaging method allows one not only to obtain the effective characteristics but also to investigate the nonhomogeneous distribution of mechanical stresses in different materials and structures, which is of great significance for evaluating their strength.
1.5 Renormalization Regrettably, the simple averaging of small-scale movements is not always applicable, either. There occur such problems wherein several different-scale movements show up markedly even at the macroscopic level. Among these is, for example, the study of what is known as critical phenomena related to phase transitions, or the study of turbulence. In this case, a number of successive averaging procedures for all scales has to be carried out. This is the very essence of the renormalization procedure which forms the renormalization group method. A rigorous renormalization of the procedure, however, involves considerable technical difficulties. A practical solution of the problem is offered by a quite unexpected asymptotic method. The fact is that in a four-dimensional imaginary world these problems do not occur, and this makes it possible to carry out an ordinary averaging. This case could be considered as a limiting one, with the quantity e = 4 - d (where d is the spatial dimensionality) as a small parameter. In the real three-dimensional world d = 3, and e = 1, which is not small. Nevertheless,
8
1. Introduction: Some General Principles of Asymptotology
an asymptotic expansion in the parameter proved to be quite effective in solving the most complicated problems of critical-phenomena physics.
1.6 Localization Real system deviations from the limiting (i.e., ideal) system may be of a different nature. Sometimes these deviations are small over the entire range of the system parameter variations: it is not infrequent, however, that the deviations are high, although localized within a small region. This is true for the above instance of a body streamlined by a fluid. Another example is the transition (reducing) from the three-dimensional model of an elastic body to the two-dimensional model (plates, shells), or to the one-dimensional model (rodes, beams). In this case a narrow boundary layer (of the order of the plate or shell wall thickness; or of the cross-sectional characteristic size of the rod or the beam) exists near the body boundaries wherein the three-dimensionality of the original problem manifests itself. Upon reducing the three-dimensional problem to the two-dimensional one, it is still possible to isolate the so-called end effects concentrated at the shell boundaries or its structural inhomogeneities. The concept of the boundary layer is closely related to the so-called St. Venant principle that says that in the analysis of a structure it is possible to digress from the detailed load distribution pattern in fixing its elements. In fact, however, the distribution pattern is essential, but within narrow zones only more extension is defined by the element crosssectional characteristic sizes or by the load-variation period. Mathematically, defining a boundary layer is due to the fact that a simplified differential equation is of a smaller order than the original one. The asymptotic approach in this case is termed singular.
1.7 Linearization If the equations of a physical theory are nonlinear, then even a small number of degrees of freedom or a localized solution do not ensure the overcoming of mathematical difficulties. The problem is solved by linearization - an asymptotic method - that relies on the concept of low-intensity processes. A linear approach (to the problem) allows one to formulate such fundamental concepts as the normal vibration, the eigenfunction and the spectrum. For a linear system with n degrees of freeciom with no damping, one can always choose such "normal" coordinates which describe the system by n oscillations for pendulums not linked to one another. In other words, any motion of a linear system is represented by a linear combination of normal oscillations (or waves), that is, by the so-called expansion in a Fourier series.
1.8 Parle Approximants
9
It is of fundamental importance that the oscillations are singled out not only mathematically but also physically. Thus, it is precisely the normal oscillations that will resonate under the influence of the periodic force. If we consider a linear system as the first approximation to the nonlinear one (that is the crux of local linearization) then, when taking into account the nonlinear corrections in the equations of the second and following approximations, there appear dummy external loads that bring about the normal oscillation resonanCes. This can be avoided by "touching up" the parameters of the normal linear oscillation. However, nonlinear systems, especially high-dimensional ones, quite often do not lend themselves to correct the description in the approximation of the local linearization method. Thus, the combination of high dimensionality with strong nonlinearlity was until recently considered an insurmountable difficulty in carrying out a structural study of a physical system. But a fairly extensive class of multidimensional nonlinear systems that permit such a study has been recently discovered. These systems, known as "integrable systems" , have particular solutions as stable solitary waves - solitons - that are in a way analogues of normal oscillations defined in linear systems. Thus, a nonlinear generalization of the Fourier method - the method of the inverse scattering problem - is based on solitons which playa fundamental role taking the place of the usual Fourier components. The method of the inverse problem of scattering can be treated as the nonlocallinearization of the original nonlinear equation. In other words, the latent instability of a nonlinear system makes it possible to find a transformation that reduces the construction of an extensive class of solutions to the analysis of linear equations. The integrable systems can in their turn act as an approximation in the analysis of the systems that approximate them, but are nonintegrable within the framework of an asymptotic approach.
1.8 Pade Approximants So far we have assured ourselves that practically any physical problem, whose parameters include the variable parameter c, can be approximately solved as c approaches zero or infinity. How is this "limiting" information to be used in the study of a system at intermittent values of c, say, c = I? This problem is one of the most complicated in asymptotic analysis. As yet there is no general answer to the tricky question of how far the parameter € can be considered small (or large) in the problem involved, though in many instances this problem is alleviated by the so-called two-point Pade approximants. The two-point Pade approximant is represented by the rational function F(c), in which the m + 1 coefficients of expansion in the Taylor series when c ~ 0, and the m coefficients in the Lorentz series when c ~ 00 coincide with the corresponding coefficients of the governing series for F(c).
10
1. Introduction: Some General Principles of Asymptotology
Experience shows that the Pade approximants do indeed quite often allow the limiting expansions to be "sewn together" after defining the regions of "small" and "large" values of e. This resembles a known interpolation procedure, that is, the reconstruction of the intermediate values of a quantity by its two extreme values. The role of such known values is played in this case by the asymptotics as e tends to zero and to infinity (see Fig. 1.5).
F(e)
e--+oo
e--+O
two-point Pade approximants
CD
® e
Fig. 1.5. Matching of limiting asymptotics by two-point Parle approximants
For instance, for equation (1.1.2) the"Pade approximant for the first root YI
=
(2
+ O.57e + O.12e 2 )(1 + O.12e)-1
derived on the basis of the asymptotics of the form (1.1.3) as e tends to zero or of the form (1.1.5) as e tends to infinity, defines satisfactorily the exact solution at any value of e (Fig. 1.2, curve 1, in this case the exact (numerical) solution and the two-point'Pade approximant solution practically coincide).
1.9 Modern Computers and Asymptotic Methods The reader must have repeatedly asked themself a question: are the asymptotic methods of any practical use at all when there are computers? Is it not simpler to write a program for any original problem to solve it numerically by using standard procedures? This may be answered as follows: first, asymptotic methods are very useful in the preliminary stage of solving a problem even in cases where the principal aim is to obtain numerical results. The asymptotic analysis makes it possible to choose the best numerical method and gain an understanding of a vast
1.11 Problems and Perspectives
11
body of numerical material, though not properly arranged. Secondly, asymptotic methods are especially effective in those regions of parameter values where machine computations are faced with serious difficulties. Laplace used to say, not without reason, that asymptotic methods are the "more accurate, the more they are needed". Moreover, there is the possibility of developing such algorithms wherein smooth portions of solutions are obtained numerically, and the asymptotic approaches are applied to those parameter value regions where these solutions change drastically, say, within boundary layers. Thirdly, the asymptotic methods develop our intuition in every possible way and play, as noted above, an important role in shaping the mentality of, say, a contemporary scientist or engineer. Therefore, it would be more proper to consider the asymptotic and numerical methods not as competing, but as mutually complementary. Again, computers further considerably the development of the asymptotic method. For instance, defining higher approximations is a major difficulty in applying asymptotic methods. In solving complex problems by manual calculations one may succeed in defining two or three approximations at the most. Now the burden of manual calculations can be shouldered by the computer.
1.10 Asymptotic Methods and Teaching Physics "Few of the equations of physics have exact solutions which are manageable, and one usually has to have recourse either to approximate methods or to numerical solutions. Numerical work becomes cumbersome if the problem has a great number of variables, or if one is interested in a general survey of possible solutions. In those cases the natural approach is by approximation. In teaching physics we probably overemphasize the exceptional problems which have closed solutions in terms of elementary functions, and do not give enough attention to the more common situation in which approximations have to be used. Beginners are usually uncomfortable with approximations, and, even if only an approximate answer is required, often prefer to find the exact answer, if this is possible, and then to approximate. This is understandable because the art of choosing a suitable approximation, of checking its consistency (e.g. ensuring there are no oscillations) and finding at least intuitive reasons for expecting the approximation to be satisfactory, is much more than solving an equation exactly" [130].
1.11 Problems and Perspectives Naturally, we do not describe all new asymptotic approaches successfully used in the theory of nonlinear oscillations. We only mention algorithms of
12
1. Introduction: Some General Principles of Asymptotology
conjugate operators and equations [115, 57d], methods on the basis of the Lie group [165], renormalization group and intermediate asymptotics [4648], and the asymptotic-numerical approach [60d]. The last one has, in our opinion, a great future, because it gives the possibility of using the merits of the asymptotical and numerical approaches simultaneously. The most interesting discoveries ill the field under consideration, in our opinion, are closely connected with developments in the theory of summation of the extrapolation and interpolation of asymptotic series, especially in the theory of Pade approximants and its generalization, and in searching for new parameters of asymptotic expansions. We have analysed the merits and demerits of Parle approximants in Sect. 2.12. The following parameters are used as asymptotics in this book: the difference of frequencies WI - W2 (Sect. 2.10.2); the amplitude A, A small and A large (Sect. 2.12.4); the parameter fJ (Xl+ 6 , Sect. 2.11.1); the power of nonlinearity N (X N , Sect. 2.11.3); the ratio of flexible to extension rigidities (EI)I(EFR2) (Sect. 3.5.1); the ratio of the thickness to the radius of a shell, hiR (Sect. 3.5.2); the ratio of the typical size of nonhomogenity to the cell size (Sect. 3.3); the ratio of the structure size, alb (Sect. 3.5.3); II A, where A is a large frequency (Sect. 3.4); the ratio of foundation and plate rigidities, (cL5)1 D -+ 0 and (cL5)1 D -+ 00 (Sect. 3.7); and the formal small parameter c (Sect. 3.6). Successive use of the asymptotic approach in nonlinear dynamics is closely linked with the choice of parameters for asymptotic expansions, and this problem is one of the most interesting and attractive in this field.
2. Discrete Systems
2.1 The Classical Perturbation Technique: an Introduction The classical method of perturbation is based on the assumption that the influence of the nonlinear part of the considered differential equations is small in comparison to the influence of the linear part of the equations, or that the oscillation amplitude is small. The perturbation technique can also be used even if the deviations from the true (sought) solution are not small, but are localized in a small space. This is emphasized by the formal or natural introduction of the "small" perturbation parameter c to the differential equation. The solution of the equations are sought in the form of power series because of the parameter c (for c = 0 the fundamental solution - the first term of the required series - is the solution to the linear differential equation). The next solution components, standing by the successive powers of c, are obtained from the recurrent sequence of linear differential equations with constant coefficients. The main idea of the perturbation technique is focused on the asymptotic reduction of the dimension of the dynamical system, or in the language of mechanics, on the reduction of the degrees of freedom of the system. Generally, such an approach can be initiated when the system can be divided into subsystems which are different because of their dynamical characteristics (slow and quick motions, soft and hard types of stiffness characteristics, small and large damping of the system), which allows for the introduction of one or a few perturbation parameters. Another positive aspect of the perturbation technique is connected with the general properties' of nonlinear dynamical systems. Only in very rare cases are we able to find the analytical solutions of the systems governed by a nonlinear differential equation. For instance, up to now it has been impossible to describe analytically a chaotic solution. Even when we sometimes have such a solution, it is approximated by complex functions, and in practice the benefit of such an approach is doubtfuL Additionally, supposing that we have some of the particular solutions of the analysed systems, the superposition rule does not work for nonlinear systems, and it is impossible to find a solution
2. Discrete Systems
14
for the arbitrarily taken initial conditions. The asymptotic series, however, sometimes allows us to overcome these problems. Further we introduce so-called local asymptotic linearization, which is based on the linear first approximation to the nonlinear system. Recently other interesting nonlinear behaviour has been detected during an analysis of strong high-dimensional nonlinear systems. In spite of their complicated form, they sometimes possess partial solutions, called solitons, which can serve as a start to the asymptotic (non local) approach. The main advantages of asymptotic (perturbation) analysis are as follows: a) an analytical form of the solution; b) the solutions can serve as the initial solutions for numerical simulations; c) the perturbation approach can serve as a tool for establishing the physical and engineering meaning of the dynamics. The main weak point of the asymptotical approach lies in the fact that formally it is a very difficult task to prove a strict connection between an exact solution and that described by the asymptotic series. It can happen that because of the character of nonlinearities a periodic solution of the system for e = 0 does not occur for e =I- O. It can also happen that a few or infinitely many solutions for e =I- 0 correspond to a periodic solution for e = O. The most difficult task connected with the application of the asymptotic series is to prove the convergence of the series in a wide enough interval, as well as to estimate the validity of the assumption that e is a small enough parameter. Even if such a proof is successfully done, usually it is based on inequality chains, which is somehow not convenient enough for further analysis. Usually such problems are omitted, using instead numerical computations to test the validity of the results. The intr
ii + a~y = eQ(y).
(2.1.1)
In order to simplify the calculations, we consider a trivial solution at y = 0, which leads to Q(O) = O. We develop a function Q(y) in a Taylor series in the vicinity of the equilibrium point y = 0, where (dQ/dy)y=o = O. The solution to (2.1.1) is sought in the form K
y = yo(t)
+L
ekYk(t).
(2.1.2)
k=l In order to eliminate the secular terms (unrestrictedly growing in time), we introduce the additional series K
0:
2
=
o:~ +
L ekak'
k=l
(2.1.3)
2.1 The Classical Perturbation Technique: an Introduction
15
Taking into account (2.1.2) and (2.1.3) in (2.1.1), we get
f~ ekYk + (02- f~ ekok) f~ ekYk = eQ (f~ hk) .
(2.1.4)
Then we develop the right-hand side of (2.1.4) into a power series because of the small parameter e in the vicinity of e = 0 and we get
eQ(y) =eQ (~ekYk) I.~o+ {Q (~ekYk) +eQ' (~ekYk) .f?eHYk} I.~oe + {Q' (f~ekYk) f;kek-1Yk +Q' (f~ekYk) f?eHYk +eQ" (f(Yk) f;ke k-1Yk + eQ'
(t
ekYk)
k
k(k - 1)e - 2Yk } Il!:=o e;
+ ...
k=2
k=l
=
t
e2
Q(Yo)e + 2Q'(YO)Yl2" +
,
(2.1.5)
where: Q' = dQ/dy, Q" = d 2 Q/dy 2 , . Taking into account (2.1.5) and having compared the terms standing near the same powers on the left and right hand sides of (2.1.4), we obtain cO
e1 e2
Yo + a 2 Yo = 0, Yl + a?Yl = alYo + Q(yo), Y2 + a 2 Y2 = a2YO + alYI + YIQ'(yO),
(2.1.6)
The solution to the first equation of the recurrent set (2.1.6) is the function (2.1.7)
Yo = ao cos 'It, where 'It = at
+ eo.
(2.1.8)
Taking into account (2.1. 7) in the second equation of (2.1.6), we get 2 YI + a YI = aoao cos'lt + Q(ao cos 'It). (2.1.9) The function Q(ao cos'lt) is a periodic function because of 'It with the period 21r, therefore we can develop it into the Fourier series 1
Q(ao cos 'It)
=
2bo
00
+ 2: bn cosn'lt, n=l
(2.1.10)
2. Discrete Systems
16
where '11"
bn =
~/
(2.1.11)
Q(ao cos!li) cosn!li d!li.
o
Taking into account (2.1.10) in (2.1.9), we get
ih + 0:2 Y1
1
+ (cHaO + b1 (ao)) cosYi' + L 00
= "2bo
bn cosn!li.
(2.1.12)
n=2
In order to get a periodic solution we should eliminate the secular term from Yd t ), which leads to the condition
O:laO + b1 (ao) = O.
(2.1.13)
From this equation we get
b1 (ao) (2.1.14) , ao and the first unknown coefficient of (2.1.13) is estimated. Therefore, a solution to (2.1.12) has the form 0:1
= -
~
o + ed + -2 + L..,., b
Yl = al cos(o:t
n=2 0:
20:
bn (
2 -
no:
)2 cosn!li.
(2.1.15)
The above constants ao and eo are defined by the initial conditions, and we take al = e 1 = O. This leads to the determination of ao and eo from (2.1.2), which serves to obtain the cons'tants ao and eo on the basis of the assumption that Yo(t) fulfil the initial conditions, while ak and ek are found from the condition that Yk(tO) = O. Further, we take al = e 1 = 0 and from (2.1.15) we get
bo Yl (t ) = -2 2 0:
+
~
L..,., n=2
bn 2 ( )2 cos n!li. 0: .,- no:
(2.1.16)
Therefore, the right-hand side of the third equation of (2.1.5) is defined. On the basis of the condition of avoiding a secular term in the solution Y2(t),'we get the coefficient 0:2, and the solution of that equation is the next term of the sequence (2.1.2). After a limitation to the first approximation O(c2 ), we get Y = ao cos(o:t
1[1
+ eo) + c 0:2
00 b ] "2bo + ~ 1 _nn 2 cosn(o:t + ne o) , (2.1.17)
where (2.1.18)
2.1 The Classical Perturbation Technique: an Introduction
17
The period of the sought solution is 21r
(2.1.19)
T = --;::==== . / a2 _ 0
V
e bdel
(»
el()
and depends on the oscillation amplitude ao.
Example 2.1.1. For the mechanical system presented in Figure 2.1 calculate the oscillation period and find its oscillations analytically. y
Fig. 2.1. One-degree-of-freedom conservative oscillator The equation of motion has the form 2 Y.. + CioY = -ey 3 ,
(2.1.20)
where
k Llk e=-. m m The recurrent set of equations is given below 2
a'Q=-'
Yo .•
Yl
ih
+ a 2 yo 2
+ a Yl
+ a 2Y2
= 0,
(2.1.21 )
(2.1.22) 3
= alYo - Yo'
(2.1.23)
= a2Yo - CilYl - 3Y~Yl'
(2.1.24)
The solution to (2.1.22) is
Yo
= ao cos !V,
!V
= at + eo.
(2.1.25)
Taking into account (2.1.25) in (2.1.23) and on the basis of the equation 3
cos !V =
3
1
4 cos !V + 4 cos 3!V,
(2.1.26)
we get
.
Yl
+ Ci 2Yl
= ao
(al -
2) cos~.T, -
43 ao
3
1 cos 3~. .T, 4ao
(2.1.27)
2. Discrete Systems
18
From the condition of avoiding the secular form we find that al =
3
2
(2.1.28)
'4 ao ,
and taking into account
= 8 1 = 0, we get
al
a3 Yl = ~ cos 3rJt. 32a Taking into account (2.1.28) and (2.1.29) in (2.1.24), we obtain
ih + a 2 Y2
+
= a2 aO cos rJt
5 3ao5 ao 2 2 cos 3rJt - 3- 3 2 cos 3rJt cos rJt. 128a 2a
(2.1.29)
(2.1.30)
Because 1
cos2 rJt cos 3rJt = - (cos 5rJt + 2 cos 3rJt 4 therefore
ih + a 2 Y2
=
+ cos rJt),
(2.1.31)
3aB ) 3aB a2 aO + 128a 2 cos 3rJt - 128a 2 cos 3rJt
(
3a 2
_....::0--:128a 2
cos 5rJt.
(2.1.32)
From (2.1.32) we get a 2 -
and for
3a04 128a 2 ' a2 =
(2.1.33)
82 = 0 we obtain
,,
3 aB Y2 = 1024 a2'
(2.1.34)
Finally, we get Y = ao cos rJt
a3
+ c 32a, O cos 3rJt + c 2
a5
2
0
1024a
2
(3 cos 3rJt + cos 5rJt),
(2.1.35)
where a
=
2
3 4
2
3 a~ 3"" 2 128 a o + c4a6
a o + c- a o + c 2
rJt
= at + 80.
(2.i.36)
The oscillation amplitude, which corresponds here to the maximum hangout from the equilibrium position, is obtained for the following time moments _ n1T' - 8 0 tn (2.1.37) (rJt = at + 80 = n1T'). a The period of oscillation is given by the formula T =
21T'
-,=========
(2.1.38)
2.2 Krylov-Bogolubov-Mitropolskij Method
19
In order to estimate the corrections introduced by the successive approximation, we take E = ao = ao = 1 and we calculate: A. For
we have
e(O),
T(O)
B. For eO), we have
T(l)
ea~/320~
C. For
= 1.01778; we have T(2)
e(2),
= 6.28 for every amplitude a(O);
21r I Ja~
=
+ ~ea~
= 21r la = 4.7293 for
= 4.7472 for A(2)
A(1)
ao
+
= 1.035156.
2.2 Krylov-Bogolubov-Mitropolskij Method This method can be applied to systems of second-order ordinary differential equations. Its fundamental parts will be outlined on the basis of an example of one-degree-of-freedom autonomous systems of the form
ii + a~y = eQ(y, iJ)·
(2.2.1 )
Let us suppose that y = iJ = 0 is the equilibrium position and the function Q(y, iJ) is analytkal because of its variables. The main difference between this and the Krylov method lies in the assumption that the amplitude and the phase are functions of time. We are looking for the solution K
y(t)
=
2: ekYk [a(t), lJi(t)].
a cos IJi +
(2.2.2)
k=l
For conservative systems the amplitude is constant, and da = 0
dt
(2.2.3)
.
The derivative of IJi with respect to time is also constant for conservative systems and can be approximated by the series K
IJi•
= 0 =
ao +
'L...J "'
ek ak(a).
(2.2.4)
k=l
However, for nonconservative systems we introduce the following series K
a=
2: ekAk[a(t)], 'k=l
(2.2.5)
. K
tP =
00
+ 2: ekBk[a(t)].
(2.2.6)
k=l
From (2.2.2) one obtains K
iJ
=
acos IJi - atP sin IJi + 2: ek k=l
(a:ak a + aaYkIJi tP) ,
(2.2.7)
20
2. Discrete Systems
jj =
acos IJI - 2iLrj, sin IJI -
a¥ sin IJI - alJl 2 cos IJI 2 ~ k (82Yk . 2 + 8 2Yk a'ITI + 8Yk .. + 8 Yk rj,' + LJ e 8a2 a 8a81J1 '1' 8a a 81J18a a k=l
2 8 + 81J1Yk2 IJI. 2
Yk
+ 881J11J1")
.
(2.2.8)
Taking into account (2.2.2) and (2.2.8) for the left-hand of side (2.2.1), we get
L=
[a - (rj,2 -
a~)a] cos IJI - (2iLrj, + a¥) sin IJI
2 2 k (82Yk ·2 + 8 Yk rj,2 + 2' rj, 8 Yk + 8Yk .. + LJ e 8a 2 a 81J12 a aa81J1 8a a
~
k=l
2) .
8Yk ..
(2.2.9)
+ 81J1 IJI + aoYk
According to (2.2.5) and (2.2.6), we get KKK
¥ = Lek~kiL = Lek~k LekA k = e2~1 Al + O(e k=l
k=l
. 3
),
(2.2.10)
k=l
and after some transformations one obtains L ="
[cr~ (~::~ + YI) - 2croB Ia cos q; -
2[ 2 +e 0 0
(881J12
+ (AI d~l
2y 2
+ Y2
2
jpYI 8 YI + 2a oB 1 81J1 2 + 20 0 A 1 8a81J1
)
- (2croB,
2croA I sin q;]
+ B~)a) cosq;
- (2cr oA 2 +2A I B I +
d:al Ala) Sinq;] +0(,,2).
(2.2.11)
We develop the right-hand side P of (2.2.1) into a power series because of the parameter e in the vicinity of e = 0 of the form .
P = eQ(y, y)
'_ {. (88yQ Q(y, y) IE=O - e
+
dy aQ dy ) 2 } de IE=O + 8y de IE=o + O(e )
= "Q (a cos q;, -acro sin q;) + ,,2 [ : (a cos q;, -acro sin q;)Yl . IJI) ( A I cos IJI, -aB I sm . IJI + + aQ ay (a cos IJI - aoo sm +O(e 2).
Y1 00 8 81J1 ) ]
(2.2.12)
2.2 Krylov-Bogolubov-Mitropolskij Method
21
Equating the terms of the same powers of c, we get the following recurrent set of differential equations
8 2 Yl -8 2 +Yl 1ft 8 2Y2
--2
81ft
+2Alaosmlft+2BlaOacos~,
.
(2.2.13)
1
. + 2A 2a o sm 1ft + 2B2 a Oa cos 1ft).
(2.2.14)
= 2 (II
+ Y2
IT,)
1
2(10 ao
=
ao
and so on, where: (2.2.15)
fo = Q(a cos 1ft, -aao sin 1ft),
II
2
2
8 -Yl 8 Yl - (A dA-l - B 2 a) cos 1ft o = -2B l a o 2A a l l l 81ft 2 8a81ft da
+ ( 2A I B I + aA I ~I )
sin if' +
(2.2.16)
~~ (a cos if', -a",o sin if')YI
+ ~~ (a cos if', -a",o sin if') ( Al cos if' - aB I sin if' + "'0 ~ ) , The function defined by (2.2.15) is periodic with period 21T" and it can be approximated by the Fourier series of the form fo(a,lft)
=
1 "2boo(a)
00
+ 2:(bon(a) cosnlft + Con (a) sin nlft),
(2.2.17)
n=l
where the coefficients are defined below bon (a) =
~/
271'
Q(a cos 1ft, -aao sin 1ft) cosnlft dlft,
(2.2.18)
Q(a cos 1ft, -aao sin 1ft) sin nlft dlft,
n = 0,1,2, .... (2.2.19)
o 271'
Con(a) =
~/
o Taking into account (2.2.17), (2.2.18) and (2.2.19), we have 8 2 Yl ,T,2
8~
1 1 + Yl = 2 (COl + 2A 1a o) sin 1ft + 2(b ol + 2Bl a oa) cos 1ft ao ao
2: (bon cosnlft + Con sm nlft). 00
boo 1 +-22 + 2 a . a
o
0
•
(2.2.20)
n=2
The first two terms standing on the right-hand side of (2.2.20) grow unrestrictedly in time and their occurrence contradicts the assumptions of the asymptotic method. Therefore, we equate them to zero, which leads to the determination of the unknown function A1(a) and Bl(a) according to the formulas __ Col (a) A 1 (a ) , (2.2.21 ) 2ao
22
2. Discrete Systems
B (a) = _ bo1 (a). 1
(2.2.22)
200a
On the basis of these results, the general solution to (2.2.13) is
Yl = al cos(!li
boo
+ fh) + -202 0
1
1
00
+"2" '"' 2 (bon cosn!li + Con sin n!li). a L..."l-n
(2.2.23)
o n=2 Taking al = fh = 0, the above equation can be reduced to the form 00
Yl = -boo 2 20 0
. n!li ) . + 21 '"' L..." (bon cos n!li + Con sm
(2.2.24)
a o n=2
It is not difficult to see that now the function Yl defined by (2.2.24) is known thanks to (2.2.18) and (2.2.19) and it has period 211" because of the variable !li. Now we develop the function II into the Fourier series 1
lI(a,!li)
00
= 2blO(a) + L(b 1n(a)cosn!li + cln(a)sinn!li),
(2.2.25)
n=1 where
! II ! 27l"
b1n (a) = :
(2.2.26)
(a,!li) cos n!li d!li,
o
27l"
Cln(a) = :
11 (a,!li) sin n!li d!li,
n = 0,1,2, ... ,
(2.2.27)
o Making calculations in a similar way, we define
A 2 (a) = _ Cll (a),
(2.2.28)
B2(a) = _ bll (a) . 2°oa and the function
(2.2.29)
200
Y2 =
blO 1 22 +2 0 0
0 0
~ 1 L..." 1 2 (bIn cos n!li + Cln sin n!li). n=2 - n
(2.2.30)
We can define the unknown function a(t) and !li(t) occurring in the series (2.2.2) by solving the differential equations (2.2.5) and (2.2.6). Separating the variables in (2.2.5), we get dt =
da kA )' L...m=1 C k(a and then, after integration, we obtain "",M
(2.2.31)
2.2 Krylov-Bogolubov-Mitropolskij Method
23
da. (2.2.32) ""M leA () +a 0, 6m=1 C k a where ao is a constant dependent on the initial conditions. However, even in the case when the integral given by (2.2.32) can be defined using elementary functions it will be difficult to find explicit formulas necessary to solve (2.2.5). If it is possible to get an explicit expression for a(t), then from (2.2.6) one obtains t-
rl' =
!
! (<>0+ ~ okBkla(tll)
dt + eo.
(2.2.33)
Taking into account the above results we get the general solution defined by (2.2.2) with the constants ao and dependent on the initial conditions.
eo
Example 2.2.1. Determine an analytical form of the oscillations governed by the Rayleigh equation of the form .. + 0:02 Y = (2h - gy'2)'y. Y
(2.2.34)
We formally introduce the parameter c, and the above equation has the form (2.2.1) already discussed, where
cQ(y,iJ) = c(2h - giJ2)iJ.
(2.2.35)
We are looking for a solution of the form
y(t) = a cos rJt + CYI (arJt), a = cAl (a), .p = 0:0 + cB 1 ( a ),
(2.2.36) (2.2.37) (2.2.38)
and according to (2.2.13) we get
aarJt2Yl + Yl = 2
1 o:~ (fo
.
+ 2Al o:Osm rJt + 2B l O:Oa cos rJt),
(2.2.39)
where
fo = Q( a cos rJt - ao:o sin rJt) = =0
(~ga2<>~ -
2h) sin rl' -
~ga3<>~ sin 3rl'.
(2.2.40)
Substituting (2.2.40) in (2.2.39), We get
~~; + Yl = ~~ {[2A <>0 + =0 Gga2~ - 2h)] sinrl' 1
+2BlO:OacosrJt -
~ga3o:~sin3rJt}.
(2.2.41)
From the condition of avoiding secular terms, we determine the unknown coefficients
2. Discrete Systems
24
~a (2h - ~ga20!~) ,
Al =
B l = O.
(2.2.42) (2.2.43)
The solution to (2.2.41) because of (2.2.42) and (2.2.43) takes the form
Yl
= 312ga3o:o sin 3!Ii.
(2.2.44)
From (2.2.37), taking into account (2.2.42), we get
a = cah(1 - K 2a 2),
(2.2.45)
where
K 2=
~go:~
(2.2.46)
8 h If K 2 > 0, then we obtain da ----:-:---=-:---:- = c h dt. a(1 - Ka)(1 + Ka) The above equation can be presented in the form
f
(2.2.47)
da+Kf da _ K j da +lnL=cht, a 2 l-Ka 2 I+Ka
(2.2.48)
where L is the integration constant. After integration we get
aL h In --;::=::::;:;;~ Vl- K2 a 2 = c t,
(2.2.49)
and then
a=
aoe Eht
-/1 + a~K2e2Eht ,
= L -1 is 'IjJ = O:ot + eo,
where ao
(2.2.50)
a constant. The phase !Ii can be approximated by (2.2.51 )
and both ao and eo are defined by the initial conditions.
2.3 Equivalent Linearization Leaving only the first term in the series (2.2.2), we have
y(t) = a cos !Ii,
(2.3.1)
and in the series (2.2.5) and (2.2.6) we take only the terms
a = cAl (a), q; = 00 + cBl(a).
(2.3.2) (2.3.3)
Then the solution (2.3.1) will be the first simplified approximation solution of (2.2.1). We prove below that the solution (2.3.1) of (2.2.1) fulfils the equation
2.3 Equivalent Linearization
jj
+ 2he(a)y + a; (a)y
=
O(e).
25
(2.3.4)
We call the above equation the equivalent linear approximation to the nonlinear equation. Two parameters appearing in (2.3.4), the equivalent unit damping coefficient he and the frequency 0e, are defined as follows 211"
he(a) =
e
21T" 0 0a
(2.3.5)
!Q(acosrJt,-aoosinrJt)SinrJtdrJt, o 271'
Oe(a) = 00 -
e
21T"o oa
! Q(a cos rJt, -aao sin rJt) cos rJt drJt.
(2.3.6)
o
From (2.3.1) we have
y = a cos rJt - arP sin rJt.
(2.3.7)
Taking into account (2.2.21) and (2.3.5), the equation (2.3.2) will take the form
a = -ahe(a).
(2.3.8)
Taking into account (2.2.22) and (2.3.6), the equation (2.3.3) will take the form (2.3.9)
rP = ae(a). Taking into account (2.3.5) and (2.3.6) for (2.3.1), we obtain
(2.3.10)
y = -ahe cos rJt - aae sin rJt. Differentiating (2.3.7) we get ..
.h
.T,
dh e
IT,
Y = -a e cos ~ - a da cos ~
+ a he~ sm ~ .T,
•
IT,
•
•
IT,
- aae sm ~
dOe . -a da asinrJt - aoerJt cos rJt,
(2.3.11)
and after taking into account (2.3.8) and (2.3.9), we get jj = ah; cos rJt
+ 2h eaae sin rJt -
aa; cos rJt + a 2 he ~~e cos rJt
+a 2 he ~:e sin rJt - 2h;a cos rJt - 2h ea e sin rJt + a;acos rJt.
(2.3.12)
Finally, we obtain jj
+ 2h ey +
o;y = -ah; cosrJt + a he ddhea cosrJt + a he dhdae sinrJt. 2
2
(2.3.13)
The right-hand side of (2.3.13) is of order e2 , because according to (2.3.5) and (2.3.6) the expressions he, dhe/da, doe/da are of order e. To summarize, We have illustrated that on the basis of the approximations introduced by (2.3.2) and (2.3.3) We reduce the problem governed by the nonlinear equation (2.2.1) to an equivalent linear one. Equation (2.3.13) possesses a periodic solution if
2. Discrete Systems
26
(2.3.14) and the solution to the above algebraic nonlinear equation is an amplitude ao of the periodic solutions of the form Yo
= ao cos[O:e(ao)t].
(2.3.15)
In the case when he(a)
= 0, the solution is given by
y = Acos(O:e(ao)t + (}),
(2.3.16)
where A and () are constants dependent on the initial conditions.
Example 2.3.1. Using the method of equivalent linearization, determine the amplitude of oscillations of the Rayleigh equation (2.2.34). According to (2.3.5) and (2.3.6), the equivalent damping coefficient and the equivalent frequency are equal to
I
271'
he(a) = 2 1 1raoa = -h
[2h(-aaosin!li) - g(-aaosin!li)3] sin!lid!li
o
3
+ gga 22 0:0'
I
(2.3.17)
271'
O:e(a)
= ao =
2 1
1raoa
[2h( -ao:o sin!li) - g( -ao:o sin !li)3] cos!li d!li
o
0:0·
(2.3.18)
Therefore, the equivalent linear equation takes the form
Ii + ( -h + ~ga2a~ ) 1i + a~y = 0,
(2.3.19)
and the amplitude of the periodic oscillation is equal to ao
=
2- fBh. 0:0 Y39
(2.3.20)
2.4 Analysis of N onconservative N onautonomous
Systems 2.4.1 Introduction
We consider one-degree-of~freedom systems of the form
ii + o:~y
= c¢(y, il, wt),
(2.4.1 )
2.4 Analysis of Nonconservative Nonautonomous Systems
and the exciting force fulfils the periodicity conditions ¢( wt + 21r) Therefore, it can be developed into the Fourier series
27
= ¢( wt).
M
¢(y, iJ, wt) = QlO(Y, iJ) +
L {Qlm(Y, iJ) cos mwt
m=l + Q2m(Y, iJ) sin mwt}.
(2.4.2)
We also assume that the functions QlO, Qlm and Q2m are analytical because of their variables. This means that they can be expanded into a Taylor series in the vicinity of the equilibrium (here taken as the trivial one Y = iJ = 0). In the case of linear systems with a periodic excitation, the resonant oscillations can appear for m~harmonics, when mw = aD, whereas in the case of nonlinear systems, the resonance occurs when
mw = nao,
(2.4.3)
where m, n = 1,2,3, .... In dissipative systems during resonant oscillations, an increase in the amplitude of oscillations is observed. The resonance occurring in nonlinear systems can be classified as follows: 1. 2. 3. 4.
Main resonance (m = n = 1). Subharmonic resonance (m = 1, n > 1). Ultraharmonic resonance (m > 1, n = 1). Ultrasubharmonic resonanCe (m > 1, n > 1).
2.4.2 Nonresonance Oscillations Consider an oscillator governed by the equation jj
+ a~y = c [Q(y, iJ) + P(7])] ,
(2.4.4)
where c is the perturbation parameter and the exciting force P(7]) = P(7] + 2'rr), where 7] = wt. We use the KBM method described earlier to analyse (2.4.4). We are looking for the solution Y = acosW +cYl(a,W,7]),
(2.4.5)
where
a=cA1(a), .p = aD + cB 1 (a),
(2.4.6) (2.4.7)
Here we restrict ourselves to the O(c 2 ) approximation. From (2.4.5) one obtains
Ii = Ii cos lji + aq, sin lji + e (
Z:
Ii + :: q, +
~ w)
,
(2.4.8)
28
2. Discrete Systems
(2.4.9)
(2.4.10)
(2.4.11) Because .. a
= dt
d (.) a
!i}
= i (tit) = c 2 dB I Al
=c
2A dA I I da ' (2.4.12)
dt da' ·2 2 lJt = a o + caaOB I + c2B 1 ,
therefore, the left-hand side L of (2.4.4) takes the form • ! 8 2YI 2 8 2YI L = c ( 2A 1a o sm lJt - 2B 1a oa cos lJt lrI- 8lJt 2 ao + 2 8lJt81] aow 2
2 + 881]2YI W 2) + CaOYI·
( 2.4.13 )
Developing the right-hand side R of (2.4.4) into a power series in c, we have
P = c [Q(y, y)
+ P(1])].
Q } 8Q dY] + c 111 [8ay dy dc + 8y dc IE=o + ... = c [Q(a cos lJt, -aao sin lJt) + P(l1)] + O(c). . (2.4.14) Because the functions - Q(lJt) = Q(lJt + 27r) and P(y) = P(w(t + T)), . = c { [Q(y, y)
+ P(1])]
IE=o
where T = 27r /w, then We develop the right-hand side of (2.4.14) into the Fourier series of the form 1
P = c [ "2bo(a)
CXl
1
+ ~ (bn(a) cosnlJt + en(a) sin lJt) + 2PO
+ f~ (Pm cos m'l + qm sin m'l) ]
(2.4.15)
2.4 Analysis of Nonconservative Nonautonomous Systems
29
where
J ~J 2'11'
bn
=
~
Q(a coslJi, -aCto sin !It) cosn!It d!It,
o
2'11'
en =
Q(acoslJi, -aCto sin !It) sinn!It d!It,
o T
Pm = ;
J J
P(11) cos m11 d11,
o
T
qm=;
P(11)smm11 d11,
(2.4.16)
m,n=0,1,2, ...
o
Equating the terms of the same powers of c in (2.4.13) and (2.4.15), we obtain 2 2 2 2 8 YI w 8 YI w 8 YI 1 { . ,T,2 + 2- !It8 + 2 - 8 2 + YI = 2 [2A 1a o - cI(a)] sm!It 8~ ao 8 11 a o 11 ao 1 00 + [2B 1a oa - bl(a)] cos !It + 2"b o(a) + (bn(a) cosn!It n::;:::2
E
+
1
en(a) sin !It) + 2"Po + ~l (Pm cos m11 + qm sin m11)· 00
}
(2.4.17)
The first two terms standing on the right-hand side of that equation give the secular terms and we get
J J
..
2'11'
Al
= -CI(a) - - = - -1a2Cto
21T" o
Q(a cos !It, -aCto sm!It) sm!It d!It,
(2.4.18)
o
2'11'
B1
=
-bl(a) -2Cto
=- 2
1
a1T"a o
Q(acos!It, -aao sin !It) cos !It d!It.
(2.4.19)
o
From (2.4.17) we have 2 2 2 2 w 8 YI w 8 YI 8 YI 8!It2 + 2 Cto 8!It 8 11 + Ct~ 8112 + YI =
1 [1 2"b o(a) +
a~'
~ (bn(a) cosn!It 00
00
+
en (a) sin !It
1 + ~l (Pmcosm11+qmsinm11)· ] ) + 2"Po
(2.4.20)
From the condition that the solution to (2.4.20) will not contain the first harmonic of the force oscillations, we find
2. Discrete Systems
30
+ 2: (bn(a)2 cosn~ITI + 00
YI
=
1 [bO(a) -2 - a 2
o
+Po + 2
I-n
n=2
f: (
Pm 1 - (:
m=I
m)
2
en(a)2 sm~ . ITI)
(2.4.21)
I-n
cosm71 +
qm 1 - (:
m)
2
Sinm 71) ].
Taking into account (2.4.21) in (2.4.5), we get the general solution obtained with an accuracy of O(c 2 ). According to (2.4.6), (2.4.7) and (2.4.18), (2.4.19), we get
c_1 1 271'
a = __
27rao
Q(a cos rIt, -aao sin rIt) sin rIt drIt,
(2.4.22)
o
271'
rit =
-
Q( a cos rIt, -aao sin rIt) cos rIt drIt.
c
2 a7ra o
(2.4.23)
o The above equations govern the transitional dynamical state of the systems investigated. When we consider the steady state, the problem is reduced to the solution of the nonlinear algebraic equation 271'
/ Q(a cos rIt, -aao sin rIt) sin rIt drIt = 0,
(2.4.24)
o
from which We determine the amplitude"a. It can possess several solutions. The phase corresponding to each of them can be found from rIt =
1
1 2
(ao -
c
71'
2a7rao
Q(a cos rIt, -aao sin rIt) cos rIt drIt) dt+ 8 0 .(2.4.25)
o The solution y includes' two parts. The first one governs the free oscillations (harmonics of rIt), whereas the second one governs the excited oscillations (harmonics of 71). In the general case a separation of those two types of oscillations is impossible. The solution is not defined for the resonance case, Le. when ao = rru.v. This problem will be solved in the next section. Example 2.4.1. Investigate the nonresonant motion of the oscillator jj
+ a~y = c [(2h -
gii) iJ + pcoswt] .
(2.4.26)
According to (2.4.15) and (2.4.16), we get Pm = 0 for m -# 1, PI = P, qm = 0, bn = 0, C2 = 0, C3 = iga3a~ and en = 0 for n > 3. On the basis of (2.4.5), we obtain y = a cos ilt
+0
[;2 ga
3
"'~ sin 3ilt + "'~ ~ w2 cos wt] .
(2.4.27)
2.4 Analysis of Nonconservative Nonautonomous Systems
31
For this case we have from (2.4.22) and (2.4.23)
J 2111"
iL
= __c_
21T" Ct o
[2h - g( -ao:o sin !Ii 2 )] (-aCto sin !Ii) sin !Ii d!Ii
o
2 = ah(1 - K 2 a ),
(2.4.28) (2.4.29)
if, = Cto,
where
K 2 = ~gCt~ 8 h . For K a
=
2
(2.4.30)
> 0 the solutions to (2.4.22) and (2.4.23) are as follows aoe
Eht
VI + a~k2e2Eht ,!Ii = O:ot + 8,
(2.4.31)
where ao and 8 are defined on the basis of the initial conditions. Taking into account (2.4.31) in (2.4.27), We obtain Y=
c [ aoeEht ] cos(Ctot + 8) + -gCto +a~k2e2Eht 32 viI +a~K2e2Eht aoeEht
viI
. sin(3Ctot + 38) +
2 cp
0:0
+w
2
coswt.
3
(2.4.32)
In the steady state t - +00 and from (2.4.32) we obtain lim Y =
t-++oo
..!. cos(0:0t+8)+-=- gCt~ sin(30: 0t+38)+ 2 cp 2 coswt.(2.4.33) K 32 K Ct +w
o Finally, we have to remind the reader that the solution is valid for small amplitude of the excited force and when it is far enough from resonance.
2.4.3 Oscillations in the Neighbourhood of Resonance Let us consider again the oscillator (2.4.4) and now we use the equivalent linearization method described earlier to solve the problem stated in the title of the subchapter. The solution sought is of the form K
Y = a cos !Ii +
L ckYk(a, !Ii,7}),
(2.4.34)
k=l
where K
iL =
L c Ak (a, t9). k
(2.4.35)
k=l K
if,
= 0:0 +
L c Bk(a, t9), k
k=l
depend additionally on the phase shift t9. This phase is defined by
(2.4.36)
2. Discrete Systems
32
nt9(t)
nlJi(t) - mTJ(t) ,
=
(2.4.37)
which allows us to eliminate nlJi and obtain the following equations K
+ t9) +
Y = a cos ( : TJ
L ekYk(a, t9, TJ),
(2.4.38)
k=l K
a = L ek Ak(a, t9),
(2.4.39)
k=l K
m t9. = ao - -w n
+
""" e kBk(a, L...J t9).
(2.4.40)
k=l
Further considerations will be focused on the oscillations near the ultrasubharmonic resonance, which emphasizes the equation ao2 -
(m)2 ~w = eLl.
(2.4.41)
Taking into account (2.4.41) in (2.4.4), we obtain
y + (:
w)
2
Y = e [Q(y, iJ)
+ p(TJ) - Lly].
, (2.4.42)
From (2.4.37) we get
V(mw) n
ao = . /
2
+ eLl
'V
m w+ ~ e. n 2m
(2.4.43)
n
From (2.4.34)-(2.4.36) We obtain (with·an accuracy of O(c2)) y = a cos ( : TJ
a=
CAl (a,
+ t9) + eYI (a, t9, TJ),
(2.4.44)
t9),
t9. = e [B I (a, t9)
(2.4.45) Ll ] = eB I (a, t9). + 2';:w,
(2.4.46)
Differentiating (2.4.44) and taking into account (2.4.45) and (2.4.46), We get
if = -a :
w sin ( :
7) + ,,) + 0 [ Ai cos ( : 7) + ")
+
+19) + ' : Ii = -a (:w)' cos (:7) +19) + o[- 2a:wB wA sin ( : 7) +") +~:; w . -aB i sin ( : 7)
-2:
i
2
] •
(2.4.47) i
cos
(:7) +19) (2.4.48)
2.4 Analysis of Nonconservative Nonautonomous Systems
33
Taking into account the above equations in (2.4.4), we obtain - a ( : 71) 2 cos (:71 + 19)
+a
( : 71) 2 cos ( : 71
+ 19)
+c [-2a:wB1cos(:71+19) -2:wA1sin(:71+19)
+
Z l
2 W
+ (:'7)" Yl]
= c { Q [a cos ( : 71
+ P(71) -
+ 19) , -a :
W
sin ( : 71
+ 19) ]
~acos (:71 + 19)}.
(2.4.49)
Comparing the terms in c, we get
8
2
YI
871 2
+ (m) 2 YI = 2 m .!.A1 sin (m 71 + 19) nnw
(2.4.50)
n
cos (m 71 + (2a m .!.B 1 ~a) n w w2 n
+ 19) + ~2 [~bo(a) w
2
00
+ Lbnl(a) cosn' (:71
+19) + Cn,(a)sinn'
(:71
+ 19)
n'
1
+ '2Po
I .
I
+ LPm l cosm 71 + qm' smm 71 00
]
,
m'
where 2'11'
~/
bn =
Q (a cos W, -a: w sin
w) cosn'w dW,
Q (a cos W, -a : w sin
w) sin n'w dW,
o 2'11'
Cn
=
~/ o
T
Pm'
2/ P(71)cosm
=T
I
71 d 71,
o T
qm'
=; /
P(71) sin m'71 d71,
m ' , n ' = 0,1,2, ...
(2.4.51)
o
Further calculations will be made for n 2
8 YI2 -8 71
= 1.
From (2.4.50) we obtain
1 I + 2C1 1 ) sin (m71 + 19) + 29m 1 + m 2 YI = (2m-A sin m71 w w w
+ (2am'!'B I ~~ w w
+w ~bl)
cos(m71
+19) + ~PmCosm71 w
34
2. Discrete Systems
+
~2 [ ~bo(a) + ~ b
+
~po + m~:
n,
(Pm'
1
cosn' (m7) + 01) + Cn' sin n' (m7) + 01)
COS
m'7) + qm' sin m' 7))] .
(2.4.52)
m'#m
After introd ueing the following quantities 1 1 2m-A I + 2C1 w
w
1 2qm
= a,
w
1 Lla 1 2am- BI - - 2 + 2bl w
w
w
= 13,
1 2Pm = 8,
= /,
(2.4.53)
w
we get
+ t9) + 13 sin m1J + / cos (m1] + t9) + 8 cosm1J = (0: cost9 + 13 - / sin t9) sin m1J + (0 sin t9 + / cos t9 + 8) cos m1J. (2.4.54)
osin (m1]
The secular terms equal zero when 0: cos t9
+ 13 sin t9 = 0,
0: sin
+ / cos t9 =
t9
(2.4.55)
O.
Multiplying the first equation of (2.4.55) by cos t9 and the second one by sin t9, and adding both of them, we have 0+
13 cos t9 + 8 sin t9 =
2m.!.A 1 + ~2 CI w
w',
+ w~2 qm cos t9
1 . +2 Pm sm t9 = O.
(2.4.56)
w
Multiplying the first equation of (2.4.55) by sin t9 and the second one by cos t9, and adding both of them, we have
13 sin t9 - 8 cost9 - / = ~qm sin t9 - ~Pm cost9 w
w
1 -2am-B 1 W
Lla w2
+-
b w
- - l2 = O.
(2.4.57)
From the above equations we obtain qm . A 1 = - -CI- - - cos·u_0 - -Pm- smt9, 2mw 2mw 2mw Ll bl B I = --- - qm.smt9 - Pm cost9. 2mw 2maw 2maw 2maw Taking into account (2.4.58) in (2.4.45) and (2.4.46) we get . (------cost9---smt9 CI qm Pm . ) a=c 2mw 2mw 2mw ' -C.l ·u
= 00
-
mw
+c (
-
bI - qm. sm u_0 2maw 2maw
-
(2.4.58)
(2.4.59) Pm 2maw
COS t9
) .
(2.4.60)
2.4 Analysis of Nonconservative Nonautonomous Systems
35
In order to simplify this procedure, we take m = 1, Le. we are looking for a solution of the form
+ '19).
y = a cos(wt
(2.4.61 )
From (2.4.43) we get
cLl = 2w(ao - w),
(2.4.62)
whereas from (2.4.59) and (2.4.60) we have
.
(Cl
qm. _Q) a=c ----sm-u 2w 2w 211"
IQ( a cos tJi, -aw sin tJi) sin tJi dtJi - cp sin '19,
= - _c_ /
21T"W
.
'19 = Qo -
2w
o
(b1 - - - -qm cos '19 ) 2aw 2aw
+c
W
_I
(2.4.63)
= ao - w
271'
- _c 21T"wa
Q (a cos tJi, -aw sin tJi) cos tJi dtJi -
o
cp cos '19 . 2aw
(2.4.64)
Now we introduce the following quantities 271'
he (a)
=-
c 21T" Q
1 Q(a cos tJi, -aao sin tJi) sin tJi dtJi,
oa
(2.4.65)
o
1 271'
Qe(a)
= Qo
c
-
21T" Q
oa
Q(a cos tJi, -aQo sin tJi) cos tJi dtJi.
(2.4.66)
o
We now show that (2.4.61) fulfils the equivalent linear equations of the form jj + 2h e (a)iJ
+ a;(a)y =
cpcoswt.
(2.4.67)
The equation (2.4.63) can be transformed into the form 271'
ci = -ahe
+ ahe -
_c_1 Q (a cos tJi, -aw sin tJi) sin tJi dtJi - cp sin '19 21T"w 2w
o
271'
=
-ahe
+ - c2 1 Q 1T"
!
o
Q (a cos tJi, -aao sin tJi) sin tJi dtJi
o
271'
- -c2 1T"W
o
Q(a cos tJi, -aao sin tJi) sin tJi dtJi - cp sin '19, 2w
(2.4.68)
2. Discrete Systems
36
and then taking into account (2.4.62), into the form
f
211'
a = -ah e + c.d 21r (w + L-) 2w
+ cLl) sin rIt)
Q (a cos rIt, -a (w
2w
0
sin rIt drIt
211'
__c_ / Q(a cos rIt, -aw sin rIt) sin rIt drIt - cp sin '19. ~
21rW
(2.4.69)
o After expanding the second term of the right-hand side of (2.4.69) in a power series because of c, we obtain
f
211'
a = -ahe +
21r (w
f
c.d) + L2w
Q (a cos rIt, -a (w
+ cLl) sin rIt) 2w
0
sin rIt drIt
211'
__c_ ~w
= -a h e
cP
Q(a cos rIt, -aw sin rIt) sin rIt drIt - 2 sin '19
w
o -
cP'_ Q SIn·v.
(2.4.70)
-
2w Similar considerations lead to . cp '19 = Ct e - W - cos '19.
(2.4.71)
2aw From (2.4.61) we get
acos(wt + '19) - a(w + t9) sin(wt ~,t9): ij = a cos(wt + '19) - 2a(w + t9) sin(wt + '19) - aJ sin(wt + '19) -a(w + t9)2 cos(wt + '19).
iJ
=
(2.4.72) (2.4.73)
Taking into account (2.4.70) and (2.4.71) in the above equation, we find (with an accuracy of O(c 2 ))
iJ
+ '19) + cp sinwt - aO:e sin(wt + '19), 2w 2ah eO: e sin(wt + '19) + cpcoswt - aO: e cos(wt + '19).
= -ah e cos(wt
ij =
(2.4.74) (2.4.75)
The left-hand side L of (2.4.67), taking into account (2.4.61), (2.4.72) and (2.4.73), can be transformed into the form
L
= cpcoswt - 2ah~ cos(wt + '19) + he cp sinwt w
"J
cpcoswt.
(2.4.76)
Taking into account (2.4.76) and the right-hand side of (2.4.67), we see that the solution (2.4.61) fulfills (2.4.67) with an accuracy of c. Thus, the method of equivalent linearization allows us to replace (2.4.4) by (2.4.67), which is valid near resonanCe. The unit equivalent coefficient of damping he(a) and the equivalent frequency O:e(a) are functions of the amplitude a. This amplitude can be found from the formula
2.4 Analysis of Nonconservative Nonautonomous Systems
a=
cp
J(o:~(a) -
w2)2 + 4h~(a)w2
37
(2.4.77)
,
which allows us to obtain
4h;(a) [h~(a) - Ct~(a)]
w=
+c
2 2
; .
(2.4.78)
0:
For a given amplitude, we can have one, two or no value of frequency according to (2.4.78). According to linear oscillation theory, we have {} = arctan
-2he (a)w 2() 2· Ct e a - w
(2.4.79)
Therefore, for each amplitude a and w defined by (2.4.78), it is possible to find the corresponding phase {}. The exemplary results are shown in Figure 2.2. However, not all parts of the resonanCe curveS are stable. In order to check stability, let us consider the steady state defined by y = ao cos(wt
+ {}o),
(2.4.80)
where ao and {}o fulfil equations (2.4.70) and (2.4.71). Therefore, we have
-aohe(ao) - cp sin {}o 2w Cte(ao) -
= 0,
(2.4.81)
~ cos{}o = O.
W -
2aow
(2.4.82)
In order to investigate the stability of (2.4.80), we have to consider the near by solution y = a cos(wt whert~
it
+ {}),
(2.4.83)
a(t) and {}(t) are the solutions of (2.4.70) and (2.4.71) =
-ahe(a) - ;: sin {}
.
{} = Cte(a) -
cp W -
.
-
2lU.V
=
cos{}
cA [a(t), {}(t), w] ,
= cB [a(t), {}(t),w].
(2.4.84) (2.4.85)
We will consider the solutions close to the investigated solutions
a(t) {}(t)
= ao
+ oa(t),
= {}o
+ 019 (t),
(2.4.86)
where o(t) are small enough. Taking into account (2.4.85) in (2.4.82), we obtain
6a = cA [(aD + Oa(t)) , ({}o + 019 (t)) , w] , 619 = cB [(aD
+ Oa(t)), ({}o + 019 (t)) ,w],
(2.4.87)
and next, we develop the right-hand sides of (2.4.87) into a Taylor series because of oa and 019 near the point (ao , {}o), and finally we obtain
38
2. Discrete Systems
a)
i a
o
2
b)
1
2
Fig. 2.2. Amplitude of oscillations (a) and phase shift (b) versus w / 010
. = c [ A(ao,t9o,w) oa
+ aA aa (ao,t9 0)oa + aA aa (ao,t9o)o~]
.
+ aB aa (ao, t90)oa + aB aa (ao, t9o)o~ ]
[
o~ = c B(ao, '190, w)
, .
(2.4.88)
According to (2.4.80) and (2.4.82), we have
A(ao,t9 o,w) = 0, B(ao '190, w) = o. l
(2.4.89)
Solutions to the linear differential equations (2.4.88) are sought in the form
Oa = Daert , o~ = D~ert.
(2.4.90)
2.4 Analysis of Nonconservative Nonautonomous Systems
39
Taking into account (2.4.90) in (2.4.88), we obtain the following characteristic equations
r
2
- IT
+e
[~: (aD. 190) + ~~ (aD, 190)] 2
[8A 8B 8a (ao, '19 0 ) 8'19 (ao, '19 0 )
-
8A 8B ] 8'19 (ao, '19 0 ) 8a (ao, '19 0 )
=
O. (2.4.91)
The solution will be stable, if oa(t) and o~(t) approach zero with t ---. +00. This happens when the real parts of the roots of (2.4.91) are less than Zero. According to Vieta's formulas we have
8A 8A 8a + 8'19 < 0, 8A8B 8A8B 8a 8'19 - 8'19 8a > o.
(2.4.92) (2.4.93)
These conditions will be transformed into a form allowing us to estimate the stability of the solution on the basis of the reSonanCe curve given in Fig. 2.3.
a
Fig. 2.3. Resonance curve with stable ( continuous line) and unstable (dashed line) parts
According to (2.4.84) and (2.4.85), we obtain
8A 8a 8B -8 a
8~
-he(ao) - ao 8a (ao), 8h e ep = -8 (ao) + - 2 cos '19 0 , a ~
=
8A ~ 8'19 = - 2w cos '19 0 , 8B ep. 8.0 = --2- smt9 o· v
~w
(2.4.94)
40
2. Discrete Systems
On the basis of the above results, the stability conditions will take the form 8h e ( ) cp . (2.4.95) - he(ao) - aO-8 ao + -2- smt9 0 < O. a aow Taking into account equation (2.4.81), we have ) 8h e - he(ao) - ao 8a (ao < O.
(2.4.96)
This condition is transformed into the form d [a~he(ao)] > 0, for ao > O. d ao According to (2.4.82), we obtain
(2.4.97)
A [ao(w), t9o(w),w] = 0, B [ao(w), t9o(w),w] = O.
(2.4.98)
Differentiating the above equations with respect to w, we have
8A 8ao 8a 8w 8B 8ao 8a 8w
8A 8'19 0 + 8'19 8w = 8B 8'190 + 8'19 8w = -
8A 8w'
8B (2.4.99)
8w'
Multiplying the first equation of (2.4.99) by 8Bj8t9, and the second one by 8Aj8t9 and adding up both of them, we get
8A dB _ 8A dB = (dao) 8a dt9 8'19 da dw
-1
(8A dk _ 8B dA) 8'19 dw 8'19 dw .
(2.4.100)
Because
8A
cp. 8w = 2w2 smt9 o, 8B cp, = -1 + cos '19 0
8w
(2.4.101 )
2aow2
and taking into account (2.4.94), we obtain according to (2.4.93), -
(~) [;~ cosOo ( -1 + 2:: -1
2 c p2 + 4 3 sin 2 '19 0]
aow
> 0,
2
cos 0 0 ) (2.4.i02)
which, after limiting considerations to the terms in the first power of c, leads to the condition ~ cos '19 0 ~ >0. dw
(2.4.103)
2.4 Analysis of Nonconservative Nonautonomous Systems
According to (2.4.85), we have ep - cos '19 0 = ao [ao(ao) - wI , 2w
41
(2.4.104)
then, for ao > 0 we have
D:e(ao) d
~
W
0
(2.4.105)
>.
dw
On the basis of this inequality we can formulate the following conclusions: the solution is stable if
-dao > 0 an d a e (ao) > w, dw
(2.4.106)
or if
-dao < 0 an d D:e (ao) < w.
(2.4.107) dw This analysis allowed us to determine the stability of the solutions on the basis of the consideration of the resonance curve, which is illustrated in Fig. 2.3. In this figure the "skeleton line" is defined by the equation D:e(ao) = w.
Now we will analyse the slow transition through the resonance taking into consideration Fig. 2.4. The amplitude of driven oscillations is increased along sector AB of the resonance cUrve (Fig. 2.4a). In point B a sudden jump into a new branch on the resonance curve has occurred (point D) and a further increase in the frequency w is accompanied by an increase in the amplitude of oscillations up to point E. In this point a sudden amplitude change to the value defined by point J has appeared. A further increase in w causes a slight decrease in the amplitude of oscillations. In a similar way we are able to analyse the dynamics with an increase in the frequency w (Fig. 2.4b). We have to emphasize that the process of nonlinear and discontinuous changes of the amplitude corresponding to the increase in the frequency differs from a similar process accompanying the decrease in the frequency. As has been mentioned earlier, for the considered parameters of the system different kinds of oscillations can occUr (they depend on the initial conditions). Example 2.4.2. Analyse the dynamics of the system
ii + a~y
= e(-2h - (3y3
+ pcoswt)
(2.4.108)
in the neighbourhood of the resonanCe Using the method of equivalent linearization.
42
2. Discrete Systems
According to (2.4.65) and (2.4.66), we obtain
J 271'
h (a) = e
e
27raoa
[-,8(a cos !li)3 - 2h(-aaosin!li)] sin!lid!li
o (2.4.109)
= eh,
J 271'
a (a) = ao e
e
27raoa
3e
[-,8(acos!li)3 - 2h(-aao sin !li)] cos!lid!li
o
2
(2.4.110)
= ao + - a ,8.
3ao According to (2.4.78), the resonance curve is given by W
=
[0<0 + ~a 3e
(
±
2
2
2 2
f3] - 2e h
4e'h' [e'h'- (0<0 + 3~0 a'(3)'] + e?') !,
(2.4.111)
and following (2.4.79), we find the shift of the phase equal to {} =
arctg
-2ehw 2'
(2.4.112)
a 2,8) - w 2 ( a o + ;l..L 8 ao
On the basis of the last two equatioI\s and for the parameters ao = 1, e = 0.1, P = 2, ,8 = 2, resonance curves are obtained, which are shown in Fig. 2.5. The skeleton line is defined by
ao(a)
= ao + -3ea 2,8 = w. 8ao
(2.4.113)
In this figure the continuous line denotes the stable solutions, whereas the dashed line corresponds to the unstable solutions. The conditions (2.4.97) have the form 2eh < 0, which is true for h > o.
2.5 Nonstationary Nonlinear Systems In what follows, we consider nonstationary linear and nonlinear mechanical systems. We begin with a simple example of the parametric oscillations of the mathematical pendulum with a periodic change of its length (Fig. 2.6). For small cp the governing equation has the form d 2
2 dL dcp
9
+ L dtdt + L
0,
(2.5.1)
where L(t) is the time-dependent length and 9 is the gravitational acceleration. We consider small changes of the length according to the formula
2.5 Nonstationary Nonlinear Systems
a)
43
E
a
.
C.. F\ B
H
..
'
.'
w
b)
a
c
w
= L o(1 + /-Lcoswt).
L
Fig. 2.4. Nonlinear jumps of amplitudes accompanying an increase (a) and a decrease (b) in frequency
(2.5.2)
Then, by introducing the new variable x = LIp and after some transformations, we obtain 2
d x dt 2
2 ) + (g L + /-Lw coswt
o
(2.5.3)
x = 0,
where 0 < /-L « 1 is a small perturbation parameter. When the fixed point zero (trivial solution) undergoes periodic movement (Fig. 2.6b), we get the following governing equation mLoep
= -mg sin
(2.5.4)
which leads to the Mathieu equation
X + (a
+ 16qcos27") X = 0,
(2.5.5)
where 27" = wt,
x
=
Yo q = 4£0'
(2.5.6)
2. Discrete Systems
44
3~----r-----.,......----.-----.,
a) a
2
h=O.5
n=l 1 - .-- ---- -------r-~______t~~;t__::71--1
0.5
1.5
1
2
w
b)
-1r/2
o ~h.....;;O"""'''''''
o
~_--t---l-_ _-----J 1 • 1.5 w 2
0.5
Fig. 2.5. Amplitude (a) and phase (b) versus w for different values of h
In the third example (Fig. 2.6c) the excitation y(t) can be defined as follows r2
y(t) = 4L
+ r coswt -
~2
4L cos 2wt,
(2.5.7)
and the problem is reduced to the Hill equation
x + [0" + J.L4'(r)] x
=
0,
(2.5.8)
where.
0" =
~,
2r = wt,
J.L =
~o'
4'(r) = w 2 cos 2r -
TW
2
cos 4r.
(2.5.9)
If 4'( r) = cos r, then we have the following equation d2 x dr 2
+ (0" +J.Lcosr)x =0.
(2.5.10)
The periodic solutions with periods 21r and 41r appear on the stability limits of the system governed by the Mathieu equation (2.5.10). We would like to
2.5 Nonstationary Nonlinear Systems
45
y(t)
I I I I
_.- .c)
b)
a)
Fig. 2.6. Mathematical pendulum with a periodically changed length L(t) (a), with the periodic movement of the fixed point 0 (b), and with anharmonic periodic excitation (c) estimate analytically the (a, ,) values for which the system possesses a 41rperiodic solution. We are looking for the following solution
x(r,l-t) = xo(r) + I-txI(r) + 1-t2x2(r) a(l-t) = ao + I-tal + 1-t2a2 + ....
+ ... , (2.5.11)
Substituting (2.5.11) into (2.5.10) and comparing the terms with the same power of the small parameter I-t «: 1, we get
d 2xo dr 2 d2xl dr 2 d2x2 dr 2
+ aoxo = + aOXl
0,
= -(al
+ aOX2 =
-(al
+ cos r)xo, + cos r)xl
(2.5.12)
- a2XO,
The first equation of (2.5.12) gives
xo(r)
=
A o cos Jl10r
+ B o sin Jl1Or.
(2.5.13)
Because we are looking for the 41r- periodic solution, therefore
k2
ao ="""4'
k=0,1,2, ...
(2.5.14)
2. Discrete Systems
46
and from (2.5.13) we obtain
kr
kr
xo(r) = A o cos 2
+ Bosin 2·
(2.5.15)
o. From (2.5.12) we find
We consider first the case when k =
Xl (r) = C l
+ C 2r
r2 - O"lA0 "2
+ Ao cosr,
(2.5.16)
where Ci (i = 1,2) are the integration constants. Taking into account the periodicity, we have xI(r) = C l
+ A o cos r.
(2.5.17)
From the second equation of (2.5.12) we get X2(T)
~ Cs + C4T - (<72+~) An ~2 + C1 COST + ~ COS2T.
Again, the 41T"-periodicity gives equation of (2.5.11), we obtain 0"
1-L
0"0
=
0"1
= 0,
0"2
= -~ and from the second
2
(2.5.19)
';!::j - - .
2
For n
(2.5.18)
= 1 we have 0"0 = ~,
xo(r) = A o cos ~
2
and
+ B o sin~.
(2.5.20)
2
The second equation of (2.5.12) has the Ikw form
::~ + ~X1 = - (<7
1
+
DAocos~
+ (-<7 1 +
D
Bosin
~
Ao 3 B . 3 - - cos -r - - o sm -r. (2.5.21) 2 2 2 2 In order to have a periodic solution, secular terms should vanish, which leads to the conditions
- (<71+ DAn =0, ( <71 + ~ )
B o = O.
(2.5.22)
There are three different solutions of (2.5.22), which are given below': 1. A o
# 0;
2. A o = 0; 3. A o = Bo
O"i l ) = -~; O"i
2
)
= +~;
Bo
= O.
B o # O.
= 0 with O"i 3 ) temporarily unknown.
(2.5.23)
2.5 Nonstationary Nonlinear Systems
47
The first two solutions of (2.5.23) give the following estimation of the stability limits u(1)
u(2)
~
! _ tt
~
! + tt.
4
2'
(2.5.24)
4 2 For n = 2 we have the following stability limits u(1)
~ 1 + ~H2
u(2)
~ 1 - -tt2 .
12"'" ,
1
(2.5.25)
2
In all these cases We introduced the small positive perturbation parameter tt, which characterizes the modulation depth of the parametric excitation. However, linear systems are an idealization of real systems which are nonlinear. Therefore, we consider the general form of the system of equations
{±}
= [[~o]
+ tt [~I(t)] + tt 2 [~2(t)] + ... J {x} +e [F(XI' ... ,xn )] .(2.5.26)
Above, we have two independent perturbation parameters tt and e. The first one defines the parametric modulation, whereas the second one characterizes the nonlinearity. Additionally, [~o] is the constant matrix, and [~i(t)] = [~i(t + T)], where T is the period. The solution of (2.5.26) is sought in the form {x(t, e, tt)}
= x(O,O) + ttX{I,O) + tt 2x{2,O) + .. , +e (X{O,I)
+ ttX{I,I) + tt 2x {2,1) +
+e 2 (x{O,2)
)
+ tt X{I,2) + tt2x {2,2) +
(2.5.27)
)
+ ... Substituting (2.5.27) into (2.5.26), and after comparing the expressions with the same power of ttkel, we get { ±(O,O) } -
[~o] { x{O,O)} = {O},
01
{ X(O,I) } _
[~ol { X(O,I) } = [~~]
ttl
{ ±(I,O) } -
[~o]
{ x(O,2) } -
[~ol { x(O,2)} = [~~]
tt 2
{ ±(2,O) } -
[~o] { x{2,O)}
OIL
{ X(l,l)}
[~oJ {X(l,l)} = [~ll { X(O,I) } + [~]
eO, ttO
0
2
-
{x{1,O)} =
=
{x(O,O)} ,
[~l] { x{O,O) }
[~2]
{
{X(O,I)}
x{O,O)}
(2.5.28)
,
+ [~O~]
{x(O,O)} ,
+ [~l] { X{I,O) }
,
{X(I,O)} ,
48
2. Discrete Systems
The first equation of the set (2.5.28) gives the fundamental part of the sought solution (2.5.27), which can be succesfully approximated by solving the recurrent set of the rest of equations (2.5.28). To achieve a complete ordering of all of the recurrent equations, we take the following additional condition e < 1-£. The tringle below gives the ordering from the smallest to the largest values asymptotically on each horizontal row, Le. defines the sequence of recurrent equations e
1-£ e el-£ 1-£2 e3 e21-£ 1-£2 e 1-£3 e4 e31-£ e21-£2 1-£3e 1-£4 ............................................................. 2
(2.5.29)
The question arises: which initial conditions should be taken for x(O,O) , X(l,O), X(O,l) and so on. The following condition should be fulfilled {x(to, e, I-£)} = {x(O,O) (to) } + e { X(O,l) (to) } + 1-£ { x(l,O) (to) } + ... ,(2.5.30)
and usually it is useful to take {x(O,O)(to)} = {x(to,e,I-£)} ,
(2.5.31)
which leads to the equations X(I,O)(tO)
= X(O,l)(tO) =
X(2,O) (to)
Examples of two calculation to considered now [28].
= ... = O.
supp~rt
(2.5.32)
the above consideration will be
Example 2.5.1. Consider the standard example of vibrations of a system with one degree of freedom, of the form
d2 x mcit
+ (ko + k l
cos 2t),x + k 2x 3 = 0,
(2.5.33)
where m is the mass of the vibrating body, and ko, k 1 and k2 are the stiffness coefficients. After assuming the parameters >..2 = ko/m, 1-£ = k1/ko, and e = k 2/m, we obtain the equation . d2 x
dt
+ >..2(1 + 1-£ cos 2t)x + ex3 = O.
(2.5.34)
For 1-£ = 0, we obtain the Duffing equation, and for e = 0, the Mathieu equation. Let us develop the quantities >.. 2 and x into a power series in the small parameters 1-£ and e:
2.5 Nonstationary Nonlinear Systems ..x. 2 = n 2
+ a(I,O) J-L +
+e 2 (a(0,2) x
49
J-L 2a(2,0) + ... + e ( a(O,I) + J-La(I,I) + ... )
+ J-La(I,2) + J-L 2a(2,2) + ... ) + ., .
(2.5.35)
= x(O,O) + J-Lx(1,O) + J-L 2 x(2,0) + ... +e (X(O,I) + J-LX(I,I) + J-L 2 x(2,1) +
+e 2 (x(0,2)
+ J-LX(I,2) + J-L 2 x(2,2) +
)
(2.5.36)
) + . . ..
Now, consider the first unstable zone (n = 1). After substituting (2.5.35) and (2.5.36) into (2.5.34), and equating the expressions representing the same power J-L and e and their combinations, we obtain the following recurrent system of linear differential equations
+ x(O,O) £(1,0) + x(1,O) £(2,0) + x(2,0) £(0,0)
= 0,
=
_x(O,O)
cos 2t
- a(I,O) x(O,O) ,
= _a(2,O) x(O,O) _ a(I,O) X(I,O) _ a(l,O) cos 2tx(0,0) - x(l,O) cos 2t,
£(0,1)
+ X(O,I)
£(1,1)
+ X(I,I) = -3 ( x(O,O») 2 x(1,O)
= (x(O,O») 3 _ a(O,I) x(O,O),
-a (1,1) x(O,O) £(2,1)
+ x(2,1) = _3x(0,0)
-
(2.5.37)
_ a(l,O) X(O,I) _ a(O,l) x(O,O)
X(O,I)
(X(I,O») 2 _
cos 2t
cos 2t, 3 ( x(O,O») 2 x(2,0)
_ a(2,1) x(O,O)
_a(1,I)x(I,O) _ a(0,l)x(2,0) _ a(O,I)x(1,O) cos2t _ a(2,0)x(0,1) _x(O,1)a(I,O)
cos 2t
- x(l,I)a(I,O) -
X(I,I)
cos 2t ,
Let us assume the initial conditions x(t = 0) = A o, x(t = 0) = B o. From the first equation of system (2.5.37) we obtain x(O,O) =
A o cost
+ B o sin t.
(2.5.38)
Substituting (2.5.38) into the second equation of system (2.5.37), from the condition of avoiding the secular terms, we get: lOa (1,0) = and B o = 0 0 and A o = O. Let us consider the first case: then X(I,O) _ or 2 a(I,O) = 1~ A o cos 3t and, taking into account the third equation, we have
-!
!
a (2,0)
=
2.
32'
x(2,0)
= - ~ A o cos 3t + A o cos 5t. 256 768
(2.5.39)
50
2. Discrete Systems
Acting analogously, we obtain
_~A02
...!:.-A30cos3t 32 ' 5 llA3 A3 a(l.l) = _A 2 X(l,l) = 0 cos3t + _ 0 cos5t 16 0' 256 384' 75 73 235 a(2,1) = --A~ x(2,1) = A~ cos3t A5 cos5t 1024 ' 24576 73728 16A4 + 1474~6 cas7t,
a (O,l) =
4'
X(O,l) =
(2.5.40)
(2.5.41)
3 3 3 a(O,2) = --A~, x(O,2) = --4Ag cos3t + 3072Ag cos5t. 128 102 Taking the above calculation into account in (2.5.35), we obtain 2 1 7 2 2 (3 A = 1 - '2J.L + 32J.L + ... + cAo -'4 +0
2
(
+
5 75 2 ) 16J.L + 1024J.L + ...
-1~8A~ + .. -) +....
(2.5.42)
For c = 0 we obtain one branch of the first limit of the stability loss of the Mathieu equation, and for J.L = 0, the dependence of the frequency on the amplitude for a conservative system with the Duffing characteristics.
Example 2.5.2. Consider an unsteady nonlinear system with two-degree-offreedom, the motion of which is governed by the following differential equat~ns
1
•
MZ 1 + C(ZI - Z2) + k(Zl - Z2) + k1(Zl - Z2)3 = 0, (2.5.43) mZ2 - C(ZI - Z2) - k(Zl - Z2) - k 1(Zl - Z2)3 = -(k3 - k o cos 2wt)Z2, where m and M are the masses, k, k 1 , k 3 and ko are the rigidities, and C is the damping coefficient. The frequencies of free vibrations of the conservative system are calculated after assuming C = k 1 = ko = 0 in (2.5.43). They amount to a21 2 = ! , 2
3 ]. [~ + k +m k3:r (~ + k +m k 3 ) 2 _ 4kk M M Mm
'
T
(2.5.44)
Our consideration is limited to calculating the first simple parametric resOnanCe around the frequency al. Equation (2.5.43) will be rearranged to the form
ih + A~Yl
-CMA~(cIY2 - c2yI)3 - J.LOIAl(cl'iJ2 - c2iJI) +J.LPIA~(YI - Y2) cos 2r, ih + A~v2Y2 = -CMA~v2(cIY2 - c2yI)3 - J.LOIAIV(cIY2 - c2iJI) +J.LP2A~v2(Yl - Y2) cos2r, =
(2.5.45)
2.5 Nonstationary Nonlinear Systems
51
where
1
1/J =
, 11 and the dot now denotes the differentiation with respect to r. We seek the solution of (2.5.45) of the form 2 YI = y1°'0) + /lYP'O) + /l2 y 1 ,0) + ... + cyiO,I)
-'2
2 (0,2) (1,1) +c YI + ... + /lcYI + ... (0,0) +
Y2 = Y2
+c
(1,0) +
/lY2
2 (0,2)
Y2
+ ... +
/l
2 (2,0) +
Y2
(1,1)
/lcY2
...
+
(0,1) cY2
(2.5.47)
+ ...
A~ = 1 + /laiI,O) + /l2 a 12 ,0) + ... + ca1o,I) + c2a1o,2) + ... + /lca1 I ,I) + ... Al
= 1
(1,0) (0,1) (1,1) (2,0) (0,2) a a a a1 4 + _I_/l + _ I _ c + _I_/l c + /l2_1_ + c _ _ +
a
.
2 2 2 4 4" After substituting (2.5.47) into (2.5.45), we obtain the following system of linear differential equations .. (0,0) + (0,0) YI YI
0
=,
.. (1,0) + (1,0) _ (1,0) (0,0) £ YI - UI YI YI - -a l
(
• (0,0) . (0,0») cIY2 - c2YI
+PI (y1°'0) - y~o,O») cos2r, .. (2,0) + (2,0) _ (1,0) (1,0) (2,0) (0,0) YI - al YI YI YI - -a l
-01 (CIy~I,O)
(1,0) al ( . (0,0) . (0,0») 1 -cIY2 - c2YI
0
2
I - c 2y 1 ,0») + PI aiI,O)
(y1°'0) - y~o,O») cos 2r
+PI (y1 I ,0) - y~I,O») cos 2r, .. (0,1) + (0,1) _ (0,1) (0,0) YI YI - -al YI -
M
..(0,2) Y1
a(0,2)y(0,0)
+
y(0,2) __ a(O,I)y(O,I) 1
-
1
1
_
1
(0,0) cIY2 -
1
-M { 3c~ (y~O,O) )2y~0,I) -
-
(0,0»)3 c2YI ,
(2.5.48)
Ma(O,I) (<=, y(O,O) 1
q
6c~c2y~0,I)y10'0)
2
(0,0»)3
- c2YI
52
2. Discrete Systems
.. (1,1) +
Y1
(1,1) _
Y1
- -
a(l,I)y(O,O) a(O,l)y(l,O) 1 1 1 1 -
M a (1,0) 1
a(l,O)y(O,l) 1
1
M{3 c13(y(O,O»)2 2 Y2(1,0)
(0,0) (0,0»)3 c1Y2 - c2Y1 -
2 (0,0) (1,0) (0,0) 2 (0,0) (1,0) (0,0) + 6 - 6 c1Y2 Y2 c2Y1 c1 c 2Y2 Y1 Y1 (0,1) ~ al (0,0) (0,0») ~ (.(0,1) .(0,1») - c1Y2 -VI- c2Y1 - vI c1Y2 - c2Y1
2
+P1 (y~O,l) _ y~O,l») cos2r
+ P1a~0,1)
(y~o,O)
_ y~o.o») COS 2T }, .. (0,0) +
Y2
.. (1,0) +
Y2
V
0 -,
2 (0,0) -
Y2
2 (1,0)
v Y2
+ P2 v2 .. (2,0) +
Y2
V
2 (2,0)
Y2
~
2 (1,0) (0,0)
= -val
=
Y2
- V2 V
2 (1,0) (1,0)
-
Y2
.(0,0») C2Y1 -
. (0,0) . (0,0») CIY2 - c2Y1
y~O,O) )
(yiO,O) -
-val
(
-
COS
2r,
2 (2,0) (0,0) V a1 Y2
t
~
.(1,0)
v2 V \'£lY2
P2v2a~1,0) (yiO,O)
(1,0) a1 (
8
-
2V -
2
-
.(0,0) c1Y2
.(1,0») c2Y1
-
y~O,O») cos2r
-
+P2 v2 (yp,O) _ y~l,O») cos2r, .. (0,1) +
Y2
2 (0,1) V
.. (0,2) +
Y2
V
Y2
=
-v 2 aiO,1)yiO,0) _ Mv 2
2 (0,2) _
(C1Y~0,0)
2 (0,1) (0,1) (0,2) 2 (0,0) - a1 al v Y2
- -v Y2
Y2
_ c 2y i O,0») 3 _
_ C2yiO ,0») 3 ,
-
V
2M
(2.5.49)
(0,1) ( (0,0) a1 c1Y2
Mv 2 (3c~ (y~O,O) )2y~O,l)
-6c?c2Y~O,O) y~O,I) y~O,O) + 6c 1 c~y~O,O) yiO,O) yiO' 1)
+3c1c~y~0,1) (yiO,O»)2 - 3c~(yiO,0»)2yiO,1)
-3e~(y~o.O) )2e2ylo.l») , .. (I,I) Y2
+ y(I,l) 2
_ _ v2a(I,l)y(0,0)
-
1
. (c I
y~O,O)
2
-
-
v2a(O,l)y(l,O) 1
c 2y iO,0») 3 _
2
-
v2a(1,0)y(0,1) 1 1
2
- V
v2 M ( 3c~ (y~O,O») 2y~ 1,0)
2M (1,0 a1
2.5 Nonstationary Nonlinear Systems
53
(2.5.50)
54
2. Discrete Systems
(2.5.51)
2.6 Parametric and Self-Excited Oscillation
_P1 2
+
(
(-v
2a O'1) Bo i
55
+ ~Mv2c~A5Bo + ~MV2c~B3) v 2 -1
~MV2c3A2BO - !Mv 2c3B 3) 4 2 4 2 2(v 2 - 9)
BO] ° _ P1a(O,l) 1 2
°
= O.
(2.5.52)
In order to avoid time-consuming calculations, we assume that B o = O. Then (1,1) 3- (1,0) 3 3 2 P2 v2 9 2 3- 2 P2 v2 A2 A a1 = '4Ma1 c2 O- MC1C22(v2 _ 1) . gAo - SMC 1C2 (v 2 _ 9)
°
2 (0,1)
2 3- 3 1 - 3 2 P1 va 1 - 64 MC2P1Ao - 64 MC2AO + 2(v2 - 1)
1 Mv2c~A5
+ -8
(v 2
-
9)
+
°
3M 2 3A2 VC2 8(v2 - 1)
p1aiO,l) 2
(2.5.53)
Substituting the calculated values ap,O), aiO,l), ap,l) into the third equation of (2.5.47), we obtain the analytic form of the limit of the loss of stability, which depends on J.L, c and the other parameters.
2.6 Parametric and Self-Excited Oscillation in a Three-Degree-of-Freedom Mechanical System 2.6.1 Analysed System and Equation of Motion We consider a mechanical system with friction induced self-excited oscillations and parametric oscillations [14d, 17d]. The latter come from a rotor with rectangular cross-sections, which has a mass concentrated eccentrically at its centre. The rotor is fixed on the base placed on the belt moving with a constant velocity. At a certain value of the belt velocity and the frequency of rotor revolutions, parametric and self-excited'oscillations appear in addition to the forced oscillations caused by the unbalanced rotor. The diagram of the system is presented in Fig. 2.7. A weightless shaft with a rectangular cross-section and with a cylinder-like mass concentrated at its centre is supported by the base placed on the belt moving with constant velocity va. The friction coefficient between the belt and the base depends on their relative velocity. The character of this dependence (Fig. 2.8) causes self-excited vibrations. The effect is described in basic works on nonlinear vibrations. On the other hand, considering the nonidentical cross-section of the rotor at various values of its rotational speed, parametric vibrations occur. The vibrations cause a change to the normal force holding down the base to the belt in the vertical direction, and hence they cause changes to the friction force. It is assumed that despite the vibration of the rotor, the contact between the base and the belt is maintained.
56
2. Discrete Systems
m
k'l
.....-Vo Fig. 2.7. Diagram of the system
The calculation model of the system is presented in Fig. 2.9. The equations of motion of the system have the form mXc = -ewke cos
Iz"ep
= -Mo + a(-ewke cos 'Po
(2.6.1)
+ 71wk17 sin <po),
where
a,
coordinates of the centre of mass of the cylinder, mass moment of inertia of the cylider with mass m in relation to the axis z" of the system 0"x" y" z" moving with translatory motion in relation to Oxyz, coordinates of the point of puncture by the shaft in the coordinate system 0'e71, coordinate system whose axes are parallel to the main, central inertia axes of the cross-section of the shaft, shaft rigidities in the direction of the axes e and 71, driving torque reduced by all the resistance torques, parameters characterizing the position of the centre of mass of the disk C in relation to the point of puncture by the shaft.
For the states near the steady states, the torque M o is very small. Let I zIt =
·2 m'l.s'
(2.6.2)
where is is the inertia radius. Then the third equation of (2.6.1) will assume the form . c
+ 71wk 17 sin 'Po ) .
(2.6.3)
As the eccentricity a and the shaft deflection ew and 71w are small if compared to the inertia radius, the following can be assumed
2.6 Parametric and Self-Excited Oscillation
57
J.L
Fig. 2.8. Dependence of the friction coefficient on the relative velocity
ep = 0,
= w = const,
(2.6.4)
The following geometric dependences result from Fig. 2.9:
ew =
(x w - x) cos
Tlw = (x w - x) sin
+ Yw cos
Yc = Yw
+ acos(
=
+ asin(
Xc
Xw
(2.6.5)
where x w , Yw are the coordinates of the point of puncture by the shaft W in the system Oxyz. In order to write down the equation of motion of the mass M, it is necessary to determine the dynamic reactions on the shaft at its points of support. They are determined from the equations
Xl + X 2 + ew k { coswt + TlwkT/ sinwt = 0, Yl + Y2 - ewk{ sinwt + TlwkT/ coswt = 0,
(2.6.6)
where Xl, Yl and X 2 , Y2 denote the support reactions on the left and right end of the shaft, respectively. The rotor reactions on the supports are
Rx
= -X l -X2 ,
Ry = -Yl
-
Y2 .
(2.6.7)
The equation of motion of a body with mass M, on the assumption that Mg + R y > 0, has the form
Mi = -kx - eX + Rx w = Vo -
+ (Mg + Ry)J.L(w) ,
x.
(2.6.8)
The dependence of the friction coefficient on the relative velocity w can be described by the polynomial J.L(w) = e sgn w - ow
+ j3w 3 .
(2.6.9)
58
2. Discrete Systems
I·
x
I Xl x O~---4------~:""----.t-T----r---\-----r-----a--~ X2 x'
,
,
y.,=y.,
Yc=Yc -,::y(/"
x"
0" C mg
x.,
y"
Y'
Y
Fig. 2.9. Calculation model of the system
Finally, the equations of motion of the system, after assuming that x = Xl, Xw = X2, Yw = Xa, have the form t 2 Xl = -Xl (n + nl + n~ + (nl- n~) cos2wt) - HXI
-X2 (-(nl
+ n~)
- (nl- n~) cos2wt) - xa(nl- n~) sin2wt
+[g - X2(nl- n~)sin2wt - Xa (-(nl + n~) + (nl- n~) cos2f.Vt) + XI(nl- n~) sin 2wt](csgn (vo -
xd + {3( Vo - xd 3 ) , Xl (w~ + w~ + (wl- w~) cos2wt) -X2 (wl + w~ + (wl- w~) cos 2wt) + x3(wl- w~) sin·2wt - 0:( Vo
X2 =
xd
-
(2.6.10)
+a: 2 w sin(wt + CPo), x~ = -xI(wl- w~) sin 2wt + x2(wl- w~) sin 2wt +X3 (-(wl + w~) + (wl- w~) . cos 2wt) + OtW 2 cos(wt + \00) + g, 2 where: n = kiM, nl = k~/(2M), n~ = k."/(2M) , H = elM, wl k~/(2m), w; = k."/(2m).
2.6 Parametric and Self-Excited Oscillation
59
2.6.2 Transformation of the Equations of Motion to the Main Coordinates Let us introduce the following notation n~
=
nl +
2 2 WI =W~ ..
n;,
+Wq2 ,
a cos
ni = nl- n~, W22 =We2 -WTJ2 , a sin
H = J1.H 1, a: X = -, 6
{3 p= -,
( ) 2.6.11
6
J1.G = g,
where J1. = w~ / w~ = n~ / n? = (k e - kTJ) / (k e + kTJ) is the perturbation parameter. After accounting for (2.6.11) in the equation system (2.6.10), it will assume the form Xl = -Xl n 2 - Xl n~(3 + J1. cos 2wt) - J1.Hl I l +X2 n~ (1 + J1. cos 2wt) + X3n~ J1. sin 2wt +6(g - X2n~ J1. sin 2wt + Xl n~ J1. sin 2wt +X3n~( -J1. cos 2wt) )(sgn(vo - It} - X(vo - It}
+ p(vo -
X2 = -xlw~(1 + J1. cos 2wt) - x2w~(1 + J1. cos 2wt) +X3W~J1. sin 2wt + J1.(P sin wt + Q sinwt)w 2 , X3 = -xlw~J1.sin2wt + x2w~J1.sin2wt - x3w~(1- J1. cos 2wt) +J1.(Pcoswt - Qsinwt)w 2 + J1.G.
It}3), (2.6.12)
If we introduce J1. = 6 = 0 into the equation system (2.6.12), we obtain the homogeneous Linear differential equation system
Xl + Xl(n 2 + n~) - X2n~ = 0, X2 + W~(X2 - xt} = 0, X3 +W~X3 = O.
(2.6.13)
When we assume the solution of (2.6.13) in the form we find the following frequencies
Xi
=
Ai
cos pt, i = 1,2,3,
vb = ~ (n2+ nf +wf ± JU12 + nf +w?)2 - 4n2w~) , p~ = wr
(2.6.14)
Let us introduce the main coordinates el, for which at J1. = 6 = 0 the uncoupling of the linear part of the first two equations of the system (2.6.12) will occur. Let us now multiply these equations by el and e2, respectively, and add the sides. The result will be as follows 2 x16 + xt{n + n~)6 - x2n~6 + X2e2 + x2w~6 - xlw~6 = J1. [ - Xl n?6 cos 2wt - HlIlel + 6X2n~ cos 2wt -6X3n~ sin 2wt
+ 6XlW? cos 2wt -
6X2W~ cos 2wt
(2.6.15)
60
2. Discrete Systems
+6X3W? sin 2wt
+ w2e2 (P sin wt + Q cos wt) ]
+€6 [g - x2n?tLsin2wt + x3w?(1- tL cos 2wt) +xln?tLsin2wt] [sgn(vo By denoting el e 2 = (n 2 + n~)el
xd -
X(vo -
xd + p(vo - xd 3] .
- w?6,
6 e2 = -n~el + w~6
(2.6.16)
we find
(n 2 + n~ - (P)6 -w~6 = -n?el + (w? - (}2)6 = O.
0, (2.6.17)
In order for (2.6.17) to be fulfilled for following dependence must occur
n2 +-n? n? - e 2
-w? w? _ e2
6
=
=
different from zero, the
(2.6.18)
(}l = Pl' From the second equation
"Yle~,
(2.6.19)
where
"Yl
6
= o.
Hence, e? = p? and e~ = p~. Let el = e~ and 6 = e2 be denoted for of the system (2.6.16) we find e~
and
.
n?
(
•
2 2' WI - PI
Making use of the dependences (2.6.16) and (2.6.19), (2.6.15) is transformed to
fit + XlP~ + X2"Y1 + X2P~"Y1
=
tL[XlP~"Yl cos 2wt - HlXl
-P~"YlX2cos2wt +P?"YlX3sin2wt + "YlW2(P sin wt + Qcoswt)]
+€ [g - X2n~tL sin 2wt + x3n~(1 - tL cos 2wt) + xln?tL sin 2wt] . [sgn (vo - xI) - X(vo - xd + {3(vo - xd 3]. (2.6.20)
e2
Analogously, for e2 = P2 the following notation is used: = ~~, while e~ = "Y2e~,
where
e
1
= e~ and
(2.6.21)
2.6 Parametric and Self-Excited Oscillation
61
If we take (2.6.16) and (2.6.21) into account in (2.6.15), the equation will
assume the form
Xl
+ XlP~ + X2'Y2 + X2P~'Y2 = J.L[XlP~'Y2 cos 2wt - HlXl -P~'Y2X2 cos2wt + P~'Y2X3 sin 2wt + 'Y2w2(Psinwt + Q coswt)] +e[g - x2I1;tJsin 2wt + x 3I1;(l - J.Lcos 2wt) + x l I1; sin 2wt] . [sgn (vQ - xd - X(vQ - xd + p(vQ - xd 3]. (2.6.22)
Let us denote Yl = Xl
+ 'YlX2, (2.6.23)
Y2 = Xl + 'Y2 X2· The reverse dependences can be determined from (2.6.23)
Xl = IhYl - IhY2, X2 = t/J(YI - Y2),
(2.6.24)
where 1 'Y2 'Yl ,132 = - - t/J= - - 'Y2 - 'Yl 'Y2 - 'Yl 'Yl - 'Y2 Let us additionally assume that X3 = Y3. Taking (2.6.23) and (2.6.24) into account in (2.6.22), (2.6.20) and (2.6.12), we obtain the following differential equation system
131 =
iii + P~Yl
=
J.L [p~'Yl (13lYl - 132Y2) cos 2wt - HI (13dll - 132Y2) -P~'Yl t/J . (Yl - Y2) cos 2wt + P~'YlY3 sin 2wt +w2'Y1 (P sinwt + Q coswt)] +e [g - J.LI1~t/J(Yl - Y2) sin 2wt + J.LI1~(13lYl - 132Y2) sin 2wt
+I1;Y3(1 - J.L cos 2wt)][sgn(vQ - 13lYl + 132Y2) -X(vo - 132Yl
ih + P~Y2
=
+ 132Y2) + p( VQ -
13lYl
+ 132Y2) 3] ;
J.L[p~'Y2(13lYl - 132Y2) cos2wt - HI (13lYl - 132Y2) -P~'Y2t/J(Yl - Y2) cos 2wt + P~'Y2Y3 sin 2wt +w2'Y2(Psinwt + Qcoswt)]
(2.6.25)
+e[g -.J.LI1;t/J(Yl - Y2) sin 2wt + J.LI1~(13lYl - 132Y2) sin 2wt
+ 132Y2) 13lYl + 132Y2) 3] ;
+I1;Y3(1 - J.L cos 2wt)][sgn(vQ - 13lYl
+ 132Y2) + p(vQ P~(13lYl - 132Y2) sin 2wt + P~t/J(YI
-X(vo - 13lYl
ih + P~Y3
=
J.L[ - Y2) sin 2wt 2 +P~Y3 cos2wt +w (Pcoswt - Qsinwt) + G].
2. Discrete Systems
62
= wt, we obtain
After introducing the dimensionless time r
iit + A~Yl
2 = /1[A("r'I(€IYl - €2Y2) cos2r - AI H- l(1hYII - {32Y2' )
+A~"YIY3 sin 2r
+ "Yl (P sin 7" + Q cos r)] +€[g + /1[}~(€IYl - €2Y2) sin2r + [}~Y3(1 -/1cos2r)]
. [A~ sgn(Vo - {31WY~ - {32WY;) PI +wp(AI V~ - {31Y~ + {32y~)3];
AIXdAIV~ - {31Y~ + {32Y;)
ih + A~Y2 = /1[A~"Y2(€IYl - €2Y2) cos 2r - A2 f lt ({31Y~ - {32Y;) +A~"Y2Y3 sin 2r + "Y2(Psin r
(2.6.26)
+ Q cos r)]
+€ [g + /1[}~ (€1 Yl - €2Y2) sin 2r + [}~Y3 (1 - /1 cos 2r)] . [A~sgn(vo -
{31WY~ -
PI +wp(A2V~ - {31Y~
{32WY;) -
A2X2(A2V~ - {31Y~ + {32Y;)
+ {32y~)3];
ih + A~Y3 = /1[ - A~(€IYl - €2Y2) sin 27" 2 -] +A~Y3 cos 2r + P cos r + Q sin r + A3 G ,
where
Yi
dYi, =dr
€k
= 13k -1/;,
-
G
G=2' P3
~ , 1 ,• = 1, 2 , 3,
\2, -_
A
w2
1
HI
Xk
=
= X,
HI -, PI
k = 1,2,
Pk
1 I
Vo
=
Vo -, PI
•
II
Vo
Vo = -, P2
2.6.3 Zones of Instability of the First Order The procedure of solving the system of equations (2.6.26) consists in assuming two perturbation parameters /1 and € connected with the parametric excitation and friction, respectively. The sought periodic solutions of Yi(r) are presented in the form of the double power series (i) (i) 2 (i) ( (i) (i) 2 (i) ) Yi () r , Yo,O+/1YO,I+/1 YO,2+"'+€ Yl,O+/1Yl,I+/1 Yl,2+'"
(26 27) +.....
where yi~L k, l = 0,1,2, ... must fulfil the condition of periodicity. Periodic solutions are only possible for certain values of the parameters A~ presented in the form of the analogous series . A~ = n 2+/1ao,1 +/12ao ,2+' .. +€ (0:1,0 + /1al,1 + /120:2,2 + ...) + ... (2.6.28) where ak,l k, l = 0,1,2, ... are unknown coefficients, which are determined from the condition of periodicity, avoiding in the solution terms unrestrictedly growing in time. For a first-order resonance n 2 = 1 we shall determine
2.6 Parametric and Self-Excited Oscillation
63
the parametric unstability zones, for which the frequency of parameter modulation fulfils, consecutively, the dependences w P1, W P2, and w P3. In series (2.6.27) and (2.6.28) for w P1 and w P2 we shall limit our considerations to the first powers of the small parameters J.L and €. On the other hand, for w P3 we shall limit ourselves in the calculations to the second approximation. In all three cases we shall assume that sgn(vo - (31 wYl - (32WY~) = 1. Let "'J
"'J
"'J
"'J
"'J
"'J
\2 ""2 \2 ""3
2
\2
2
\2
= V2 ,1""1'
(2.6.29)
= v3 ,l""1'
where P2 V21 =-, , P1 P3 V31 = - , , P1
and let us assume that V2,1 and V3 ,1 are not integers. Let us first consider the case w ~ P1, assuming that
ya~6(r) = ya~6(r)
=
O.
(2.6.30)
The assumption is accounted for by a weak conjugation of (2.6.26) for € « 1 and J.L « 1. For J.L = € = 0 we shall obtain an unlinked system of three linear differential equations. For the resonance coordinate Y~'T), the magnitude of the oscillation of the other two main coordinates should be of the order of the small parameters J.L and €. Let us substitute series (2.6.27) and (2.6.27) into the differential equation (2.6.26), taking into consideration the dependences (2.6.19) and (2.6.30) and the expansion 0:0,1 a1,0 ( ""1 = 1 + J.L-- + €-2- + .. .. 2.6.31) 2 After equating to zero the coefficients of the same powers € and J.L, we obtain the system of recurrent differential equations \
"'J
y(1)" 0,0
+ y(1) 0,0
-
(1)"
(1) + Y1,0
= -a1,OYo,0
Y1,0
O., (1)
9 + P~
- 1 - gX1 vo
-3gwp(v1)2{31y(1)1 o 0 ,0 (1)"
YO,l
(1)1 + gX1{31Yo,0 + gwp (vo1)3
+ 3gwpv1{32 (y(1)/)2 + gwp{331 (y(1)/)3. 0 1 0,0 0 ,0 ,
(1) (1) (1) + YO,l = -ao,lYo,o + 1'l€lYo,O cos 2r -fI1{32ya~6' + 1'l Psin r + 1'1 Q cos r;
64
2. Discrete Systems
(2)11 (2) (2) Y10 + V2,1 Y1,0 ,
2
=
V2,1 - {3 (1)1 gp,2 gv 2 v V" + gV2,1X2 1Y'0,0 2 2,1/'\.2 0 2 (V,,)2 {3 y(1)1 +gWpv32,1 (V,,)3 + 3gwv2,1 0 0 1 0,0 +3gwPV2, 1V~ {3?
(2.6.32)
(Y~~6 ,) 2 + gw p{3~ (Y~~6 ,) 3 ;
2 H- {3 (1)1 cos T - V2,1 2 1YO,0 +"Y2Psin T + "Y2Q COST; (3)II + v(2) y(3) - O. 3,1 1,0 , Y1,0 (3)11 2 (3) (1) (1). 2 P Q . 2 GYO,1 + V3,1YO,1 = -V3,1 c1Yo,0 sm T + COS T sm T + V3,1 .
(2)11 (2) (2) YO,1 + v 2,1YO,1 =
2 (1) "Y2v2,1c1Yo,0
Assuming the solution of the first equation of the system (2.6.32) in the form y~16 , = a1 cos T + b1 sin T we obtain from the second equation: p2
Y~~6" + Y~~6
=
;
(2.6.33) 3
gX1v~ + gwp(v~) + 2gwP(v~){3?(a~ + b~)
-
+COST[ - a1,Oa1 + gX{31b1 -
+~ gwp{3~b1an + sin T [ -
3gwp(v~)2{31b1 - ~gwp{3~b~
0:1,ob 1 -
+3gwp ( Vo') 2 {31 a 1 + 43 gW:P{313 b21a1
gX1{31 a 1
+ 43 gwP{313 a13]
(2.6.34)
<
+~gW{3V~{3;(b~ - a~) CO~2T + 3gwpv~{3?a1b1 sin 2T 1 +COS3T[ - 4gwp{3~b~
3
+ 4gwp{3~b1a~] 3
. [-14gwp{31a1 33 32] +sm3T + 4gwp{31b1a1 . From the condition of periodicity we obtain two algebraic equations -a1,Oa1
3gwp(v~)2{31 - ~gwp{3~ADb1 =
0,
3gwp(v~)2{31-~gwp{3~ADa1-a1,ob1=
0,
+ (gX1{31
-(gX1{31-
-
(2.6.35)
where A~ = a~ + b~. For non-zero a1 and b1 , the following relation must occur -a1,0 gX1{31 - 3gwp(v~)2{31 - ~gwp{3fA~ gX1{31 - 3gwp(vO)2{31 - ~gwp{3fA~ - 0:1,0
= o.
(2.6.36)
Hence, (2.6.37)
2.6 Parametric and Self-Excited Oscillation
65
The only real solution of (2.6.37) is 0::1,0 =
0,
A21 = X
-33wP(v~)2
(2.6.38)
4 w p(3r The following function is the solution of (2.6.34) (1)
Y1,0 =
-, (')3 3 '(32A 2 p?9 - gX1 Vo + gwp Vo + "2gwpvo 1 1
+~gwpv~(3i(b~ - a~) cos 2r + gwpv~(3~a1b1 sin 2r + (3129WPi3:b~ + 3329WPi3~bla~)
. cOS 3T
(2.6.39)
1 gwP (331a13 - 32 3 gwP(331 b2 )' 3 1a1 sm r. + ( 32 The solution omits the general integral of the homogeneous equation by as1 sociating it with ya J. When we substitute (2.6.33) into the fourth and sixth equation of the equation system (2.6.32), after transformations, we obtain (2)"
Y10 ,
+ V22"1Y1(2)0
2
9
2
-"
= V2, 1-2 - V2 19X2 Vo P2'
+ gWP23, 1 (Vo")3
2 3 "(32A - (31b1 +"2V2,lgwpvo 1 1 + cos r [ v2,lgX2
-3gwPv~,1 (v~)2(31b1 + ~9Wp(3~A~b1] +sinr[ - V2,lgX2(310::1 + 3gwpv~,dv~)2(31a1 2 ] 3 " 2 2 +"43 gwp(33A 1 1a1 + "2V2,lgWPVO (3db 1 +3V2,lgwpv~(3ra1b1sin 2r
(3)"
Y1,0
2 (3) + v3,lY1,0
2) a1 cos
2r
2)
+ (
-41 gwP(313b31 +
1 3 "2gwP(31 b1a 1 cos 3r
+ (
-41 gwP(331a13 + "43 gwP(313b2)' 1a1 sm 3r;
= O.
(2.6.40) (2.6.41
)
The following functions are the particular solutions of the above equations (2) 9 - 2Vo" Y1,0 = 2'X P2
2 3 "(32A + gwpv 2,1 (Vo")3 + -2-gwpvo 1 1 V2,1
+ V?, 11-
- (31 - 3gwpv2,1 2 (")3(3 1 [ v2,lgX2 Vo 1
+ v 2. 1-
1 [ - V2,lgX2(31 a 1 +
2 1
-
"43 gwp(313 A 3] 1 b1 cosr
3gwPv~,1(v~)2(31a1
(2.6.42)
66
2. Discrete Systems
and (3)
(2.6.43)
YI0 , = O.
By substituting (2.6.39) into the third equation of the system (2.6.32), we obtain
y~~I" + y~~1 = ( -CtD,lal + ~'l'1olal + ( -CtD,lb l + ~1'lolbl
HilMI + 'nQ) - Hlihal
+ 1'I
cOST
P)
sin T
(2.6.44)
+ ~"YIClal cos 3r + ~"YIClbl sin 3r. We avoid terms unrestrictedly growing in time in its solution if the following equations are fulfilled (CtD,1 -
-
~1'101) al + H l l3l bl = 1'IQ,
-Hl{31 a l
+
(
aO,1
1)
+ "2"YI CI
.
\
(2.6.45)
b1 = "YIP.
For the case P = Q, after transformations, we obtain from (2.6.45): 2
P 2 aO,1 = ± ( Ai"Y1
±
1
2 2
+ 4"Yl cl
P (AI "Yd
4
- 2 2
- HI {31
P2)!
'2 P
+ "Yl C1 A~ - 2H- 1{31 A~ "Yl3 Cl 4 2
(2.6.46)
The particular solution of (2.6.44) is (1) 1 YO,1 = -16 "Yl c l(al cos3r
.
+ bl sm3r).
(2.6.47)
Taking (2.6.33) into consideration in the fifth and seventh equation of the system (2.6.32), we find the particular solutions.
y~~1 =
1 V?1 _ 1 GV?'I1'20 I a l -1I2,IH213Ibl
+ vl
1 _ 1 (l
+ 1'2Q) COST
~V~,I1'20 1bl + V2, I H2131 a I + ')'2P)
sin T
2.6 Parametric and Self-Excited Oscillation 2
2
V2,11'2 Cl V2,11'2 c l . + (2 9) al cos 31' + ( 2 ) bl sm3T; 2 v 2,1 2 v2 ,l - 9
y~~l =
G+
v 2 1_1 a,l
(
67
2.6.48)
(-~Vj'lclbl + p) COST
1 + 2 1_ (-2 vj,lclal + va,l 1
Q) sinT
Va2 lCl Va2 lCl + (2' )blcos3T+ (2' )alsin3T. 2 va,l - 9 2 va,l - 9
(2.6.49)
We have thus determined the particular terms of the series (2.6.27) and (2.6.28), limiting the calculations to the first approximation. Let us now concentrate on the analysis of the case w rov P2. The solutions will be sought, as has been done previously, in the form of the series (2.6.27) and (2.6.28) for i = 2. From (2.6.28) we obtain al,O
.A2 = 1 + c-2 Let us denote \2 "'1
=
0:0,1
+ J.L-- + . . . . 2
(2.6.50)
2 \2 VI ,2"'2,
\2 2 \2 "'a = va,2"'2'
(2.6.51)
where Vl,2 = Pl/P2' Va,2 = Pa/P2' and Vl,2 and Va,2 are not integers. Analogously to (2.6.30), we have (1)
Yo,o( 1')
(a) = Yo,o
= O.
(2.6.52)
Substituting (2.6.27) into (2.6.26), with (2.6.50), (2.6.3) and (2.6.52) taken into account, after equating to zero the coefficients of the same powers of J.L and c, we obtain (1)" 2 (1) 9 - 2' {3 (2), a ( ,)a Yl,O + Vl ,2Yl,O = p~ - gxv l ,2 vO- gX Vl,2 2Yo,o + gwpv l ,2 Vo 2 (V')2 Yo,o (2), + 3gWPVl,2 {322VO '( Yo,o (2),)2 +3 gwp{32Vl,2 o
+gwp{3~ (Y~~6,)a j (1)" 2 (1) 2 (2) - {3 (2), YO,l + Vl,2YO,1 = -1'lVl,2c2YO,O cos 21' + Vl,2 H l 2Yo,o +(3l Psin l' + 1'1 Q cos Tj y(2)" + y(2) - O., 0,0 0,0 -
(2.6.53)
(2)" + (2) (2) + 9 -" - {3 (2), ( ,,)a Yl,O Yl,O = -O:l,OYO,o P~ - gX2Vo - gX2 2Yo,O + gwp Vo
+3gWp(v~)2{32y~~J' + 3gwpv~{3~(Y~~6,)2 + gwp{3~(y~~J,)a; (2)" (2) YO,l + YO,l
(2)
(2)
-
(2),
= -ao,lYo,o - 1'2c 2Yo,O cos 21' + H2{32YO,0 +1'2PsinT
+ 1'2Q COST;
68
2. Discrete Systems Yl,O
(3)11
2 y(3) + V3,2 1,0
- O.
(3)11 YO,l
3 (3) + V3,2YO,l
= V3,2€2YO,0
-
, 2
sm 2T + P COS7
(2).
-
Q' 2 G sm T + V3,2 .
After substituting the following expression in the fourth equation of the system (2.6.53) y~2J ,
= a2 cos T + b2 sin T
(2.6.54)
and using the trigonometric relations, we obtain (2)"
Y1,O
+ Y1,0 (2) _ .!!.... _ 9X v" + gwp(v")3 + ~gwp(vll)2(a2 + b2) - p~ 2 0 0 2 0 2 2 +( -
al,Oa2 -
gX2j32b2 +
3gwp(v~)2j31b1 + ~gwpj3~b~
+~gWpj3~b2a~) cos T + ( -3gwp (vo")2j3 2a2 3 gwPVo"j32(b2 +2" 2 2
al,ob 2 + gX2j32a2
. T 2 sm + "43 gwp j323b22a 2 - "43 gwp j332a3)
2) - a2
cos 2T
+3gwpv~ j3ia2b2 sin 2T + ~(b~ - 3a~)gwpj3~b2 cos 3T +41 (2 a2 -
3 . 3T. 3b2) 2 gwpj32a2 sm
(2.6.55)
From the condition of periodicity of the S9lution, we get
-"I,oa. + ( - g5l..130 + 3gwp( v~)' (3. (gX.{3. - 3gwp(';,{)' (3. -
\
+ ~ gw p~A~ )
~ gw p~A~ ) ao -
"I
b. = 0,
,Db. = O.
(2.6.56)
where A~ = a~ + b~. For the non-zero a2 and b2 the main determinant of equation system (2.6.56) must equal zero. From this condition we obtain a1,O = 0, A 2 _ X2 - 3wp(v~)2 2 ~wpj3~
(2.6.57)
The particular solution of (2.6.55) is
y~~J
=
g2 P2
gX2v~ + gwp(v~)3 + ~2gwpv~j3iA~
-2"1 gwpvo"j322 (b22
2) - a2
cos 2T + gwPVo"j32. 2a2b2sm2T
( 2.6.58)
+ 32 (3a~ - b~)gwpj3~b2 cos 3T + ;2 (3b~ - a~)gwpj3~a2 sin 3T. 1
2.6 Parametric and Self-Excited Oscillation
69
Making use of (2.6.54) in the first and sixth equation of system (2.6.53), we obtain their particular integrals 3 '{32A2 + gwp v 1,2 (')3 Vo + -2-gwfJVo 2 2 V1,2
(1) 9 -, Y10 = ""2 - gXl Vo , P1
+ II~ 2 _ 1
1 (-9;\1111,2130
,
+ 3gwPI30Il~,2(V~)2 + ~gwP{j~A~) b2 COS T
{33 2)
.
(gX1 - V1,2{32 - 3gwp{32V1,2 2 (Vo')2 + -43 gwP 2A 2 a2 sm r + 21 V12 1 , 3V1 2 2, 2 2 + 2( 2' ) gwp{32 v o(b2 - a 2) cos 2r v 1,2
+ 2(
-
3V1
2 2 '_ v 1,2
(2.6.59)
1
2 ,
)
1
•
gwP{32 v oa 2b2 sm 2r
1 2 2 gwp{3~ 1 2 2 gwp{3~ . +-4 (b 2 - 3a2) 2 9 b2 cos 3r + -4(a2 - 3b 2) 2 a2 sm 3r V 1 ,2 -
V 1 ,2 -
9
and (3) -
Y1,0
-
0
(2.6.60)
.
The substitution of (2.6.54) into the fifth equation of system (2.6.53) gives (2) (2) YO,l + YO,l
=
(
-QO,l a2
1 - '212c2a2 + H- 2{32 b2 + 12Q ) cos r
+ ( -aD, 1b2 + ~'Y2g2b2 -
H2{j2a2
+ 'Y2P) sin T
+~12c2a2cos3r - ~12c2b2sin3r.
(2.6.61)
The following expression is obtained from the condition of periodicity after transformations and after assuming that P = Q: 4 QO,l
2 (- 2 2 1 2 2 p2 2) ( - 2 2 1 2 2) 2 + 20: 0 ,1 H 2{32 - 412c2 - A~ 12 + H 2{32 - 412C2 1 4 2 p2 2 _ p2 ( +"212c2 A2 - 212 {32 H 2A2 H 2{32 - 12c 2) = 2
o.
(2.6.62)
2
Hence
(2.6.63)
70
2. Discrete Systems
The particular integral of (2.6.61) is the following
Yo(2)1 ,
12 e ). =-2 ( a2 cos 3'T + b 2 sm'3'T 16
(2.6.64)
On the other hand, after substituting (2.6.54) into the second and seventh ,equation of (2.6.53), we find the particular solutions (1) YOI = ,
2
1 (1 1 (1 1
V 1 ,2 -
2 ) --2Vl,2/le2a2+Vl,2Hl{32b2+/lQ COS'T
2 -VI2 2/l e 2b2 - VI 2H- l{32 a 2 + lIP) sin'T 2v 1,2 - 1 2 ' , Vl,2/le2 Vl,2/l e 2 . . - ( 2 )a2cos3'T - 2( 2 _ 9)b 2 sm3'T, 2 v l ,2 - 9 v l ,2
+
(a) YOI =G+ ,
1 (1 1 (1 2
Va,2 -
+ 2 va,l - 1
1
(2.6.65)
2 e -Va2 2b2+ P ) COS'T 2 '
2 -2 va,2e2a2 -
Q) sm'T .
(2.6.66)
2 2 Va,2e2 Va,2 e2 . + ( 2 ) b2 cos 3'T - 2( 2 ) a2 sm 3'T. 2 va ,2 - 9 va ,2 - 9
Finally, let us consider the case of W ~ Pa. Periodic solutions are possible for the particular value of the parameter A,a: A, ~ 1 + e al,O + aO,l + 2 a2 ,O + 2 aO,2 + al,l + (2.6.67) J-L 2 e 2 J-L: 2 eJ-L 2 ... a2
.
I
Let us denote \2 2 \2 Al = vl,aAa, A,22 = V2,a 2 A,2a,
( 2.6.68 )
where Vl,a = PI/Pa, V2,a = p2/pa, and Vl,a and V2,a are assumed not to be integers. Similarly to the previous considerations, assuming that
y~~J('T) = Y~~6('T)
= 0
(2.6.69)
we obtain the following recurrent differential equation system from equation system (2.6.26) (1)" 2 (1) 2 (a) Yo,o - + Vl,aYl,o = (g + (}l Yo,o)
(1)"
2
(1)
2
(a).
[
Vl,a 2 -, a, a 2 p~ - Vl,aXl Vo + WPVl,a(Vo) ] ; .
YO,l +Vl,aYo,l = IlVl,aYo,osm2'T+/l(PSm'T+Qcos'T); (1)" Y2,O
2 (1) 2 (1) ( r'l2 (a)) [al,o 2 ~ , + Vl ,aY2,o = -Vl,aal,OYl,O + 9 + J&lYO,O p~ - Vl,aXlal,OVO - ( {3 (I), {3 (2), 2 (Vo')2 (al,O -Vl,aXl - lYl,O + 2Yl,O + WPVl,a -2-Vl,aVo,
2.6 Parametric and Self-Excited Oscillation
71
(2)')] + J'I £"}2 YI,O (a) [ P~ 1 - VI,aXI 2 - Vo , + WPVI,a a (Vo, )a] ; + {3 IYI,O (I)" 2 (l) _ 2 (I) 2 ((I) (2») 2 YO,2 + v I ,aYo,2 - -vI,aQo,IYO,I + VI ,a'1'I €IYO,I - €2Yo,I cos r (I) - {32YO,I (2)') +VI,a H- 1 ({3 IYO,I (a) + YO,I (a»). +VI2,a'1'I ( ao,IYo,o sm 2r; (I)" 2 (l) 2 ({I) (I») 2 ( (I) YI,I + VI,aYI,I = -VI,a aO,IYo,O + aI,oYO,I + VI ,a'1'I €IYI,O -
€2yi~J) cos 2r +
VI,aHI
({3Iyi~J' - {32yi~J')
0 +V~,a'1'1 (QI 'Oy~aJ + y~aJ) sin 2r + (g + {}~y~aJ) [a 2,1 " , Pa - a O,I Vo, - VI,aXI - ( - {3IYO,I (I), + {32YO,I (2»)] -VI2,2Xl 2 ( ')2 (2 V, QO,I {3 (I), {3 (2)') +WPVI,a Vo VI ,a o-2- - IYO,I + 2YO,I
(2.6.70)
£"}2( (a) (a) )( 1 2 -, a ( , )a) +Jel YO,I + Yo,o cos2r P~ - vI,aXIVo +WpvI,a Vo ; (2)" 2 (2) 2 (a) V2,a -" 2" a 2 YIO + V2 aYI 0 = (g + {}IYO 0) -2 - V2,aX2 Vo + WPV2 a(Vo) ] ; , J' ' P[ 2 ' (2)" 2 (2) 2 (a). 2 (P . Q ) YO,I + v2,aYo,I = v2,a'1'2Yo,0 sm r + '1'2 sm r + cos r ; (2)" 2 (2) 2 (2) ( £"}2 (a») [QI,O 2 - " Y2,0 + V2,aY2,0 = -V2,aQI,oYI,0 + 9 + JeIYO,O P~ - V2,aX 2Vo QI,O - ( - {3 IYI,O (l), (2)') + WPV 2,a ( Vo")2 -V22,aX2 + {32YI,0 2 . (aI,o 2 v2,a v" 0 - {31y{l), 1,0 + {32y(2)') 1,0 ] £"}2 YI(a)0 [ -2 1 2 -V " V2 ( II)a] ; +Jel , P-aV2' aX2 o + WP 2,a Vo (2)" 2 (2) _ 2 (2) 2 ((I) (2») YO,2 + V2,aYo,2 - -V2,aaO,IYo,I + V2,a'1'2 €IYO,I - €2YO,I cos 2r - ({3 (I), {3 (2)') +V2,a H 2 IYI,O - 2YO,I 2 ( (a) (a»). +V2,a'1'2 ao,IYo,o + YO,I sm 2r; (2)" 2 (2) 2 ((2) (2») 2 (l) YI,I + V2,aYI,I = -V2,a aO,IYI,O + aI,OYO,I + V2,a'1'2(€IYI,0
+€2y~~J) cos 2r - v2,aH2({3Iy~~J' - {32yi~J) 2
(
(a)
(a).
+V2,a'1'2 aI,oYo,o + YI,O) sm 2r £"}2 (a») [aO,I 2 " - ( {3 (2), +( 9 + JeIYO,O P~ - V2,aX2 a O,I VO - V2,aX2 - IYO,I
7'2
2. Discrete Systems + -
(2),) (32YO,1
2 ( ")2 ( ,,00,1 + WPV2 ,3 V o l.I2,3 V O
2
(l), (2)') ] 2( (3) (3) {3lYO,l + {32YO,0 + [11 YO,l - Yo,o
2-
cos 2) T
2 ( ")3] .,
1 - V2,3X2VO"+ WPV2 ,3 Vo . [ p~
o·
(3)" + y(3) 0,0 , Y 0,0 (3)" (3) _ (3). Yl,O + Yl,O - -Ol,OYO,O'
y~~l" + y~~l
=
-00,lY~~6
(3)" (3) Y2,0 + Y2,0
= -Ol,OYl,O - 02,OYO,0'
+
Y~~6 COS2T +
(3)
QsinT
PCQ;ST -
+
OJ
(3).
(3)" (3) (3) (3) (I) . 2 YO,2 + YO,2 = -ao,lYo,l - 00,2Yo,0 - clYO,l sm T (2) . (3) +c2YO,1 sm 2T + OO,lYo,o
cos 2T
(3) + YO,l
cos 2T
+ 00,1
a-.,
(3)" (3) (3) (3) (3) (3) Yl,l + Yl,l = -Ol,lYO,O - Ol,OYO,l - 00,lYl,o - clYl,O (I) . 2 + (3) -ClYl,O sm T OO,lYo,o
cos 2T
+
(3)" (3) (3) (3) (3) Yl,l + Yl,l = -Ol,lYo,o - Ol,OYO,l - OO,lYl,o (2) . (3) +c2Yl,0 sm 2T + Ol,OYo,o
cos 2T
00,1
G~',
+ 01,0
a- .
cos 2T + (I) . 2 ClYl,o sm T (3) YO,l
(3) + Yl,O
cos 2T
After substituting
y~36 ,
= a3
cos T
.
,
+ b3 sin T
(2.6.71)
into the twelfth equation of system (2.6.70) we get (3)" (3) Yl,O + Yl,O
For non-zero 01,0
a3
=
( -01,0 a3 COST +
and
b3
b3 sm . T ).
(2.6.72)
from the condition of periodicity we get
o.
(2.6.73)
= O.
(2.6.74)
=
Hence, (3) Yl,O
Making use of (2.6.71) in the first and sixth equation of (2.6.70), we obtain their particular solutions (l) Yl,O =
[g n? -2- + 2 Vl ,3
1 (a3
COST + b3
. T) ] sm
Vl,3 -
2 - + WP V l,3 2 (Vo')3] ,
. [ V?,3 p~ - Vl,3Xl VO
(2.6.75)
2.6 Parametric and Self-Excited Oscillation
73
2
n
9 + 2 1 (aa cos 'T + ba sin 'T) ] yi~J = -2[ v2 ,a v2 ,a - 1 .
[v~,a
2 -" p~ - V2,aX2Vo
" + WPV22 ,a(V o)
a]
(6
2..76)
.
After substituting (2.6.71) into the thirteenth equation in system (2.6.70), we have
y~~I" + y~~1 ~ {; + ( -aa,la3 + ~a3 + + ( -aa,lb3 - ~b3 -
Q)
p)
cosr
sin r
(2.6.77)
+ ~ (a3 cos3r + b3 sin 3r).
The condition of periodicity gives ~ 2 1
",,(1) _
'-L0,1 -
(2) _
a O,l -
+
P
,
a3
Q b '
-"2 -
(2.6.78)
3
The particular solution of this equation is the following
y~al,
= (; -
1 (aa cos 3'T + ba sin 3'T)
16
(2.6.79)
•
When substituting (2.6.71) into the second and seventh equation of system (2.6.70), we obtain their particular integrals (1)
YO,l
=
"Y1 2 v 1,a -1
v2
+ 2( ~,a
(v?,a) - ba + Q 2 "y
1 )
v1a , - 9
(2) _ YO,l -
"Y2 2 v2 ,a -
v2
1
(v~,a) -2 ba + Q
+ v 2 a"Y11,
-
ba cos 3'T)
cos 'T
1
(2.6.80)
,
+ v 2,a"Y2-
1
2
"V
+ 2( V 22,aa -12 9) 2,
(aa sin 3'T
COS'T
(
• 3 aa sm 'T
-
ba cos 3) 'T .
(2.6.81)
After substituting (2.6.71) and (2.6.73) into the fourteenth equation of system (2.6.70), we obtain the following from the condition of the existence of periodic solutions: a2,0
= O.
(2.6.82)
Analogously, taking (2.6.71), (2.6.79), (2.6.80), (2.6.81) into account in the fifteenth equation of system (2.6.70), we obtain equations which, after transformations, will assume the form
74
2. Discrete Systems
1 1) 1'2C2V~,a (1 1)
(1) _ 1'l c 1V?,a (v?,a 0:0,1 4 2
+
+ v?,a - 9
~22,a -
4
1+
~22,a - 9
1 (1) 1 P (C11'1 C21'2) +"2 00 ,1 - 32 - aa 2(v?,a - 1) - 2(v~,a - 1) ,
c V o (2) - - 1'l 1 ?,a 0,2 -
4
1'2 C2v~ a + 4'
~0(2)
_
- 2 0,1
(1 (1 v~,a
v?,a - 1
~_
1) 1) v~,a
+--=-2-v 1,a - 9
- 1+
- 1
Q ( C11'1 + c21'2 ). ba 2(v?,a - 1) 2(v~,a - 1)
32
(2.6.83)
(2.6.84)
The following algebraic equation system will be obtained from the condition of periodicity of the solutions of equation system (2.6.70) after substituting (2.6.71), (2.6.74), (2.6.75), (2.6.76) and (2.6.79) into the sixteenth equation of system (2.6.70): c2 c 2!1? c1c1!1? -O:l,l a a - 2(v?,a _ 1) ba + 2(vl _ 1) ba = 0, a C1 c1!1? c2c2!1~; -O:l,l ba - 2(v?,a _ 1) aa + 2(v~,a _ 1) da = 0,
(2.6.85)
where C1 =
2 v1,a
-2 -
2 - VI + 2 ( ,)a v 1a X1 O wpv1a Vo ,
PI'
,
2
v2,a
2 -
I
C2 = - 2 - v2 aX2 Vo P2'
+ wpv ' 2 a( V ,)a . 2 o ,
(2.6.86)
From the condition of a non-zero solution of equation system (2.6.85) in relation to aa and ba, we obtain
0:(1,~) = 1,1
± [!1? ( 2C2 C2 _ 2
v 2,a
-
1
C1 C 1 )]. 2 - 1 v 1,a
(2.6.87)
The coefficients of the series (2.6.67) are determined by expressions (2.6.73), (2.6.78), (2.6.83), (2.6.84) and (2.6.87).
2.6.4 Calculation Examples The analytically obtained diagrams of parametric instability zones are presented below in order to illustrate the influence of particular parameters of
2.6 Parametric and Self-Excited Oscillation
75
the system on their magnitude and position. The physical parameters of the system are given in the form in which they occur in differential equation (2.6.12). Figures 2.10 and 2.11 present the influence of the unbalance J.LP, the damping (J.LHt) , and the shape of the friction characteristic (0:/{3) on the magnitude of the parametric instability zones for PI and P2, for the following data: [J2 = 900 s-2, = 480 s-2, = 4800 s-2, 9 = J.LG = 9.81 m s-2, VQ = l O.4ms- , e = 0.2. On the basis of (2.6.14), PI = 73.32s- 1, P2 = 28.35s- 1, and P3 = 69.288- 1 have been obtained. The adequate coefficients assume the form ')'1 = 0.833, ')'2 = 0.12, {31 = 0.126, {32 = -0.674, el = 1.176, e2 = 0.376. The other quantities characterizing the system have been marked in the figures (when 0:/ (3 = 0:/ p). The parametric instability zones presented in Figs 2.10 (for pd and 2.11 (for P2) expand with the increase of the unbalance J.LP, while, depending on the value of the quotient 0:/ {3, this tendency can have different intensity. In the case of 0:/ {3 = 0.5 m 2s- 2, the doubling of the unbalance has caused the instability to be expanded twice for the zones corresponding to PI and P2. For 0:/ {3 = 1 m 2s- 2, the unbalance has been increased three times, which has brought about a comparatively small expansion of the instability zones for PI, while for P2 the expansion is still almost doubled. In the case of large unbalance of the rotor, the changes of the quotient 0:/{3 do not influence the magnitude of the parametric instability zones. The influence of damping on the magnitude of the instability zones corresponding to the frequencies PI and P2 is also very different. Small damping (J.LH I = 0.05 s-l) causes considerable shift of the zone for P2 in the direction of the growing value of the modulation depth J.L (J.L > 0.15). In the case of the double increase of damping the zone will not occur for J.L ~ 0.3. The magnitude and position of the instability zones for PI are not sensitive to the changes of the damping coefficient. In the case of J.LH I = 10s- 1 the proper zone of the frequency PI exists for J.L > 0.034. When J.LHI = 20s- 1 damping is doubled, and the lower border of the occurrence of the zone is shifted to the value of J.L = 0.07. The parametric instability zones for P3 are presented in Fig. 2.12. The magnitude of the zones depends on the initial conditions of the system motion. The diagrams have been prepared on the assumption that a3 = b3 = 0.01 m, where a3 = Y3(0), b3 = Y3(0). The calculations, in the case of the resonance coordinate Y:3, have been performed with with a precision of up to second order, hence the inclination of the unstability zones in the direction of the growing values of the parameter ..x~ has appeared. For the first approximation, the zones remain symmetrical in relation to the straight line ..x~ = 1. As in the cases considered above, the increase of the unbalance considerably expands the instability zone. The changes of the value of the quotient 0:/ {3 and damping have a negligible influence on the magnitude of the zone. Figure 2.13 presents the parametric instability zones for various
n?
w?
76
2. Discrete Systems
a)
0.7 0
b)
0.1
0.2
0.3
1.3 - r - - - - - - - - - - - - - - - - - - - - - , H1=0, /-LP=21·1Q-4 m .x 21 HI =0, /-Lp=7·1Q-4 m 1.2 HI=P=O
1.1 1.0
0.9 0.8 0.7+------.....--------,r---------l
o
c)
0.1
0.2
0.3
1.3,...--------------------,
.x~ 1.2
0.8 0.7+-------.--------,.--------l
o
0.1
Fig. 2.10a-c. Instability zones for PI
0.2
0.3
2.6 Parametric and Self-Excited Oscillation
a)
1.02 rHl=O, ttp=7·1Q-4 m
,\~
rHl=O, ttP=3.5·1Q-4 m
1.01
r
HI - P - O
c.
1.00
'-P:
O. ttHI 0.05s- 1
0.99 I
0.98
b)
o
0://J=0.5m2s- 2
1
0.2
0.1
J-L
0.3
1.02
,\~
rHl=O, ttP=14.1Q-4 m
1.01
rHl=O, ttP=7·10-4 m
r Hl =P=O ~
1.00
""'C.p 0, ttHI 0.05 S-1
0.99
10://J= 0.65 m 2s- 2 1 0.98
c)
o
0.2
0.1
J-L
0.3
1.02 ,\2
rHl=O, ttP=21.1Q-4 m
2
1.01
rHl=O, ttP=7·1Q-4 m , H1
1.00
p 0 ~ '-p 0, ttH} 0.05s- 1
0.99
0.98
2 2 I 0://J= 1 m s- ]
o
0.1
Fig. 2.11a-c. Instability zones for P2
0.2
J-L
0.3
77
78
2. Discrete Systems
2.0.......----------------~
\2 "'3
ttP=ttQ=21.1Q-4 m ttP=ttQ =7.10- 4 m
1.5
l.0t========:::::::::::=;;;;~
0.1
0.2
J-L
0.3
Fig. 2.12. Instability zones for 2 2 P3 (0./(3 = 0.5m s- , 0./(3 = 0.65m2 s- 2 , 0./(3 = 1 m 2 s- 2 )
values of the parameters {}2, {}? and w? For the zones denoted by 1 we get {}2 = 14400s- 2 , {}? = 1920s- 2 , w? = 19200s- 2 ; for the zones denoted by 2 we have {}2 = 3600s- 2 , {}? = 480s- 2 , w? = 4800s- 2 , and for the zones denoted by 3: {}2 = 900s- 2 , {}? = 120s- 2 ,"w? = 1200s- 2 . In all the cases the magnitudes of the other parameters are as follows: J-LHI = lOs-I, e = 0.2, o/f3 = 0.5m2 s- 2 , VQ = 0.5ms- 1 , J-LP = 0.0015m. As shown in Fig. 2.13a,b the growth of the squares of frequencies {}2, {}?, w? (resulting from the increase in rigidity of the elastic elements in the system, or from the decrease in the values of the masses) causes the instability zones for PI and P2 to expand. For example, when the parameters {}2, {}?, and w? increase by four times, it brings about an approximately doubled expansion of the zones. The unstability zone for pa is not influenced by the frequency changes in the system (Fig. 2.13c). Figure 2.14a, b presents the influence of the velocity changes of the belt VQ on the magnitude of the instability zones for PI and P2. Calculations have been performed for the data denoted by 1, except for the velocity VQ, whose value has been changed. In each case the increase in the belt velocity causes the expansion of the instability zones. For VQ < 0.3 JIlS- 1 these changes are less evident. The influence of the velocity changes VQ on the parametric instability zone for Pa is practically negligible. We can summarize the obtained results as follows. (1) The method of seeking a solution as a power series of the two perturbation parameters J-L and e used in the considerations makes it possible to
2.6 Parametric and Self-Excited Oscillation
a)
79
1.5 1
2
c
1.0
3
0.5J------..---------,..----------l o 0.1 0.2 {L 0.3
b)
1.5
\2 ""2
1 2
0.5
c)
3C
C
1.0
o
0.1
0.2
0.3
1.5
1.0
0.5+--------..---------,..-------~
o
0.1
0.2
{L
0.3
Fig. 2.13. Influence of parameter changes on the instability zones for: (a) Pl; (b) P2; (c) P3
80
2. Discrete Systems
investigate the single resonances of any order for the systems with weak nonlinearity and weakly modulated systems (J.L « 1). When we perform calculations with a p,recision of up to second order, it turns out that the limits of instability zones incline in the direction of the growing values of the parameter >.~ (Fig. 1.7). For the first approximation, the limits remain symmetrical in relation to the straight line >.~ = 1. (2) The parametric instability zones for Pll and P2 expand with the increase in the rotor unbalance. Depending on the value of the quotient alp, this tendency has different intensity. In the case of a/ (3 = 0.5 m 2 s-2, the double increase in the unbalance has brought about a considerable expansion of the unstability zones, for PI as well as for P2. For a/(3 = 1 m 2 s-2 the unbalance, causes a rather small expansion of the instability zones for PI, while for P2 the expansion is still almost doubled. In the case of a lack
a)
1.5
r
tb=O.4ms- I
; ; '\)=0.35ms- 1 I Vo 0.25ms-
r
"- tb-0.15ms-I
1.0
! \
0.5
b)
0
0.1
1.5
\2
A
2
c
1.0
0.5
o
c:
0.1
0.2
0.3
,;;=r
'\)=0.4 ms-
/r
vo=0.25ms- I
~
I
tb= 0.35 ms- I
tb°.15ms-I
0.2
Fig. 2.14. Influence of the belt velocity changes PI; (b) 112
J.L Vo
0.3
on the insta.bility zones for: (a)
2.7 Modified Poincare Method
81
of rotor unbalance the changes of the quotient a/{3 do not influence the magnitude of the parametric instability zones. The influence of damping on the magnitude of the instability zones corresponding to PI and P2 is also very different. The minimum damping (J.LHI = 0.05s- 1 ) causes a considerable shift of the zone for P2 in the direction of the growing values of the modulat,ion depth (J.LHI = 0.15s- 1 ). The magnitude and position of the instability zones for PI are not so sensitive to the damping coefficient changes. The regularities indicated here are the more clear, the greater the difference between the values of the frequency PI and P2 (i.e. for PI » P2) is. The increase in the unbalance also produces a considerable expansion of the instability zone for P3; however the changes of the parameter and of damping have no essential influence on the magnitude of the zone. The growth of the frequency squares n2 , n? and w~ causes the expansion of the instability zones for PI and P2. The parametric instability zone for the frequency P3 is not sensitive to the frequency changes in the system. In the case of the belt velocity increase (vo), the instability zones for PI and P2 are expanded. This property is noticeable within the range of great velocities (vo > 0.4ms- 1 ). The influence of the velocity changes Vo on the parametric instability zone for P3 is practically negligible. (3) For the frequencies PI and P2, the position of the instability zone limits does not depend in the first approximation on the initial conditions of the system motion. The magnitude of the instability zone limits for P3 depends on these conditions.
2.7 Modified Poincare Method 2.7.1 One-Degree-of-Freedom System Consider a one-degree-of-freedom nonlinear system governed by the equation [33, 117]
ddt + 2
y
2
W
2
Y = eQ
(d y, dt ,e
Y )
.
(2.7.1)
The above e > 0 is a small perturbation parameter and the function Q is analytical with regard to its arguments y, dy/dt and e for 0 < e < co. We are going to find a periodic solution to (2.7.1) depending on the perturbation parameter e. The system under consideration is autonomous, therefore we can arbitrarily take
dY(O) = 0 (2.7.2) dt ' because starting with the initial conditions for to, (2.7.1) will not change with a shift of time.
82
2. Discrete Systems
For c = 0 we have d2 y _ + w2 Yo - 0, Yo(O) = 0, dt and a solution to (2.7.3) is given by -2
(2.7.3)
yo(t) = A o cos wt,
(2.7.4)
where the amplitude A o is not yet defined. However, contrary to the previous investigations and following Proskuriakov [65d] , we are going to find an invariant orbit depending also on the second parameter b = b(c), where b(O) = O. Therefore, we have
y(O) = A o + b(c).
(2.7.5)
On the basis of the assumption of analycity of F, this function will be analytical also with respect to A o + b. We are focused on finding a periodic solution
y(t) = y(t + T), y(t) = y(t + T),
(2.7.6) (2.7.7)
and the period T is defined as T = To
+ a(c),
(2.7.8)
where
To = 21rw- 1 , a(O) = O.
(2.7.9) (2.7.10)
From (2.7.6) and (2.7.7) we get
y(To + a, A o + b, c) = y(O, A o + b, c) = A o + b, y(To + a, A o + b, c) = y(O, A o + b, c) = O.
(2.7.11) (2.7.12)
A general solution form of (2.7.1) is a function of the two parameters b and c: K
y(t, Ao + b, c) = (A o + b) coswt
+ LYk(t, A o + b)c k,
(2.7.13)
k=l
and K
y(t, Ao + b, c) = -w(A o + b) sin wt + LYk(t, A o + b)c k .
(2.7.14)
k=l
From (2.7.13) for t
Yk(O, Ao + b) = 0, Yk(O, Ao + b) = O.
=
0 and taking into account (2.7.2) and (2.7.5) we get (2.7.15) (2.7.16)
2.7 Modified Poincare Method
83
We develop the right-hand side of (2.7.1) into a power series of e in the neighbourhood of e = 0 putting fJ = 0 into series (2.7.13) and (2.7.14). We obtain (8 Q dy 8Q dy ad + F de e=O=O y e e=o=O y e=o=O 2 y + 8Q ) + e2(~ 8 Q (d )2 8e e=o=O 2 8 y2 de e=O=O 2 2Q +! 8 Q2 Y) + ~ (8 2 ) 2 8y de e=o=O 2 8e e=O=O 2 2 8 Q dy dy 8 Q dy 8 2Q dy +---+ --+ --8y8y de de e=o=O 8y8e de e=O=O 8i/8e de e=o=O 2 2 y 8Q d 8Q d y ) + O( e + - -2 +-2 8y de e=o=O 8y de e=O=O
. {. eQ(y,y,e) = e Q(yo,yo,O)
+e
(d
3)}
= e{ Q{yo,Vo)o +e (( ~~). YI + (~~)o VI + (~~)J
+e
2
[~ (~:~) °y? + ~ (~:~) °V? + ~ (~:~)
2 82Q ) . ( 8 Q) + ( 8y8i/ Yl Yl + 8y8e
2Q ( 8 ) 0 Yl + 8fj8e
+ (~~) °Yl + (~~) °V2] + 0{e
3 )}.
0
0
. Yl (2.7.17)
Now we demonstrate that using the periodicity conditions we are able to solve the problem, Le. to find A o, o:(e), and then fJ(e). To show this, let us begin with (2.7.12). From (2.7.14) and (2.7.2) for T = To + 0: we obtain y(To +
0:,
A o + fJ,e) = -w(A o + fJ) sinw(To +
0:)
(2.7.18)
K
+ LYk(To + 0:, Ao + fJ)e k = 0, k=l
which leads to the equation !<
- w(A o + fJ) sinwo:+ LYk(To +
0:,
A o + fJ)e k = O.
(2.7.19)
k=l
0:
The above solution allows us to find o:(e). For this purpose, we first express in an analytical form of e (2.7.20) o:(e) = eO:l(Ao + fJ) + e2O:2(Ao + fJ) + ... + ekO:k(Ao + fJ).
Each of O:k depends on A o,To and fJ and can be developed into the Maclaurin series in the neighbourhood of fJ = 0:
84
2. Discrete Systems
8Cik 1 82ak '2 ( ) 8A b + 2 8Ag b + ... , k = 1, ... , K. 2.7.21 o The corresponding ak can be obtained by differentiating (2.7.19) with respect to e and putting a = b = e = O. We find Cik(A o + b)
= ak(Ao) +
e
-w' An (~;) 0 + Yl(To,Aol = 0,
e'
-w'Ao
(~:~)o +2 [Yl (~;)o +y.] =0,
3
-w' An
(~:~)
e
+21
0
(2.7.22) (2.7.23)
+ 6{!i3 + (~~) 0 Yl + (~;)/.
(8a) 2 [oo. 8e 0 Y I
+"31 W 2 YI ] } ,
(2.7.24)
where 8 ka/8e k = Cik. If Yk are known then we can find Cik according to the subsequent equations (2.7.22), (2.7.23) and (2.7.24). Now we illustrate how we obtain Yk' For this purpose we have to reconsider the left-hand side of (2.7.1). Taking into account (2.7.14), we obtain
y(t, A o + b, e) = -w 2(A o + b) coswt +
K
L Yk(t, A o + b)e k.
(2.7.25)
k=l
Finally, from (2.7.1), (2.7.17) and (2.7.25) we get the following recurent set of linear differential equations (for b =r= 0) YI + W 2YI = Q(yo, Yo), . (2.7.26) Q Q Q .. (2.7.27) Y2 + w2 Y2 = (8fJy ) 0 YI + (88y )0.YI + (88e ) 0'
2Q 2Q 2Q 1 (8 ) 2 1 (8 ) .2 1 (8 ) Y3 + w Ya - 2 8y2 0 YI + "2 8 y 2 0 YI + 2 8e 2 0 2 2 82 Q ) . ( 8 Q) ( 8 Q) . + ( 8y8y 0 YI YI + 8y8e 0 YI + 8y8e 0 YI ..
2
_
+ (~~) /2+
(~~)
0
y.,
(2.7.28)
and so on, which allows us to find Yk(A o,To). To conclude, using condition (2.7.12) we have found ak, and therefore the unknown period T = To + a(e) is defined. Using condition (2.7.11) we illustrate how to find A o and b. For this purpose let us develop (2.7.11) into the Maclaurin series with respect to Ci
y(To,Ao +b,e) + Ciy(To,Ao +b,e) +
~a2Y(To,Ao +b,e)
1 a'" a Y (To, Ao + b, e) + ... - A o - b = O. +6
(2.7.29)
2.7 Modified Poincare Method
85
The left-hand side L of (2.7.29) can be described by the series
L = eLl (To, A o + b)
+ e 2 L2(To, A o + b) + ... + e k Lk(To, A o + b).(2.7.30)
Each of the Lk(To, Ao + b) term can be developed into the Maclaurin series with respect of b
8Lk 182Lk 2 Lk(To, Ao+b) = Lk(To,A o)+ 8A b+2" 8Ag b +... , k = 1 ... , K.(2.7.31) o Taking into account (2.7.31) in (2.7.30) we obtain the following evident form
~ (L k (rr' A) L = e L...J oLO, 0
o, A o)b + ~ 8 Lk(To, Ao) b2 + + 8 L k(T 8A 2 8A2 .. , 2
k=l
0 2
+
~ 8 Lk(To, A o)bk) k!
&A~
e
0
0 -.
k-l -
(2.7.32)
From (2.7.29) for e = 0 we get y(To, A o + b) = A o + b. Each of the terms y(m) = dmy/dt m depends on A o, To and b, and can be developed into the Maclaurin series in the neighbourhood of b = 0: (m) _ (m) 8y(m) ~ 8 2y(m) 2 y (To,Ao + b,e) - y (To,Ao) + 8A b + 2 8Ag b + ... o k 1 8 y(m) k + k! 8A~ b (2.7.33) Now for b = 0 we compare the terms standing by the same powers of e in (2.7.29), taking into account (2.7.20) and (2.7.30) and (2.7.32). First, we take b = 0 and from (2.7.29) we get
y(To, A o, c)
1
+ oy(To, A o,c) + "20?y(To, A o, c) + 6"1 a 2'"Y (To, A o, c) + ... - A o =
(2.7.34) O.
Comparing the coefficients of the same powers of e in (2.7.32) and (2.7.34), we obtain cO
el e2
Ll(To,A o) = yt{To,Ao) = 0, L2(To,Ao ) = Y2(To,Ao) + 0lYJ (To, A o ) - ~ll:~Aow2, . 2 L3(To, A o) = Y3(To, A o) + ll:2Yl (To, A o) + ll:lY2(To, A o) -Aow2ll:lll:2 - ~ll:~Yl(To,Ao).
(2.7.35) (2.7.36)
(2.7.37)
Equation (2.7.35), further called the amplitude equation, can possess two solutions: either yt{To, A o) = 0 or yt{To, A o) = O. In the first case this equation allows us to find the amplitude A o. If A o is real and not a multiplied root of the above equation, then it corresponds to the only real solution of (2.7.1).
86
2. Discrete Systems
In the latter case we divide (2.7.32) bye and L2(To, A o) = 0 will serve as the amplitude equation, and so on. We discuss further the first case. From (2.7.35) we calculate A o, and then (2.7.32) defines the following implicit fUl1ction of the two parameters band e
8L l 2 8L 2 1 282L l L(e, b) = eL 2 + b 8A + e L 3 + eb 8A + 2b 8A5 o o 3 2 8L3 1 2 82 L2 1 3 83 L l eb 4 +e L + e b 8A + 2 8A~ + '2 b 8A~ +.... o
(2.7.38)
According to the implicit function theorem, the number of branches of the implicit function b(e) is defined by the smallest power of bk in the following series 8L l 1 282 L l 1 383 L l 1 k 8k L l _ b b L(O, b) = b 8A + 2 8A5 + '6 8A~ + ... + k! b 8~ - O. (2.7.39) o According to (2.7.35) Ll (A o) = ydA o) and if all the derivatives up to the kth order 8LI/8Ao, 8 2 LI/8A~, ... , 8 k LI/8A~ vanish, then A o has kmultiplicity, Le. A~. In what follows k-multiplicity of Ao defines the number of branches of b(e). Each of the k-branches can be described by the 'following series b(e)
= Al/me l / m + A 2 / m (e l / m) 2 + A 3 / m (el/m) 3 + ... ,
(2.7.40)
where m equals one of the numbers 1,2, ... , k. In the case of fractional powers, the sum of the denominators cannot be greater than k. · For m = 1 the problem is reduced to the well-known situation of the classical series \
(2.7.41)
2.7.2 General Nonlinear Systems 2.7.2.1 Introduction. Let us consider an autonomous nonlinear system in the general form
dX -at =
F(X),
(2.7.42)
where:
10 F is continuous and X
E
IRn ;
2° F fulfills the Lipschitz condition: IIF(X) - F(Y)II ::; CIIX - YII, where C is constant; 3° a solution X is defined as: X = X(t, XCO)), where XCO) = X(to) (further we take to = 0).
2.7 Modified Poincare Method
87
Now we consider a set of first-order equations, which can govern not only mechanical, but also biological or chemical nonlinear dynamical systems. Of course, the set of second-order differential equations obtained using Newton's laws or the Lagrange equations can be easily reduced to (2.7.42). We are going to find only the periodic solutions to (2.7.42) with a period T > 0 taking the initial conditions X(O) in such a way that Z(T, X(O») = X(T, X(O»).
(2.7.43)
X(T,X(O») is the solution to (2.7.42), and X(O,X(O») = X(O). If we find the real X and T which fulfil the equation Z(T,X) = X,
(2.7.44)
the above map reduces the problem of finding the periodic solution of (2.7.42) to finding the fixed points of the map (2.7.44). Such an idea is often used during an application of numerical techniques to exhibit periodic solutions. 2.7.2.2 Autonomous System. Consider now the case where the righthand side of (2.7.42) depends on a small parameter e
s
= 1, ... ,n,
(2.7.45)
where the same assumptions as for (2.7.42) are valid and 0 ~ e < eo. Because (2.7.42) is autonomous, then for an arbitrary to if we take t + to instead of t in (2.7.45) this equation will not be changed. It implies that a solution to (2.7.45) does not depend on to and we can take to = O. Let us assume that for e = 0 the linear system (2.7.46)
has a family of periodic orbits with a period To. Let us assume that it corresponds to the existence of K pairs of V s eigenvalues ±ri/To, r = 0,1,2, ... , i 2 = -1. Even if they are multiple values but with simple elementary divisors, system (2.7.45) can be reduced to the form dxs
dt
=
-vsYs
+ eips(Xl,'"
,xk, Yl,···, Yk, Zl,···
,ZM, e),
dys . dt = vsxs + e1/Js(Xl,"" Xk, Yl,···, Yk, Zl,···, ZM, e),
d: = ~
dz·
L.- PijZi
+ ehj(Xl""
,Xk, Yb" . ,Yk, Zl,· .. , ZM, e),
i=l
s = 1,2, ... , K,
j = 1, 2, ... , M,
M
+ 2K =
n.
(2.7.47)
88
2. Discrete Systems For e
= 0 the linear system
dx s = -VsYs, dt
-
dys = VsX s ,
(2.7.48)
dt dz M - = LPijZi dt i=l
has a family of To periodic solutions depending on 2K arbitrary constants C s and D s of the form
Xs = Cs cos vst - D s sin vst, Ys = Cs sin vst + D s cosvst, Zj
(2.7.49)
= 0,
s = 1, 2, ... , K,
j
= 1, ... , n -
2K.
Because the system under consideration is autonomous, we can arbitrarily take one of Ys = 0 for to = O. For s = 1 we get (2.7.50)
Y1 = D1 = O. This means that instead of 2K variables we have to obtain 2K - 1. The solutions to system (2.7.47) have the following form
! ! t
Xl = (C 1 + Ci) cos v1t
+e
[CP1 COS)!"l (t - r) - t/J1 sin V1 (t - r)] dr,
o
t
Y1 = (C1 + Ci) sin V1t +e
[CP1sinv1(t - r) - tP1COSV1(t - r)] dr,
o
Xs = (C s + C:) cosvst - (D s + D:) sin vst t
+e
!
.
[CPs cos vs(t - r) - tPs sin vs(t - r)] dr,
o
Ys
= (Cs + C:) sin vst +
(D s + D:) cos vst
t
+e
J
[CPs sin vs(t - r) + t/Js cosvs(t - r)] dr,
s
=
2, ... , K,
o
! t
{z} = e[P]t{z·}
+e
o
e[P](t-r) {h}dr.
(2.7.51)
2.7 Modified Poincare Method
89
The above solutions for t = 0 have the following initial values
Xl(O) = C l +Ci(e), Yl(O) =0, xs(O) = Cs + C;(e), Ys(O) == D s + D;(e), {z} = {z·(e)},
s
= 2, ... ,K, (2.7.52)
where {z} = col(Zll ... ,Zn-2K), {z·} = col(zi,···,z~_2K)' [P] {h} =col(h l , ... , hn - 2K ), and these solutions possess a new period
+ a(e).
T = To
= [Pij], (2.7.53)
For e ---. 0 we have a(e) ---.0, C: ---.0, D: ---.0, {z·} ---. O. Now we show that solutions (2.7.51) fulfil (2.7.47). For this purpose we put (2.7.51) into (2.7.47) and we get
:t
{!
['P. cos 1I.(t - r) - ,p. sin 1I.(t - r)] dr } t
=
~
J
-vs
{!
[CPs sin vs(t - r)
+ 'l/Js cosvs(t - r)] dr + CPs,
o
['P. sin 1I,(t - r) +,p, cos 1I,(t - r)] dr } t
=
V
s / [CPs cos vs(t - r) - T/Js sin vs(t - r)] dr o
s= 1, ... ,K,
~
[j
e[P)(t-T) {h}
dr] = [P]
!
e[P)(t-T){h}
dr +
+ T/Js'
{h}.
(2.7.54)
'Ib prove the identity of the above equations, we use the Laplace transformation, remembering that t
.
Xl (r) * X2(r) == / Xl (r)X2(t - r)dr, o L[Xl(t) * X2(t)] = Xl (s)X 2(s), L[cos at] = L[sin at] =
s
(2.7.56)
2
+a
2'
(2.7.57)
2
+a
a
2'
(2.7.58)
S S
(2.7.55)
90
2. Discrete Systems
Therefore, we find that in fact the following equations are fulfilled 82 82
+ V;L(cps) -
8Vs 82
+ l/;L(tP s )
= - v [ 2 Vs 2 L (cp s) 8
+ Vs
+ 8 2 +8 V 2 L (tP s)] + L (CPs), s
2
8VS 8 L (CPS) + 2+ 2 L (tPS) 2 2+ 8 V 8 V
s
s
= Vs [ 8 2: V 2L (CPS) s
88 _
1 [p]L(h)
= [P] 8 _
8
2~
1 [p]L(h)
2 L (tP s )]
+ L(tPs),
Vs
+ L(h).
(2.7.59)
The temporarily unknown quantities C;, D: and z·, which depend on c, should be obtained in order to fulfil the periodicity conditions. As system (2.7.47) is analytical, therefore also its solutions, as well as the period, are analytical because of the initial conditions and the parameter c. The periodicity condition applied to solutions (2.7.51) will take the form
J T
Xl
(T) -
Xl
(0) = c
[CP1 cos V1 (T - r) -tP1 sin V1 (T - r)] dr
o +(C1 + C;)[cosv1T - 1] = 0,
J T
Y1 (T) - Y1 (0) = c
[CP1 sin V1 (T -17) - tP1 cos V1 (T - r)] dr
o +(C1 + C;) sin V1T = 0, T
xs(T) - xs(O) = c
J
[CPs cosvs(T - r) - 'l/Js sin vs(T - r)] dr
o +(Cs + C;)[cos vsT - 1]
= 0,
T
Ys (T) - Ys(O) = c
J
[CPs sin vs(T - r) + tPs cos vs(T - r)] dr
o
.
+(Ds + D:) [cos vsT - 1] = 0,
J T
{z(T)} - {z(O)}
= (e[P]T -
[I]) {z·} +c
eIP](T-r){h}dr = O. (2.7.60)
o According to the above equations, we have n equations with the unknowns
Ci, T, C:, D: (8 = 2, ... , K). It is easy to show that sin vsT = sin VsD:, cosvsT = COSVsD:,
(2.7.61)
2.7 Modified Poincare Method
91
and from the second equations of (2.7.60) it follows that
a(e)
=
ea:1(e).
(2.7.62)
For Ci = C: = D; = e = 0 (8 = 2, ... ,K) we get from (2.7.60) n equations with the unknowns 2K - 1, the initial conditions Cs , D s (D1 = 0), and the period To. In order to simplify further considerations we denote the left-hand side of (2.7.60) by F i (i = 1, ... , n), and additionally we introduce the new variables
XK = XK,
Y1
= XK+1,
To = aK+1, D2 = aK+2,
(2.7.63)
DK = a2K, •K+11 zl• = a2
Zn-2K =
Xn ,
The fundamental idea of the Poincare method is to find n parameters ai (i = 1, ... , n) and n functions ai(e) (i = 1, ... , n) from the equations
Fi(a1 +ai,· .. , a2K +a2K, a2K+1"'" a~, e)
= 0,
i = 1,2, ... , n.(2.7.64)
Because the general form of solutions (2.7.51) can not always be expressed by elementary functions (after necessary integrations), therefore to omit the potential difficulties, we look for the solution as a power series of the perturbation parameter e of the form co
Xi = ~,o
+
L Ri,l(a1 + ai, . ..
I
a2K
+ a2K, a2K+1"'" a~)el,
1=1
where
R1,O = (a1 R 2,o = (a2
+ an cos v1t, + a2) cos V2 t -
(aK +2
+ ai< +:G) sin V2 t ,
RK,O = (aK +ai<)cosvKt- (a2K +a2K)sinvKt, RK+1,O = (a1 + ai)sinv1t, R K+2,O = (a2 + a2) sinv2t + (aK+2 + ai<+2) cosv2t,
(2.7.65)
92
2. Discrete Systems
R2K,O
=
(aK
+ ai<) sin vKt + (a2K + a~K) cos vKt,
{:~:.o } e(P]'{a2~~+1 =
}.
(2.7.66)
For t = 0 from (2.7.65) and (2.7.66) we obtain
Rl,O(t = 0) = al + ai(c), R 2,o(t = 0) = a2 + ai(c), RK,O(t = 0) = aK + ai«c), RK+l,O(t = 0) = 0, RK+2,O(t = 0) = aK+2 + ai<+2(c), R2K,O(t = 0) = a2K + aiK(c), R 2K+l,O(t = 0) = a2K+l(c), Rn,o(t = 0) =
(2.7.67)
a~(c),
and ~,l(t = 0) = 0 for i = 1, ... ,no In general it can happen that some of the functions r i have the following structure F i = cf'i. The necessary periodicity condition is then equivalent to either c = 0 or f'i = O. The first assumption implies that the function Ri,O(t) is periodic for all parameters ai. In the second case we get <Xl
f'i
=L
R i ,l(al
+ ai,·.'·, T, ... ,a2K + aiK' aiK+l' a~)c'-l = O.
1=1
For c = 0 we have
(2.7.68)
a: (0) = 0, and from (2.7.68) we obtain
f'i(al, ... ,a2K,O, ... ,O,0) =0,
i = 1, ... ,2K.
(2.7.69)
Let us suppose that the algebraic nonlinear set of equations possesses one or a few solutions which we denote by
(2.7.70) and which can be multiplied. Then, we constrain such a solution to the periodicity condition
2.8 Hopf Bifurcation
Fi(iil +ai, ... ,li2K+a2K,a2K+I, ... ,a~,c)=0 i=l, ... ,n.
93
(2.7.71)
When we introduce ai = ... = a~ = c = 0, then we have the lefthand side of (2.7.71) equal to zero. We are going to find the functions ai(c) (i = 1, ... , n) on the basis of the theory of the implicit functions with the condition ai (0) = O. The functional determinant of (2.7.64) for (2.7.70) and for ai a~ = c = 0 has the form
an aal
Zr
1
an
i1 0 =
=0. ~rn al
(2.7.72)
~rn
an
If for iiI, ... ,iin we get i1 0 #- 0, then we have a simple solution, and in this case the unknown functions can be expressed by a~(c) = cAil +c 2A i2 +c 3A i3 + ... , (2.7.73)
which fulfils the condition ai (0) = O. If we get i1 0 = 0, then the solution is multiple and if the rank of the matrix built on the determinant is equal to n -1, then from (2.7.71) we exclude n - 1 quantities. We finally obtain a new nonlinear algebraic equation of the form
+ ai, c) = 0, where for 2K + 1 < i ~ F(iii
i
= 1, ... ,n,
(2.7.74)
n we have iii = 0, and i is equal to one of the numbers
from 1 to n. For the unknown iii the problem reduces to finding ai(c) as the implicit function of c in the form of the power series of either the total or fractional powers of the parameter c. The number of the solution branches as well as their series representations depend on the multiplicity solutions of (2.7.69). The latter is defined by the first nonvanishing term of 8 k F/ 8ai . In the case when the rank of the matrix is equal to n - 1, n is large, or when the rank of that matrix is less than n - 1, there is still a method proposed by McMillan [64d].
2.8 Hopf Bifurcation Bifurcation from equilibrium to dynamics is associated with the so-called Hopf bifurcation [116]. Up to now there have been many papers addressing this phenomenon, which is encountered in many branches of science [26, 27, 89, 91, 98, 15d, 16d, 18d-20d, 45d, 46d]. Here, we would like to present some examples of the analytical approach based on both the perturbation and harmonic balance methods similar to those obtained by Huseyin and his co-workers.
94
2. Discrete Systems
We begin with an analytical method of determining the post-critical family of the periodic solutions of some autonomous ordinary nonlinear differential equations dependent on one bifurcation parameter. It is known [116] that when the bifurcation parameter reaches the critical point, and when the system of linearized differential equations originating from the initial system of equations has complex conjugate eigenvalues of the Jacobian (the other values have non-zero real terms), and after fulfilling additionally the Hopf conditions, equilibrium path stability is lost and a family of periodic solutions appears. Denoting the bifurcation parameter as "T/", the following case has been considered: for T/ < T/c (where 1]c is a critical point) the system has a stable equilibrium path. For T/ = T/c the complex conjugate eigenvalues of the Jacobian of the characteristic equation occur (the others have negative non-zero terms), and together with an increase in T/ (T/ > T/c) the real terms of the eigenvalues become positive. The Hopf conditions ensure that the intersection of the imaginary axis with the complex conjugate eigenvalues occurs with non-zero velocity. A method for searching for bifurcation solutions was presented in [91,98]. Nevertheless, it is troublesome to use because of the time-consuming calculations for equations of dimension bigger than two. As the Hopf bifurcation occurs in nonlinear systems, it would be desirable to use for their analysis the analytical methods widely known in the field of nonlinear vibrations. The most popular and effective are the methods of harmonic balancing and the perturbation method. The main defect of the harmonic balancing method, which makes it useless for Hopf bifurcation analysis, is the necessity of a priori knowledge about the solution. An adv..antage of the perturbation method lies in constructing the solution by the subsequent solving of the perturbation equations of the linear differential equations, when it is only necessary to know the solutions of the undisturbed differential equation system. The combination of the two methods makes it possible to solve the Hopf problem (the method of harmonic balancing is used to solve each of the perturbation equations of the linear, differential equations). We consider the system of differential equations, whose characteristic equation is of the form (u - ut)(u - (2)P(U) = 0,
(2.8.1)
where
Ul.2 = e(T/) ± iW(T/),
w(T/c)
i= 0,
e(T/c) = 0,
°
a~c) i= 0,
and the roots of the polynomial P (u) = have negative real parts. It results from the centre manifold theorem [88] that the critical subsystem is mainly responsible for the bifurcation and bifurcated solution, and for the qualitative assesment of the bifurcated solution it is possible to limit oneself only to the solution of the two-dimensional critical differential equation. Here this solution serves as the initial approximate solution of the full nonlinear differential equation system, and the "detailed'· solution is determined by
2.8 Hopf Bifurcation
95
the method of slllccessive approximations. The latter also makes it possible to solve the problem where there are nonanalytical nonlinearities. Let us consider the differential equation system having the form
d dt (x) = F(l1, x),
n
x E IR
(2.8.2)
,
where T/ is the parameter vector and F (T/, x) is a nonlinear function, analytical in the state variables T/ and x. For the purpose of the further analysis it has been assumed that T/ is a one-dimensional bifurcation parameter. Let XQ fulfil the equation F(T/, XQ) =
o.
(2.8.3)
Examination of the stability of the equilibrium pathxQ is known to be limited to the determination of the eigenvalues of the Jacobian Fx(T/, xQ), where Fx(T/,xQ)
(~:~)
=
]
, (i,j = 1, ... ,n).
(2.8.4)
x=xo
Let the equilibrium path XQ for T/ < T/e (the critical value of the parameter) be the stable solution of system (2.8.2). On the other hand, when T/ = T/e the two complex conjugate eigenvalues cross the imaginary axis with non zero velocity, Le. let 0"1
= e(T/) + iW(T/),
w(T/e) = We
i= 0,
(2.8.5)
For T/ > T/e, the real parts 0"1, 0"2 become positive. A family of periodic solutions is created at the critical point. Let us assume that equation system (2.8.2) can be presented in the form .
Ku(T/)u
+ Kv(T/)v + K(T/,u,v),
U
=
iJ
= Su(T/)U + Sv(T/)V + S(T/,u,v),
2
IR , v E IRn - 2 , U
E
(2.8.6)
where x = colon(u,v), and the matrices KC*)(T/), and SC*)(T/) are the linear parts of the expansion of F(T/, x) into the Taylor series in the equilibrium path XQ. Let the characteristic equation (2.8.6) have the form of (2.8.1), while p ((j) = 0 has roots with negative real parts, and 0"1 and 0"2 are the eigenvalues of the two-dimensional matrix Ku(T/). The matrix K u, known also as the critical matrix, decides about the Hopf bifurcation. From the centre manifold theorem it follows that in the neighbourhood of the equilibrium path XQ = 0 there exists a function v = f(u) which in a sufficiently close neighbourhood XQ = 0 has the property 8f/8u = O. This allows us to assume to a first approximation that
v = f(u) =
o.
(2.8.7)
96
2. Discrete Systems
Taking into account (2.8.7) in (2.8.6), we obtain the following .
U
= Ku(Tf)u
+ K(Tf,u,O),
U
E
2
lR .
(2.8.8)
Later we shall assume that K(Tf, u,O) = K(Tf, u). Let us develop the matrix K(Tf) into a Taylor series in the neighbourhood of the critical point Ku(Tf)
= Ku(Tfc) +
1 2 K U17 (Tf - Tfc) + "2KUT1T1(Tf - Tfc) + ... ,
(2.8.9)
where etc. Let u(Tf) = 0 be the solution of (2.8.8) for 7J < Tfc, and for Tf = Tfc the periodic solution u(t; e) = u(t + T; e) of the period T bifurcates, which is dependent on one formally assumed small parameter e connected with the amplitude. After the transformations this parameter can be arbitrarily assumed to be e = 1. In order to obtain the period T = 211", we shall introduce the dimensionless time r = wt and, as a result, we will obtain the following expression from (2.8.8): d W(e) dt u(r;e) = K(Tf)u(r;e) + K(Tf, u(r;e)).
(2.8.10)
The periodic solution u( r; e) will be sou~ht in the form of a certain Fourier series, where the amplitudes and the freqttencies depend on the parameter e: K
ui(rj e) = 2:)Pik(e) cos kr
+ Tik(e) sin kr).
(2.8.11)
k=O
Because the system (2.8.10) is autonomous, then Tl1(e) = O. Moreover, pik(e), Tik(e), Tf(e) and W(e) are developed into a power series of the parameter e of the following form .()_ ' +1,,2 P,k e - Piko +Pike "2Pike
.()_ T,k e - Tiko Tf (e") -_ Tfc
W(e) =
We
+ ... ,
+' 1"2 + ... , Tik e + "2Tike
+' Tf e + "21 Tf " e 2 +
(2.8.12)
... ,
1 + W'e + "2w"e2 + ... ,
where Piko = Tiko = 0, because ui(rj 0) = 0 at the critical point. The solution of ui(rj e) is also sought in the form of a power series ui(rje) =
u~(r)e) + ~u~'(r)e2 + ... ,
(2.8.13)
2.8 Hopf Bifurcation
97
where K
up (r) = 2:)p~~ cos kr + T~J sin kr).
(2.8.14)
k=O
We can now proceed in two ways. We can either introduce relations (2.8.11)-(2.8.14) into (2.8.10) and by comparing terms in the same power c obtain the perturbation equations of the linear differential equations, or obtain these equations by means of successive differentiation of (2.8.10) with respect to c. As an example, let us consider the mechanical system with 1 ~ degrees of freedom, presented in Fig. 2.15. The vibration equations of the system have the form
fil Ultt = -k l Ul - k(Ul - ua)a + (au~ CUat = -kaua - k(ua - ut}a.
1J)Ult,
(2.8.15)
c Fig. 2.15. Mechanical system with 1 ~ degrees of freedom
The bifurcation parameter 1J is related to damping, and the other coefficients in (2.8.15) are positive. After applying the first step of the perturbations, we obtain
1 1 a 1 -3- aul + -1JUl! fil fil fil = -k l Ul - k(Ul - ua)a, ka k a = --Ua - -(ua - ut} . c c
WUl T = -U2 WU2T WUaT
(2.8.16)
98
2. Discrete Systems
The roots of the characteristic equations (2.8.16) are 0"12 ,
~2 (..2L.. ± (..2L..) 2 _ ml ml
=
4k l ), ml
k3
(2.8.17)
= --.
0"3
c
= TIc = 0 we have
For TI
±iJ
l (2.8.18) k . • ml The first two equations of (2.8.16), after the assumption of U3 = 0, have the form 1 1 a 3 WUl r = -TlUl + - U 2 - -3-Ul' ml ml ml (2.8.19) WU2r = -k l Ul - ku~, <112
= ±iwc =
and we get
fff fff
1, ml
l'
- u lr = -u2'
ml
' (2.8.20) -l u' = - klUI' ml 2r Having taken into account Til = 0, we will obtain the following solution of Eqs. (2.8.20)
,
,
u l = Pu cos T, U2 = -.JklmlP~l sin T.
(2.8.21)
From the second system of perturbation equations, we obtain " P2l
" = T2l " = T22 " = TI , = W ' = 0. = P22
(2 .8.22)
Finally, the third syste~ of perturbation equations, after taking into account (2.8.22), has the form 1 1 1 1 a -w u'" + -w"u' = --u'" + --n"u' - __ (u'}3 6 c lr 2 lr 6ml 2 2ml 'f 1 3ml 1 , 1
'"
6WcU2r
W U2r + 21"
=
-61 k'" lUI -
k(u ,l )3 .
(2.8.23)
Comparing the terms in sin T and cos T in (2.8.23), we obtain 1",
- 6 WcPll
1",
1",
= '2 w Pll + 6ml T2l ,
1 '" 1" , 9a ( , }3 + -2TI Pll - -4- Pll , ml ml ml w c P2l '" = k lTll' '" 1 ", kl '" 3 k( ') 1" ~ , 6WcT2l = -"6Pu - 4 Pu + 2w V 1\;1mlPUl 1
",
-6 WcTll = -6-P2l
(2.8.24)
2.8 Hopf Bifurcation
99
and therefore we have '" = T fI' = P21 21 = 0 ,
'" TH
'"
9 k ( , )3 -4:k PH , 1
"
="43 Vkk1 ml (')2 P11
P11 =
W
(2.8.25)
I
, ):2 . 7J " = '920: (PH
Comparing the terms in sin 3r and cos 3r, we obtain 1
'"
-'2WCP13
1
1 III 6ml T 231
=
'"
1
III
-2WcT13 =
-6- P23
1
'"
k IT IN 131
1
'"
'2WCP23 = '2WcT23 =
ml
kIlO
-(fP13
-
30: ( , )3 - 4 PH'
(2.8.26)
ml
)3 + 4:1 k(p'11·
and then we get ", 3 k ')3 P13 = -16 k (P11 , 1 '"
T 13 = III
P23
27 0: (')3 16 Vm1k 1 PH ,
(2.8.27)
9 (')3 = 160: Pu I
III
T 23 =
9
16k
Vff!!!(')3 k; P11 •
From the third equation of (2.8.16), we obtain k3 k 3 + -U 1 · e e After equating the terms in sin r and cos r of (2.8.28), we obtain WU3T
", T 31 -
= --U3
(2.8.28)
3k(p'11)3
---::----~~--__:_-
4C( ~+ (~r~)' (2.8.29)
On the other hand, after equating the terms in sin 3r and cos 3r of (2.8.28), we obtain
100
2. Discrete Systems rill -
k(pll)3
-~-_-.::::....=..::~---:=---~
33- 4C(3~+~(!;r~)' k 3k(pll)3
pili _
(2.8.30)
!;)
33 - 12c2 (3~ + ~ ( 2) .
We shall limit ourselves to terms in (Pll)3 in calculations. The periodic bifurcation solution has the form 3k ( I )3 , 1 (PH ')3 cos 3T UI = Pl1 cOST - 8 kl Pl1 cOST - 32 a (p/)3' (2831) r=-r:11 sm 3T, •. 32 ymlk l ~ I 3 I 3 3 k §I( I )3 . 3 u2=-ymlkIPl1sinT+ 16 o (Pl1) COS3T+ 16 VIC; Pl1 sm T,
9 +-
U3
=
3k3k(pll)3
4c2( ~ + (~ r) +
cOST
+
3k(pll)3
4c( ~ + (!; r~)
k(pl1 )3
4C(3~+~(!;r~)
sin T
cos 3T
i I
+
kk3 (pld
3
12c2(3~ + ~(!;
sin 3T
r) ,
Finally, we obtain the amplitude parameter-frequency relations
(k;
3
k
"
W=V~+8~(Pll)
2
(2.8.32)
and the parameter-amplitude relations 1]
=
9 2 0 (Pll) . 4 I
(2.8.33)
2.9 Stability Control of Vibro-Impact Periodic Orbit 2.9.1 Introduction It is well known that mechanical vibro-impact systems have been widely employed in both theoretical and applied mechanics for a long time. Vibroimpact dynamics can be observed in many real engineering systems, such
2.9 Stability Control of Vibro-Impact Periodic Orbit
101
as hammer-like devices, ball-and-race dynamics in a ball bearing assembly, wheel-rail impact dynamics, etc. [41d, 63d]. Nowadays again this field of research has attracted strong interest, but in the framework of theories of modern dynamical systems. Recent industrial examples (tube fretting wear through vibro-impact behaviour in nuclear reactors or impacts between old and high buildings excited by earthquakes) belong to additional but not satisfactorily solved questions of discontinuous dynamical systems. There are two parallel branches of investigations in the framework of vibro-impact dynamics. The first one is based on a better approximation of laws for impact motion and restitution coefficients, and it is more involved in the physics of materials. The second branch includes control of steadystate vibro-impact motion with the possibility of stability changes (either to destabilize or to st,abilize the vibro-impact attractor). Recently many papers have appeared, which are devoted to control of nonlinear oscillators, including also control of chaotic orbits [66d, 145]. In general, these methods could be devided for feedback control with a time delay [71d], sliding mode control [146], repetitive control [43d] , iterative learning control [25], adaptive control [146], and so on. The main purpose of these methods is to control complicated systems, even with imprecise knowledge of their mathematical models. However, the control of the attractor or repeller is based on the numerical observations of the results by the introduction of "helping" control coefficients. Theoretical prediction are rather not given. Here we address one, not yet satisfactorily solved problem of vibro-impact dynamics control with delay feedback and we give an analytical prediction of the proper choice of control parameters.
2.9.2 Control of Vibro-Impact Periodic Orbits We analyse the following one-degree-of-freedom vibro-impact system with one clearance presented in Fig. 2.16. The equation of dynamics is as follows:
x + ci + o:2 x = Po coswt + A [x(t) +B [x(t) - x(t - T)] and
x+ = x_,
x+ = -Rrx_
for
x(t - T)] + for x < x>5
5,
(2.9.1)
where: Po = Yok 2 /m, c = cI/m, 0: 2 = (kl + k 2 )/m, A = k 2 aI/m, B = k 2 bI/m, and T = 21f / w is the period of the periodic orbit being stabilized. A key point of this control is that a periodic solution possesses the same period as the excitation, Le. Xo = xo(t - T) and Xo is a particular solution of both the controlled and uncontrolled system [52d]. The delay loop is switched off where perturbations are not present. In the case of perturbations the controller causes the perturbation to vanish more quickly than in
102
2. Discrete Systems
x
a)
s
b)
Y
I
X ...
,(" Y - mX+Ct X + (k t +"=2)x= k 2 y I '< ~
~
~
'I'
Ys memory storage
-
.-
--
x
Ys = at[x(t) -x(t-T))+ f+-
+ bt[x(t) -x(t -T))
---
Fig. 2.16. One-degree-of-freedom kinematid,lly excited vibro-impact system with one clearance (a) and its control diagram (b) (8 denotes the clearance)
the case without control. The problem of analytical estimation of the influence of control coefficients for periodic orbit stability cannot be solved in a standard way. Here we propose the following approach. Because in practise the differences x(t)-x(t -T) and x(t) -x(t-T) are small we express them by introducing the small parameter c, which allows us then to apply the KBM method formally and next to take c = 1 [27d]. We assume damping of the same order as c and from (2.9.1), we obtain
x + Q2 X = Po coswt + cA [x(t) -
x(t - T)]
+cB [(1 - ~) x(t) - x(t - T)] .
(2.9.2)
Po 2 cos wt, -w
(2.9.3)
Introducing
x = z+
a
2
2.9 Stability Control of Vibro-Impact Periodic Orbit
103
we get Z + a 2z
= ef(a, 7], 1/J),
(2.9.4)
where ef(a, 7], t/J) = eA [z
-
+ 0: 2 Po 2 coswt -w
a2~w2COSW(t-T)] HB[(l- ~)
.(z 7] =
wt,
z(t - T)
a
[ow 2 sinwt) - i(t - T) + -w
a
[ow 2 sinw(t - T)], -w
't/J = at.
Using the KBM method we have truncated the e series up to order O(e) and we have 0 btained da 1 Aa . Ba dt = "2(B - c)a + 2a slnaT - """'2 coso:T,
dt/J A AI. = 0: - - + -cosaT+ -Bsmo:T. (2.9.5) dt 20: 20: 2 For A = B = 0 we get the uncontrolled solution, which supperts the validity of our approach. Therefore, we analyse the following equivalent solution -
x =
a
2 Po 2 cos(wt - w
+ rp) + eRt(G cos o:ot + D sin aot),
(2.9.6)
where: G = s-Pocosrp/J(0:~-w2)2+c2w2, D = (G/sin2/3'x)(e13c cos 2/3'x) , ,X2 = 0:2 - w 2, /3 = rrk/w, cosrp = (J(0:2 -w 2)2 +c2w 2/PO)[s(R r + l)x_ sin 2/3,X/'x(2 cos 2/3,X - e- 13c - e13C )]. After integration of (2.9.5), we get a(t) = Goe Rt , t/J(t) = aot + ()o, 1
A
B
= "2(B - c) + 2a sin aT - 2" coso:T, A A 1 ao = a - 2a + 2a cos aT + "2 B sin aT,
R
(2.9.7)
and according to (2.9.7) and (2.9.6) one obtains G
= Gocos()o,
D
= -Gosin()o,
Go
= JG2 + D2.
2.9.3 Stability Control From (2.9.6) it is seen that when R < 0 the assumed solution is stabilised more quickly in comparison to the case of R = O. However, the problem of the stability investigation of the vibro-impact state is much more subtle.
104
2. Discrete Systems
Before impact number l, the mass possesses the velocity Xl-. This causes the following perturbation solution to occur
+ bXl = e RTI
x
[(C + bC,) cosaO'Tl
+F COS(W'Tl
+ cP + bcpl),
+ (D + bD,) sinao'T,] F
=
Po 2' a -w
(2.9.8)
2
A new time 'T is measured from the l-th ilmpact 'Tl = 'T + b'Tl. For example, the next impact occurs for 'Tl+l = 21r /w + b7" where b7i denotes the perturbation period T = 21r /w. After some calculations we get
+ bC, cos ao'T + Do:obn cos ao'T +bD, sin ao'T + IM'TlC cos ao'T + R8'TlD sinao'T) -Fbcpl sin(w'T + cp) - Fwb'Tl sin(w'T + cp), e RT [2Ra ob'Tl (D cos 0:0'T - C sin aoT) + mC, cos 0:0'T
bXl = e RT (-Ca:ob'Tl sin ao'T
bIl =
+IMD, sin 0:0'T + R 2 b'TlC cos 0:0'T + R 2 b'TlD sin 0:0'T -a5b'Tl (CcosO:O'T + Dsinao'T) - bC,o:osino:o'T + bD,o:O cosao'T]- Fwbcpl cos(W'T + cp) - Fw 2 b'Tl cos(W'T
+ cp).
The following boundary conditions are introduced l: l
'T
+ 1:
= 0, 'T
=
bTl = 0, bXl = 0, bIl = bXl+, 21rk - + b'Tl, b'Tl = b7;, bXl = 0, w
(2.9.9) bXl
= bI(l+I)-'
After some calculations we obtain the loequations bC, - Fbcpl sin cP = 0,
e 2f3R
(2.9.10)
. 2,8ao + w {Cv C l cos 2,80:0 + bD, sm 1 (6cpl+1I. -
bcpl)
. [(ao D + RC) cos 2,6<>0 + (RD - aoC) sin 2,6aol } - 6CI +1 = 0, R r e 2f3R { 6C, (Rcos 2{3ao - 0:0 sin 2,80:0) + bD, (Rsin 2,8ao 1 + ao cos 2,80:0) + - (bcpl+lI. - bcpl) [(R 2 C - 0:5C w
+ 2Ro:oD)cos2{3ao + (R 2 D - 0:5D - 2RaoC) sin2,8ao] } +R8C +1 + ao bD,+lI. - (R r + 1) Fwbcpl+l coscp = 0, ' where ,8 = 1rk/w and Rr < 1 as usual denotes the restitution coefficient. Assumming that l
bCPl = bcpo
+
E wbT
i ,
i=1
(2.9.11 )
2.9 Stability Control of Vibro-Impact Periodic Orbit
105
and introducing ~D, = a21',
bG, = a1 1',
b'{J, = a31'
(2.9.12)
we get the following characteristic equation
b2 1 2 + bll. 1 + bo = 0,
(2.9.13)
where
b2 = aD { F sin", -
+ (RD -
~ e2J'R [(aD D + RC) cos 213ao
aoG) sin 2}3a:o] },
bl = e 2J'R { ~ "'" [( RC + aDD)
+ (RD -
(cos 213a o - R,e2J'R)
aoG) sin 2}3ao] - (R r + I) Fw cos '{J sin 2}3a:o
+F sin", I( R, bo = R,aoe4J'R
1) aD cos 213ao + (R, + 1) Rsin 213ao I },
[~ (Re + aDD) -
(2.9.14)
F sin "'] .
Note that G(s) could be obtained using a similar approach but without the perturbations. Therefore, the problem of stability is reduced to consideration of the second-order characteristic equation (2.9.12). If the roots of (2.9.13) are 111,21 < 1, then according to the assummed solutions (2.9.12) bG" bD, and b'{J, approach zero for l ---+ 00, and the solutions will be asymptotically stable. We can easily estimate the stability regions, which are defined by the following inequalities
b2 < 1 bo
and
b1
bo + b2
<1.
(2.9.15)
Taking into account (2.9.14) it is now easy to find parameters of the system (or a delay loop) which fulfil inequalities (2.9.15). Additionally, for mechanical reasons, we have x( t} < s. 2.9.4 Simulation Results During numerical simulations we used the MATLAB-Simulink package and the MATLAB-model for (2.9.1) with the boundary conditions (Fig. 2.17). We have taken the following parameters: m = l[kg], k 1 = 3[N/m], k2 = I[N/m], C1 = 0 [Ns/m], Xo = l[m], T = O.47l'[s], I4 = 0.65, s = O.OI[m], = -O.OI[N/m], b, = -0.045[Ns/m]. For these parameters according to (2.9.7) we get R = -0.04, which shows that the delay loop control coefficients A and B allow us to obtain quicker
a,
106
2. Discrete Systems
Pocos(wt} Object
Delay x
o +....- - - - - - - - - 1 L:J-_........_~
Clock
"'-~~"-_--I Delay x'
Fig. 2.11. MATLAB-Simulink model of the vibro-impact system presented in Fig. 2.16
damping of free oscillations in the solution (2.9.6) than without control (Fig. 2.18). Additionally, for the given parameters we have found from Eq. (2.9.13) that 1"Y1,21 lie closer to the origin for the system with the control coefficients than without control (with the delay loop; 1"Y1,21 = 0.62, whereas without the loop 1'1'1,21 = 0.65). " For the given parameters numerical simulations confirm the analytical predictions. It can be seen in Fig. 2.18 that with control the transients vanish more quickly than in the case without control. In the case presented above the periodic orbit is achieved after about 24.1 seconds for the system analysed without the delay loop and after 21.8 seconds for the system analysed with the delay loop (IJ.LI < 10- 4 ), respectively.
2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom 2.10.1 Definition The normal mode of a finite-dimensional system behaves like a conservative system having a single degree of freedom. In this case all position coordinates can be well defined by
2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom
107
3,-------.---------r-----r------r-----, x
10
i
4
~-.l . .~- . ---.\, ,
2
,I
, I
\
I
\
I
\
I
1
\
- - - - ,\ - -
,
---1
- ---
--
1 1 1
.--.
h
".
.~-~---I-~~
_/ without control ...,,
~
1.---.. . .
0
I
~
--
--+-11---'"'"'=1:=-1
'-'
H I
-
with control
.--. -1
----~----~-l-
~
'-'
H
-2
-4 L...20
--'-
-'-
24
22
J...-
26
----'-
28
--'
t[sec} 30
Fig. 2.18. Difference between two transients x(t) - x(t - T) approaching a periodic orbit for the system (al = -0.01, bt = -0.045) Xi
= Pi(X),
X -
Xl,
i
= 2,3, ... , n,
(2.10.1)
Pi (x)
being analytical functions. Rosenberg [135] gets credit for being the first to introduce broad classes of essentially nonlinear conservative systems allowing normal vibrations with rectilinear trajectories in a configurational space (2.10.2)
"A phenomenon like 'vibration in normal modes' exists in nonlinear systems; in a purely verbal manner, this motion may be described as vibration in which all masses move periodically "in unison", among all the properties of normal vibrations we consider asYmptotic the basic one that the normal coordinates decouple the equation of motion" [135]. For instance, homogenous systems whose potential is an even homogenous function of the variables refer to such a class. It is interesting to note that the number of modes of normal vibrations in the nonlinear case can exceed the number of degrees of freedom. This remarkable property has no analogy in the linear (non-degenerate) case. Naturally this definition may now be called "naive", and the problem of a more accurate definition was studied by many authors [55d]. But it is very clear from the intuitive point of view and we accept Rosenberg's definition form the asymptotic standpoint in our further investigations.
2. Discrete Systems
108
In systems of a more general type, trajectories of normal vibrations are curvilinear. Lyapunov [54d] showed that solutions of this kind exist in nonlinear finite-dimensional systems with an analytical first integral, which are close to generating linear systems.
2.10.2 Free Oscillations and Close Natural Frequencies! For systems with two degrees of freedom which have quadratic nonlinearities, the internal resonance at the frequency ratio 1:2 has been studied, along with the internal resonance for systems with cubic nonlinearities and the frequency ratio 1:3 [119]. In recent years attention has turned to mode interactions (of an internal resonanse type) for close natural frequencies. Experimantal observations and solutions of particular problems show that this effect is relevant to the description of oscillatory processes in suspension bridges [1], cylindrical shells and other constructions [41-52,54-77, 28d-35d]. However, the literature does not contain any general analysis of mode interactions of free oscillations in nonlinear systems with close natural frequencies. In particular, we do not know what types of oscillation modulation are possible, what determines the degree and period of energy exchange in a system, what is the number of steady-state regimes (without modulation), which of them are stable, etc. Consider a nonlinear oscillatory system (initially, in general, with damping), described by the equations
.. Uk
+ 2J.L.Uk . + W.2 Uk = bkkUk3 + b12UIkU~, 3-k
k =, 1 2.
(2 .10.3 )
The frequencies WI and W2 are assumed to be close, and the damping factors for the two modes are taken to be the same. Equations (2.10.3) represent the general case of a system with symmetric potentials (when J.L. = 0) that include terms of the second and fourth degree. They are similar to the Duffing equation for systems with one degree of freedom and describe a broad class of mechanical systems (in general, no restrictions are imposed on the coefficients bij ). In accordance with the method of multiple scales we introduce the "fast" and "slow" times To = t, TI = eTo, T 2 = e2TO (the time TI will not be necessary below). We will seek a solution of system (2.10.3) in .the form of an expansion Uk = eUkl(To, TI, T 2, ... )
+ e3uk3(To, T 1, T 2, ...) +...
(2.10.4)
(terms of order e 2 vanish for a system with cubic nonlinearities). The smallness of J.L. and the difference in frequency are introduced through the conditions (2.10.5) 1
By courtesy Ye.V. Ladygina, A.I. Manevitch [1061
2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom
109
Taking into account that
d 2 a a D a Cit = Do + eD l + e D 2 + ... ; Do = 8T ' Dl = aT ' 2 = aT ' l 2 o 2 d 2 2 2 dt 2 = Do + 2eDoDl + e ( 2DoD2 + D l ) + ... , we obtain the following systems of equations for the two approximations
D5ukl + w2Ukl D5uk3 + w2Uk3
= =
0,
(2.10.6)
-2D o(D 2Ukl + J.Lukd + bkkU~l +b12utlu~lk - c52kO'ukl,
(2.10.7)
where c5ij is the Kronecker delta. The solution of system (2.10.6) is written in the form
Ulk = Ak(T2) exp(iwTo) +
Ak(T2) exp( -iwTo),
(2.10.8)
where the bars denote complex conjugats. Substituting (2.10.8) into system (2.10.7), from the condition that there are no secular terms in the resulting equations, we have ,
2 -
-
-2iw(A k + J.LA k ) + 3bkk A kA k + 2b1 2Ak A 3-k A 3-k +b12AkA~_k - c5 2k O'A k = 0,
(2.10.9)
where the prime denotes differentiation with respect to T 2 . Putting the complex amplitude in the exponential form A k = 1/2ak exp(iOk) (k = 1,2), we separate (2.10.9) into real and imaginary parts and obtain a system of equations governing the amplitude and phase modulation of both modes:
(a%)' + 2J.La% = (-I)kb12(4w)-la~a~sin2')', 8w8~ = -3bkka~ - b12a~_k(2 + cos 2')') + 4c5 2k O',
(2.10.10) ')' = 82 - 81. (2.10.11)
Having eliminated sin 2')', we obtain from (2.10.10) the integral (2.10.12) a~ +a~ = Eexp(-2J.LT2) = Eexp(-2e 2J.Lt), where the arbitrary constant E is proportional to the energy of the system (in the first approximation). In particular, for a conservative system (J.L = 0) a~ + a~
= E.
(2.10.13)
From (2.10.11) we obtain the equation for the phase difference ')': Bw')" = (3bl1 -2b12)a~ + (2b12 -3b22)a~ +b 12 (a~ -a~) cos 2')' +40'.(2.10.14)
For further analysis it is convenient [1] to introduce the new variable = a?/ E (0 < < 1). Then, from (2.10.11) with k = 1 and (2.10.14) we obtain a system of equations governing the amplitude-frequency modulation in variables:
e
e
e, ')'
110
2. Discrete Systems ~' = -2J1.e ')" =
+ r oe(1- e) sin 2')',
er1 + r2
1
+ r o( 2 -
(2.10.15 )
e) cos 2')'.
Here
ro = r2
=
b12E
r _ (3bll - 4b12
4w'
8w
1-
(2b12 - 3b22)E 8w
+ 40-
+ 3b 22 )E '
(2.10.16)
.
Without loss of generality we assume ro :/= 0, because otherwise b12 = 0 and system (2.10.3) decomposes into two decoupled equations. We will perform the further analysis for the case of a conservative system (J1. = 0). Dividing the second equation of (2.10.15) by the first one, we obtain d')' e (rl - r o cos 2')') + r2 + 1/2ro cos 2')' (2.10.17) de r oe(1 - e) sin 2')' The solution of this ordinary differential equation is roe(1 - e) cos 2')' + r 1e 2 + 2r2e = C, (2.10.18) where C is a constant ofintegration and determines the trajectory in the (e, ')') plane of the "amplitude-phase portrait" of the system (the AP-portrait). Eliminating')' from the first equation of (2.10.15) and (2.10.18), we obtain 1 (de) r~ dT 2 = F12(e) - F22(e),
2
(2.10.19)
2 rle +1,2r2e - c. ro The form of this equation is identical with that derived in [1] for the case when W2 = 3wl, but the functions Fl(e) and F2(e) are of a different form. The condition Fl (e)
= e(1 -
e),
F2(e)
=
(2.10.20) for the solution of (2.10.19) to exist means that solutions correspond to the parabolic segments F2(e) inside the domain bounded by the parabolic arcs ±F1(e) over the interval [0,1] (Fig. 2.19). The points of intersection of the parabolas F2(e) with the axis e, as can be seen from (2.10.18), correspond to the condition cos 2')' = 0, Le. ')' = ±(2n+ 1}rr/4 (n = 0,1,2, ... ), or the values ~ = 0 and e = 1. The points of intersection of the parabolas F2(e) and F 1 (e) correspond to extremal values of the functions e(T2 ) and ak(T2 ), respectively (k = 1,2). It follows from (2.10.15) (for J1. = 0) that at these points sin 2')' = 0 (if e:/= 0 and e:/= 1), Le. ')' = ±n1r/2 (n = 0,1,2, ...), and that the points on the lower curve (Fl < 0) correspond to the even values of n and those on the upper curve to the odd values. Consequently, the minimum and maximum values of e determining the amplitude modulations and the degree of energy exchange between the modes are equal to the roots of the equations F 1(e) = ±F2(e), Le. the equations
2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom
111
(2.10.21 ) where the upper and lower signs correspond to the points of intersection of the parabola F 2 (e) with the upper and lower curves ±Fl (e), respectively, while the value of C is governed by the initial values eo and /0. 0.4 .,----,......----., ~~
F1
A
!
,
.-
,~,.,
"
,
.~-.
,
~--r-------,
3\.
o ~._-~_. ~j"..-:~._. ~
,
0.4
0.4 .,...---'.,•.1-----, Fl
o
~._-j------
-~/~.
.. k"·~"
2"><~--~ F 1
-0.4 +--.--...-------1 o 0.5 ~ 1
-0.4 - f - - - - t - - - - - - i 0 0.5 1
a)
e
1/'
-0.4 +--"'---1------1 o 0.5 e 1
b)
c)
Fig. 2.19a·c. Graphical representation of the oscillatory regime
The solutions of (2.10.21) and the construction of a "characteristic graph" (Fig. 2.19) give a graphical representation of the oscillatory regime. There are two basic ways in which the curves Fl(e) and ±F2(e) can intersect in a "coarse" system, corresponding to the two basic oscillatory regimes: 1. both the points of intersection lie on the same parabola +Fl (e) or - Fl (e) (curves 1 and 2 in Fig. 2.19a); 2. the parabola F 2(e) intersects both the parabolas +Fl(e) and -Fl(e) in the interval [0, 1] (curve 3). In the first case, the phase difference / oscillates about the value / = ±mr/2. Synchronization of the oscillations proceeds "on average" over the modulation period: at the times when the extrema al and a2 are achieved the oscillations on both degrees of freedom proceed either in phase or in antiphase, if both the points of intersection lie on the lower branch, or the phase difference in these instants is equal to rr/2, (3rr /2) if both the points lie on the F l > 0 branch. In the second case the phase difference increases monotonically, running through the sequential values nrr /2 (n = 0, 1, ... ) at the extremal times. These two types of oscillatory regime with oscillating and monotonically increasing phase differences will respectively be called modulations of the first and second type. The solution of (2.10.19) has the form
± Irol J( ~
1
Fl2 (e) - F22 (e)) -1/2 de
= T2 -
T20,
eo = e(T20).
(2.10.22)
~o
Suppose el,'" ,e4 are the roots of the fourth-degree polynomial Fr(e)arranged in increasing order, with 6 and 6 lying inside the domain
Fi (e),
112
2. Discrete Systems
bounded by the parabolas ±Fl(e). The modulation semiperiod (for the oscillating phase case) corresponds to e varying over the interval (6, e3), and so the modulation period is equal to {3
T· = IFJ _ 2Frll/2
f ((e - 6)(e - 6)(e - 6)(e -
e4))-l/2 de· (2.10.23)
{2
For "non-coarse" systems one must consider singular cases for the position of the F 2 (e) curve (Figs 2.19b and c): the passage of F 2 (e) through the points e = 0 or e = 1 (Fig. 2.19b), and "external" touching of the parabolas F l (e) and F 2 (e) (corresponding to lines 1 and 2 in Fig. 2.19c). In these cases two of the roots ej coincide: in the first case 6 = 6 = 0 or 6 = e4 = 1, and in cases 2 and 3 6 = 6. Because the improper integral in (2.10.23) diverges when two of the roots ej coincide, the modulation period tends to infinity as these regimes are approached. These are "boundary" regimes separating the modulations of the two types distinguished above (lines 1 and 3) and associated with separatrices in the plane (e, 1), or regimes of stationary oscillations without modulation (curve 2). We remark that the" aperiodic" oscillations described in [ld] correspond to these boundary regimes. Consider the possible AP-portraits in the plane (e,1) which are given by integral (2.10.18) and which graphically describe the oscillatory modes of the system. Stationary points corresponding to oscillations with no modulation are found using (2.10.15) from the system of equations
.
I \
eo - e) sin 21 = 0, eFl
+ F2 + F2(1/2 -
e) cos 21 = 0,
which can have the solutions 2F2 = 0, cos 21 = - - , Fo
e
e= 1
1,
cos 21 =
= ± ~1r (n =
(2.10.25)
2(F~ + F2) , Fo
0, 1,2, ...),
(2.10.24)
e= e: = ±~/~ ~oF2.
(2.10.26)
(2.10.27)
These solutions exist when the following conditions are satisfied: (1) 12F21 < IFol, (3) 0 ~ e-:- < 1,
(2) 21Fl + F21 < IFol, (4) 0 < e: ~ 1.
(2.10.28)
Using the periodicity with respect to 1, we confine ourselves to the plane rectangle (0 < e < 1; 0 < 1 < 1r). The stationary points (2.10.25)-(2.10.27) can be positioned on the boundary lines of this rectangle and on the mid-line 1 = 1r /2 (with not more than one point on aline). It is easiest to investigate the nature of a stationary point with the help of (2.10.18), considering the
2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom
113
form of the integral curves in the neighbourhood of the stationary point. The stationary points at ( = 0 and = 1 are saddle points and therefore unstable. From this it follows that the presence of a second degree of freedom makes oscillations along the first generalized coordinate unstable if the stationary points (2.10.25) or (2.10.26) exist. In the neighbourhoods of the stationary points on the lines 1 = ±n1l" /2 the trajectories can be of either elliptic or hyperbolic type, and consequently these stationary points can be stable or unstable. The stability conditions for odd and even n, respectively, have the forms
e
(5) Fo(Fo + FI) > 0,
(6) Fo(Fo - FI) < O.
(2.10.29)
On the characteristic graph (Fig. 2.19c) the "externally" touching hyperbolas (curve 1) correspond to the stable stationary points and the "internally" touching ones (curve 2) to the unstable points. The stationary points on the lines 1 = ±n1l" /2 correspond to synchronous single-frequency modes, Le. normal oscillations of the nonlinear system. It follows from (2.10.3) and (2.10.8) that the points on the lines 1 = 0 and 1 = 11" correspond in the (U1, U2) configuration space to the two straight lines U2 = ±hU1' where
h
=
/ a2 = (1_e;)1/2 = (Fo/2 - F1 - F2)1 2 a1 Fo/2 + F2
e;
The stationary points on the lines 1 = 11"/2 and 311" /2 correspond to the ellipses
u2
1
et
+
u2 2 1- et
= Eg2,
e
e
and the points on the lines = 0 and = 1 correspond to straight lines along the axes OU2 and Ou 1. The separatrices pass through the possible unstable stationary points. For separatrices passing through the "left" points (2.10.25) one should put C = 0 in (2.10.18). We obtain equations for two branches: (2) cos 21 = -
eF1 + 2F2 no(1- ),
e
(2.10.30)
which exist when condition 1 of (2.10.28) is satisfied. The "right" separatrix, passing through the stationary points (2.10.26), exists when condition 2 is satisfied. The equations of the branches of this separatrix are obtained from (2.10.18) with C = F 1 + 2F2 (1)
e=
1,
(2) cos 21 =
(e + 1~~ + 2F2 .
(2.10.31)
The central separatrix (CS) passing through the stationary points (2.10.27) for odd (or even) n exists when condition 3 (condition 4) is satisfied and condition 5 (condition 6) is violated. Substituting the coordinates of points
114
2. Discrete Systems
(2.10.27) into (2.10.18), we obtain C = (-F2 ± Fo/2)2 /C~Fo - Fd and the equations for the branches of the central separatrix
e= B± VB2_D, Fo cos 21' + F2 - Fo cos 21' - Fl'
B -
D -_
(F2 =F Fo/2)2 . (Fl ± Fo)(Fl - Fo cos 21')
(2.10.32)
The stationary points and separatrices possess the following properties. 1. If the "left" stationary points (2.10.25) exist (i.e. condition 1 is satisfied), then in the rectangle (0 < ~ < 1, 0 ~ l' < 'IT) there is at least one "intermediate" stationary point (2.10.27) on the line l' = 1r /2 or l' = 0, and this point
is stable. Indeed, when condition 1 is satisfied the sign of the numerators in condition 3, 4 is given by the sign of their first term, and for their moduli we have I ± Fo/2 - F21 < IFol. If the signs of Fl and Fo are the same, then the sign of the denominator in condition 3 is the same as the sign of the numerator, and because we then have IFl + Fol > IFol, condition 3 is satisfied and, clearly, condition 5. In the case of opposite signs for Fl and Fo, the signs of the numerator and denominator in condition 4 are the same and IFl - Fal > IFo!' so that conditions 4 and 6 are satisfied. A similar assertion holds for the "right" stationary points (2.10.26).
2. If one stationary unstable point (2.10.27) exists on the line l' = 1r /2 (or l' = 0), then a stable stationary point exists on the line l' = 0 (or l' = 1r /2); and there are no separatrices (2.10.30) ati.d (2.10.31). Suppose condition 3 is satisfied and condition 5 is not satisfied (Le. the stationary point at l' = 1r /2 is unstable). Then Fo and Fl have opposite signs and IFll > IFol. It follows from condition 3 (because the sign of the denominator is governed by the sign of F 1 and is opposite to the sign of Fo) that the signs of Fo and F2 are the same and IF21 > IFl l/2. Condition 1 is therefore violated. Considering the case F l > 0 and F l < 0 separately, and taking into account that the sign of F 1 is opposite to the signs of Fo and F2 and that IFll > IFol, we find that in both cases condition 2 is violated, and the (right) inequality in condition 4 is also violated, which proves the assertion. These properties enable us to describe the various possible AP-portraits in the 'plane (e,1'). Each side separatrix (SS) joins two unstable stationary points at = 0 or ~ = 1. The branches of these separatrices surround a single stable stationary point at l' = 1r /2 or l' = 0 (0 < e< 1). One can verify that if, for example, between the "left" separatrices there is a point on the line l' = 0, then the abscissa of the point of intersection of the separatrix with the line l' = 0 is twice the abscissa of the stationary point it is obvious that ~ < 1/2. A similar property is satisfied by the right separatrix: here it is necessary for the stationary point surrounded by its branches to be in
e
e;
2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom
115
e
the right half of the rectangle. The separatrix originating from = 0 cannot intersect the line ~ = 1, and conversely. The branches of the CS join the two unstable stationary points (2.10.27), corresponding to even or odd values of n, and surrounding the stable stationary points. The CS cannot intersect the lines = 0 or = 1. Inside the domains surrounded by the SS or CS a modulation regime of the first type exists, and outside these domains, there is a regime of the second type. Thus, four qualitatively different types of the AP-portrait, governed by conditions 1-6, are possible and they are shown in Fig. 2.20
e
1 0
a)
b)
e
1 0
c)
1
d)
Fig. 2.20a-d. Four qualitatively different types of the AP-portrait
1. Conditions 1 and 2 are satisfied. There are stable stationary points at "y = n1l" /2 (2.10.27) for even and odd n, in the left section (e < 1/2) and right section (e > 1/2) of the rectangle, Le. three stable normal modes exist (and two trivial unstable ones Uk = 0, k = 1,2). Each of the stationary points is surrounded by the corresponding SS; there are no CS (Fig. 2.20a). 2. Only one of conditions 1 and 2 is satisfied. There is a stable stationary point (2.10.27) only for an odd or even n, and only one SS (on the left if condition 1 is satisfied, and on the right if condition 2 is satisfied); there are no CS (Fig. 2.20b). All the normal modes, apart from the single Uk = 0, k = 1 or k = 2 mode, exist and stable modes also exist that are either rectilinear (if condition 4 is satisfied), or elliptic (when condition 3 is satisfied). 3. Neither condition 1 nor 2 is satisfied, but condition 3 is satisfied. The stable and unstable stationary points (2.10.27) alternate (with the point for odd n being stable if condition 5 is satisfied). There is a CS, but no SS (Fig. 2.20c). Three normal modes exist, where either the rectilinear one is stable (when conditior.. 6 is satisfied), or the elliptic one is stable (when condition 5 holds). 4. Conditions 1-3 are not satisfied. There are no stationary points (normal modes) or separatrices. All oscillatory modes are of modulation type 2, with the modulation being relatively small if compared with cases 1-3 (Fig. 2.20d).
116
2. Discrete Systems
In cases 1-3 one can distinguish subcases. In. case 1 there are two subcases distinguished by the position of the left stationary point: on the line '"Y = 0 or on '"Y = 11"/2. Similarly, in case 3 the stationary point can be stable at '"Y = 0 or at '"Y = 11"/2. Four subcases are possible for case 2: a left or right separatrix, and a stationary point at '"Y = 0 or '"Y = 11"/2. The corresponding AP-portraits can be obtained from those shown in Fig. 2.20. We introduce the parameters
°
bu
b22 40' 02 = b12' 0' = b 2E ' 1 Then, conditions 1-6 can be represented in the form 01
= b12'
(1) (2) (3) (4) (5) (6)
302 - 3 < 0'0 - 301 + 1 < 0'0 - 301 + 1 < 0'0 30 2 - 1 < 0'0 302 - 3 < 0'0 - 301 + 3 < 0'0 01 + 02 > ~, 01+02<2.
< 302 ~ -301 < 302 < -301 < -301 ~ 302 -
1, + 3, 1 + 1 +3 3
for for for for
01 01 01 01
+ + + +
02 > ~, 02 < 3' 02 < 2, 02 > 2,
(2.10.33)
(2.10.34)
Unlike 0' and E, the dimensionless frequency detuning parameter 0'0 does not depend on the choice of c and can be written in the following form 0'
°
-
40'. --"....------,,,..--
- bI2(U~(0)
+ u~(O))'
(0'.
= c 20' = w22 -
2)
WI .
(2.10.35)
As can be seen from (2.10.34), the ty~e of AP-portrait is determined by the relative positions of the points Cl = 302 - 3, C2 = 302 - 1, d1 = -301 + 1, d2 = -301 + 3
(2.10.36)
and the quantity 0'0. Four possible positions of the intervals (Cl' C2) and (dl,d 2) are shown in Fig. 2.21 (C2 < d 1,Cl < d 1 < C2,Cl < d 2 < c2,d2 < C2)' The type of the AP-portrait (easily determined from (2.10.34)) is shown above the intervals. In case (a) the interval (C2' dd contains the stable stationary point at '"Y = 0 (11"), Le. the rectilinear normal mode is stable, and the unstable one is at '"Y = 11"/2 (311"/2) (i.e. elliptic). In case (d) these points (and normal oscillations) "exchange" stability. Figure 2.21 graphically demonstrates the influence of the parameter 0'0 on the'system behaviour. If 0'0 lies in the interval
01 <
0'0
< 02,
01 = min(cl' dd,
02 = max(C2' d 2),
(2.10.37)
then we have the AP-portraits of types 1-3 with stationary points and pronounced modulation of the amplitude and phase (energy exchange). If 0'0 lies outside this interval, the AP-portait of type 4 is indicated with relatively small modulation. Thus, condition (2.10.37) allows one to specify the smallness of the frequency detuning parameter. The minimum width of interval (2.10.37) is 2. The centre of the interval is the point
2.10 Normal Modes of Nonlinear Systems with n Degrees of Freedom
a)
ci
2°
3°
2°
4°
~
1°
2°
'"'"
d)
V~
~
3°
2° ci
~~M ~~
~~~
~
dl
2°
4°
~
ci
dl
Fig. 2.21a-d. Four possible positions of the intervals
4° ~
ttW#A
~
(Cl l C2)
and (d l , d 2 )
3 b22 - bll
3
a. = 2(a 2
~
dl
4°
2°
4°
2°
W/~
~ "'~~
dl 4°
1°
2°
4° ci
W§Aa
c)
b)
4°
117
-al)
=2
b12
.
In the case when b u = b22 we have al = a2: a. = 0, Le. interval (2.10.37) is symmetric in relation to the origin. If al 1:- a2 the interval is displaced relative to the origin and for sufficiently large la2 - all (or Ib22 - buD the point crO = 0 can turn out to lie outside the interval. One must also take into account that the sign of crO is governed by the sign of b12 (one can always put cr. > 0, Le. W2 > WI)' The signconstancy condition on crO indicates either the positive or negative part of interval (2.10.37) (if it exists). Two conclusions follow from this: 1. it is not necessary for the large modulation to correspond to the smaller value of cro: combinations of the coefficients bij are possible with the AP-portraits of types 1-3 for intervals with crO far from the point OJ 2. for certain combinations of bij only the type 4 AP-portraits are possible, irrespective of the energy and frequency separation. In the above analysis there is a natural separation of the influence of the oscillation energy and the ratio of the initial amplitudes of the two modes on the energy exchange. The quantity E acts on crO according to (2.10.35) (increasing E is equivalent to decreasing cr), and together with cr. it therefore determines the type of AP-portrait. The initial amplitude ratio ~o determines the phase trajectory in a given AP-portrait. Consider the special case when bu = b22 = 0, b1 2 1:- O. Then al = a2 = 0, CI = -3, C2 = -1, d l = 1, d 2 = 3, Le. we have Fig. 2.21, case (a). Condition 5 is not satisfied, while condition 6 is satisfied. For -3 < crO < -I' we have the type 2 AP-portrait with a left separatrix and stable stationary point at 1 = 0 (1r), Le. with rectilinear normal oscillations. For -1 < crO < 1 the AP-portrait is of type 3 and has a stable stationary point at 1 = 0 (1r) and an unstable one at 1 = 1r /2 (31r /2), Le. with stable rectilinear normal modes and an unstable elliptic mode. When 1 < crO < 3 the AP-portrait is of type 2 with a right separatrix and stable rectilinear normal modes. Finally, for crO < -3 and crO > 3 the AP-portrait is of type 4.
118
2. Discrete Systems
In conclusion we note that the numerical integrations of (2.10.3) performed for the purpose of estimating the accurancy of the solution obtained by the multiple scales method demonstrated almost complete agreement between the analytic and numerical solutions in all the cases considered with an arbitrary choice of e ~ 0.1 and amplitudes of up to 0.5 (the error in determining the amplitude was of the order of 0.1%). But when e was increased beyond 0.1, the error increased rapidly. For example, when e = 0.15 the error in the amplitude computation reached 30%.
2.11 Nontraditional Asymptotic Approaches 2.11.1 Choice of Asymptotic Expansion Parameters Introducing a small parameter into the nonlinear problems is a very delicate and nontrivial matter. We consider in this section an elementary illustrative problem: finding the roots of a fifth-degree polynomial [49, 10d]. We are concerned here with finding the real root Xo of the polynomial equation
+x =
x5
1.
(2.11.1)
We have chosen the degree of this polynomial to be 5 because it is high enough to be sure that there is no quadrature formula for the roots. However, one can be sure that there is a unique real r09t Xo and that this root is positive because the function x 5 + x is monotone fucreasing. Using Newton's method we compute Xo = 0.75487767 ...
(2.11.2)
There are several conventional perturbative approaches that we could use to find xo. One such approach, which we will call the weak-coupling perturbation theory, requires that we int[loduce a perturbative parameter e in front of the x 5 term [49]: ex5
+x
= 1.
(2.11.3)
Now, x depends on e and we assume that x(e) has a formal power series expansion in e: '
X ( c) =
ao + ale + a2e 2 + a3e3 +....
(2.11.4)
To find the coefficients we substitute (2.11.4) into (2.11.3) and expand the result in an asymptotic series in powers of e. We ,find that the coefficients an are integers which oscillate in sign and grow rapidly as n increases: ao = 1, al = -1, a2 = 5, a3 = -35, a4 = 285, a5 = -2530, as = 23751,
(2.11.5)
etc. In fact we can find a closed-form expression of an valid for all n,
2.11 Nontraditional Asymptotic Approaches an
= [( -1)n(5n)!] [n!(4n + I)!],
119
(2.11.6)
from which we can determine the radius of convergence R for the series in (2.11.4):
44
R = 55 = 0.08192.
(2.11.7)
Evidently, if we try to use the weak-coupling series in (2.11.4) directly to calculate x(I), we will fail miserably. Indeed, using the seven coefficients in (2.11.5) for e = 1 gives 6
x(l) =
2: an =
21476,
n=O
which is a poor approximation to the true value of x(l) in (2.11.2)! Of course, we can improve the prediction enormously first by coupling the [3/3] Pade approximants and then by evaluating the result at e = 1. Now we obtain the result x(l)
= 0.76369,
(2.11.8)
which differs from the correct answer in (2.11.2) by 1.2%. A second conventional perturbative approach is to use a strong-coupling expansion. Here, we introduce a perturbative parameter e in front of the term x in (2.11.1) [49]:
x 5 + ex = 1.
(2.11.9)
As before, x depends on e and we assume that x(e) has a formal series expansion in powers of e: x(e) = bo + b1e + b2e 2 + b3e 3 + .... (2.11.10) Determining the coefficients of this series is routine and we find that 111 bo = 1, bl = -"5' b2 = - 25' b3 = -125' _ 21 _ 78 b4 = 0, b5 - 15625' b6 - 78125' (2.11.11) etc. Again, we can find a closed-form expression for bn valid for all n,
r (4n-l) :
bn
= - 5r (¥) n!
(2.11.12)
from which we can determine the radius of convergence R of the series in (2.11.10) 5 R = 44 / 5 = 1.64938. .. . (2.11.13) Now, e = 1 lies inside the circle of convergence so it is easy to compute x(l) by summing the series (2.11.10) directly. Using the coefficients listed in (2.11.11), we have
120
2. Discrete Systems 6
x(l)
=L
bn
(2.11.14)
= 0.75434,
n=Q
which differs from the true result in (2.11.2) hy 0.07%, a vast improvement over the weak-coupling approach. Now we use the o-expansion method to find the root XQ. We introduce a small parameter 0 in the exponent of the nonlinear term in (2.11.1) [49],
+x
x 1+6
1,
=
(2.11.15)
and seek an expansion for x(o) as a series of powers of 0:
+ CIO + C202 + C303 + ... .
x(o) = CO
(2.11.16)
The coefficients of this series may be computed easily. The first few are 1 1 1 CO = 2' Cl = 4 1n2 , C2 = -gln2, 1
1
3
1
C3
2 = - 48 In 2 + 32 In 2 + 16 ln 2,
C4
= 32 In 2 - 64 In 2 - 32 In 2,
C5
= 480 In 2 - 768 In 2 - 128 In 2 + 64 In 2 + 64 In 2,
1
3
3
1
7
5
1
5
C6 = -192 In 2
1
2
3
4
35
+ 1536 In
4
3
3
2
1
5 3 5 2 1 2 + 768 In 2 - 128 In 2 - 128 In 2,
etc. The radius of convergence of the 0 series in (2.11.16) is 1. A heuristic argument for this conclusion is as follows. The radius of convergence is determined by the location of the nearest singularity of x(o) in the complex -0 plane. To find this singularity we differentiate (2.11.15) with respect to 0 and solve the resulting equation for x'(o): x 1+6 1n x
x'(o)
=
-1
+ x 6 (1 + 0)"
Since x( 0) is singular where its derivative ceases to exist we look for zeroes of the denominator 1 + x 6 (1
+ 0)
= O.
We solve this equation simultaneously with (2.11.15) to eliminate 0 and obtain a single equation satisfied by x:
o=
x In x
+ (1 -
x) In (1 - x).
The solution to this equation corresponding to the smallest value of 101 is x = O. From (2.11.15) we therefore see that 0 = -1 is the location of the nearest singularity in the complex-o plane. In fact, as 0 decreases below -1, (2.11.15) abruptly ceases to have a real root. This abrupt transition accounts for the singularity in the function x(o).
2.11 Nontraditional Asymptotic Approaches
121
Clearly, to compute Xo it is necessary to evaluate series (2.11.16) at fJ = 4. For this large value of fJ we use the coefficients in (2.11.1) and convert the Taylor series to [3/3] Pade approximants. Evaluating the Pade approximant at fJ = 4 gives x(fJ
= 4) = 0.75448,
(2.11.17)
which differs from the exact answer in (2.11.2) by 0.05%. The fJ series continues to provide excellent numerical results as we increase the order of the perturbation theory. If we compute all the coefficients up to C12 and then convert (2.11.16) to a [6/6] Pade approximant, we obtain x( fJ
= 4) = 0.75487654,
(2.11.18)
which differs from XQ in (2.11.2) by 0.00015%. Last but not least, we may introduce a small parameter in our equation in the following way [10d]:
xe
-1
+X
c « 1.
= 1,
After substituting x = ye one obtains y + ye
= 1.
Taking into account the relation c e = 1 + c In c
+ ...
we may represent y in the form y = c
+ o(clnc).
Then, we have
and for c = 1/5, x 3.9%).
~
0.724780 (the error of the first approximation is only
2.11.2 6-Expansions in Nonlinear Mechanics (49] Let us start with the solution of a simple nonlinear differential equation. Consider the nonline,ar ordinary differential equation problem
y'(x) = [y(x)]n, y(O) = 1.
(2.11.19)
The exact solution of this problem is
y(x) = [1 - (n - l)x]-l/(n-l).
(2.11.20)
To solve (2.11.19) approximately using the fJ expansion, we let n = 1 + fJ and solve
y'(x) = [y(x)P+6.
(2.11.21)
122
2. Discrete Systems
To solve (2.11.21) perturbatively, we can seek a solution y(x) in the form of a. series in the powers of 8: (2.11.22) For example, Yo(x) satisfies the linear differential equation problem
Yo = Yo(x),
Yo(O) = 1,
whose solution is x
Yo = e .
Indeed, all functions Yn (x) satisfy linear differential equations which are easy to solve. We find that 1 x x, 2 Y2(X) = e x [1"3x3+ gX 1 4] , Yl(X) = 2'e etc. The reason for using a perturbative approach if that, in general, even when one cannot solve the nonlinear differential equation, the differential equation for the perturbation coefficients Yo(x), Yl(X), Y2(X), ... are always linear and therefore can be solved in quadrature form. . For the particular problem (2.11.21) a closed-form solution exists. Therefore, we can determine the radius of convergence R of series (2.11.22): 1 R=~.
We have computed the series in (2.11.22) up to the 8 10 term. Let us examine the numerical accuracy of the 8 s~ries. The exact value of y(x) at x = 1/4 for the case n = 4 (8 = 3) is y
n)
= 1.587401.
(2.11.23)
Directly summing the 8 series 2:~ 8nYn(1/4) gives 1.284 when n = 0 (19% error), 1.404 when n = 1 (11.5% error), 1.470 when n = 2 (7.4% error), 1.5099 when n = 3 (4.9% error), 1.5626 when n = 6 (1.6% error), and 1.58128 when n = 10 (0.39% error). We can also compute a Pade approximant from the 8 series and then set 8 = 3. The [3/3] Pade approximant gives 1.58692 (0.03% error) and the [5/5] Pade approximant gives 1.587395 (3.7 x 10- 4 % error). It is numerical results such as these that encourage us to use the 8 expansion to solve difficult nonlinear differential equations. Now we turn to a more complicated problem. The classical anharmonic oscillator is defined, by the nonlinear ordinary differential equation d2y dt 2
3
+ Y + ey = O.
(2.11.24)
2.11 Nontraditional Asymptotic Approaches
123
We impose the conventional initial conditons
y(O) = 1,
(2.11.25)
y'(O) = O.
Our objective here will be to find the period of the anharmonic oscillator. It is well known that the initial-value problem (2.11.24) can be solved exactly in terms of elliptic functions and the period T can be expressed exactly as an elliptic integral ,..
J ~
T = 4
dO [1 + ~ (1 + sin 2 0)] -
1
~.
(2.11.26)
o
The integral in (2.11.26) can be expanded as a series in the powers of e 21 ] 3 T = 21T" [ 1 + -e + _ e 2 + ... 8 256
-I
(2.11.27)
One cannot use the conventional perturbation theory to find the period T for small lei. It is true that when lei is small the exact solution y(t) approximates the motion of a harmonic oscillator of period 21T". However, solving the Duffing equation perturbatively requires some subtlety. If we seek a conventional perturbative solution for y(t) as a series in powers of e, we find that there is a resonant coupling between successive orders in the perturbation theory. As a result, the coefficient of e in the pertubation series for y(t) grows linearly with t, the coefficient of e 2 grows quadratically with t, the coefficient of e 3 grows like t 3 , etc. Thus the perturbative solution is only valid for times t which are small compared with 1Ie. At such short times we cannot use the perturbation expansion for y(t) to obtain the series expansion in (2.11.27). More sophisticated perturbative methods have been devised which enable us to calculate y(t) perturbatively for times t "'" lie and thus to obtain the series in (2.11.27). One such method is called multiple-scale perturbation theory. We will attack (2.11.24) using the 6 expansion and will find that here, too, the methods of multiple scale perturbation theory must be used. To use the 6 expansion we replace y3 by yl+26 and consider the differential equation d2 dt 2
+ Y + (w 2 -
1)yl+26 = 0,
y(O) = 1,
y'(O) = O.
(2.11.28)
In (2.11.28) we have found that it is convenient to set e = w2
-
1,
(2.11.29)
so that when 6 = 0, (2.11.28) describes a classical harmonic oscillator whose frequency is w. Note, also, that y26 is to be interpreted as the positive quantity (y2)6. Thus when we expand y26 as a series in powers of 6 we obtain
y26 = 1 + 6In(y2)
+
62
63 2" [In(y2)]2 + "6[ln(y3)]3
+"',
(2.11.30)
124
2. Discrete Systems
in which the argument of the logarithm is always positive and no complex numbers appear. Let us begin by trying to find a conventional perturbative solution to (2.11.28) as a series in powers of 6 00
y(t)
=
L 6nYn(t).
(2.11.31)
n=O
Substituting (2.11.31) into (2.11.28) and using (2.11.30) we obtain a sequence of linear equations and associated initial conditions which must be solved. The first few read as
~~o + w2yO = ~~l
+W
2
2
y~(O) = 0,
Yl = _(w 2 -1)Yoln(Y5),
~;'2 + W 2 Y2 = ddtY3 2
Yo(O) = 1,
0,
+ W 2Y2
1) {Ylln(y~)
_(w 2
-
Y2(0)
= Y2(0) = 0,
(2 ) { Y2 1n (Yo2) - 1
= - W
Yl(O)
(2.11.32)
= y~(O) = 0,
+ 2Yl + ~Yo[ln(y~)]2 } ,
(2.11.34)
y~ + "2Y1 1 [1n (Yo2)]2 + 2Y2 + Yo
2 + "61Yo [ln(yo)] 23} , + 2y 1 ln(yo) Y3(0) =
(2.11.33)
(2.11.35)
y~(O) = O.
The solution to (2.11.32) is
Yo = cos(wt).
(2.11.36)
We can solve (2.11.33) using the method of order. We let
Yl(t) = cos(wt)u - 1(t).
(2.11.37)
The equation satisfied by Ul (t) is d 2 ul cos(wt) dt 2
dUl
-
2w sin (wt) dt = _(w 2 - 1) cos(wt) In [cos2 (wt)], (2.11.38)
which has cos(wt) as its integrating factor d -[cos 2 (wt) dUl] dt dt = -(w 2 - 1) cos 2(wt) In[cos 2 (wt)].
(2.11.39)
Two integrations of (2.11.39) give from (2.11.37)
J
J
o
0
t
Yl (t) = - cos(wt)(w 2-1)
~() cos ws
8
dr cos 2(wr) In [cos 2 (wr)].(2.11.40)
2.11 Nontraditional Asymptotic Approaches
125
The integral with respect to s in (2.11.40) can be performed by interchanging the orders of integration: 2
- cos (wt) (w - 1) Yl = w
f t
2
dr cos2 (wr) In[cos (wr)][tan(wt) - tan(wt)]
o =
cos(wt)
w
2
1 ~ {cos 2 (wt) - 1 -In[cos 2 (wt)] cos 2 (wt)} 2w wt
-Sin(wt)W~~1 f
(2.11.41)
o Note that the integral in (2.11.41) grows linearly with t for large t because the integrand is a positive periodic function. However, we know that the exact solution to (2.11.28) is a bounded function. Hence, (2.11.41) can only be valid for times that are short compared with I/o. This problem appears because each successive order in the perturbation theory is resonantly coupled to the previous orders. To see this, note that (2.11.33) is the differential equation for a driven harmonic oscillator of the natural frequency w. The driving term (the inhomogeneous part of the differential equation) also has the frequency w because it is a functional of Yo. Thus the oscillator described by (2.11.33) is driven on resonance and the solution exhibits secular behaviour (it grows with t). We can still try to use the expression in (2.11.41) to infer a value for the period of the oscillator accurate to order O. We assume that the period T of (2.11.28) is itself a series in powers of 0: 21r a (2.11.42) w w To determine the coefficient a in (2.11.42) we require that after a quarterperiod the amplitude of the oscillator T
= - + -0 + .. '.
y(t) = yo(t)
+ 0Yl (t) = 1 at t = 0 to Y = 0 at t = T /4.
will fall from y Evaluating the expressions for Yo in (2.11.36) and Yl in (2.11.41) we obtain, to order 0,
aD ~ 6W~~
,..
1 ]
a
= -4
w
f i
2
w -1 2
o
dx cos2 x In (cos 2 x)
= 1r 1-w 2 w
2
(1 ~ 2ln 2).
(2.11.43)
126
2. Discrete Systems
Thus, to the leading order in 6, the period of the oscillator is T = -21r [ 1 + 6 w
w
2 -
2
1 (21n 2 - 1)] .
(2.11.44)
2w Let us reexamine the problem in (2.11.28) llsing the methods of multiplescale perturbation theory. We assume that there are two time scales in the problem: a short-time scale described by the variable t and a long-time scale described by the variable 7"
=6t.
We then seek a solution to (2.11.24) of the form
y(t) = Yo(t, 7")
+ 6Yl (t, 7") + ... ,
(2.11.45)
where the initial conditions in (2.11.25) become 8Yo
at (0,0)
Yo(O,O) = 1,
8Yo
Y1 (0, 0) = 0,
87" (0,0)
=
0,
8Yl
+ at
(2.11.46)
(0,0) = 0,
etc. H we substitute (2.11.45) into (2.11.28), we obtain
82 at 2Yo(t, 7")
+ W 2 Yo(t,7")
=
°
(2.11.47)
to the zeroth order in 6 and
82 at 2Yl(t,7")
8 2 Yo 2 + W Y l (t, 7") = -2 at87"
"
+ (1 - w2)yo In(y5)
(2.11.48)
to the first order in 6. The most general real solution to (2.11.47) is Yo(t,7") = A(7")eiwt + A*(7")e- iwt , (2.11.49) where A(7") is to be determined. Using (2.11.49) we can e~aluate the right-hand side of (2.11.48):
(1 _w2)[A(7")eiwt + A*(7")e- iwt ]1n [(2IAI2) (1 + A2e2iwt + A*2 e -2iwt)] 2 21AI 21AI 2 -2[iweiwt A'(7") - iwe- iwt A*'(7").
(2.11.50)
Nov..> we expa~d the logarithm in (2.11.50) to identify all terms proportional to e1wt and e 21wt • Such terms oscillate at the frequency wand thus give rise to the seculiar behavior in Yl . The coefficient of e iwt is -2iwA'(T)
+ (1
- w 2 )A(T) In(2IAI 2 )
1- w ~ 1 4 + 2 A(7") 2k + 1 2
6
k [
2
_
1 -2 w A(T)
2k + 1 ] k+1 .
~ ~4-1 [
2; ]
(2.11.51)
2.11 Nontraditional Asymptotic Approaches
127
Evaluating the sums in (2.11.51) gives
-2iwA'(r)
w 2 )A(r) In(2IAI 2 )
+ (1 -
- (1 - w 2 )A(r) In 2 + (1 - w2 )A(r).
(2.11.52)
Thus, the condition that there is no secular behaviour in Y1(t, r) is that the expression in (2.11.52) (as well as its complex conjugate) vanishes:
- 2iwA'(r)
+ (1 -
w 2 )A(r)[1 -In(IA2 1)]
= O.
(2.11.53)
To solve (2.11.53) we let
A( r) = R( r )ei9 ('T"),
(2.11.54)
substitute (2.11.54) into (2.11.53), and decompose the result into its real and imaginary parts: R'(r) = 0, (J' ( r)
=
w 2 -1
2w
(1
+ 2ln R).
(2.11.55)
Hence, R(r) is a constant R(r) =
and
(J ( r)
Flo,
(2.11.56)
is a linear function of r,
(J(r) =
w 2 -1
2w
(1
+ 2ln Flo)r + (Jo.
(2.11.57)
The initial conditions in (2.11.46) imply that Flo our final result for To(t, r) is
¥Ott, r)
= cos [wt + r
w2 2W
= 1/2 and
1 (1- 21n 2)] .
(Jo
= 0,
thus
(2.11.58)
Finaliy, we eliminate r in favour of 8t to obtain the MSA result TMSA =
w - 6w
2: ~1r, (21n 2 - 1)
(2.11.59)
:1
which we expand to order 6:
TMSA=~ [1+6w~~1(2ln2_1)]
+0(6 2 ).
(2.11.60)
To our surprise, (2.11.60) agrees exactly with the order-6 result we obtained in (2.11.44) using the 6-perturbation method at a quarter-period. It is a long but routine calculation to carry out the 6-perturbation series to order 62 . Using the quarter-period method we find that at 6 = 1,
T =
~ [2'-+ 0.5238 w~~ 1 + 0.6041 (w~~ 1)'] .
(2.11.61)
128
2. Discrete Systems
Table 2.1. Comparison of the exact value of the period of the anharmonic oscillator with the period calculated from the order-8 quarter-period method (same 88 MSA) and the order-8 2 quarter-period method f:
1 3 8
W
J2 2 3
T( exact) 4.76802 3.52114 2.41289
T( order 82 ) 4.73488 3.50794 2.40871
T(order 8) 4.87195 3.59669 2.45397
In Table 2.1 we compare three results: the exact numerical calculation of the period T; the order-8 quarter-period calculation, which is the same as the order-8 MSA result in (2.11.60); and the order-8 2 quarter-period calculation in (2.11.61). We set 8 = 1 and look at three values of c = w 2 -1. As expected, the MSA and order-8 results are excellent, having an accuracy of about 2%. The order-8 2 results are even better, having a relative error of less than 0.5%.
2.11.3 Asymptotic Solutions for Nonlinear Systems with High Degrees of Nonlinearity As an example, one can consider the equation
x.. + xn= 0,
n = 2k
+ 1,
k = 1,2, ... ,
for which we will seek a single-parameter family of periodic solutions which are skew-symmetric with respect to the origin of coordinates in the limit as n
--+ 00.
e
.
Let us introduce the function = xj!A (A is the amplitude) for which the inequality 0 < lei < 1 holds. Note that the function e is continuous and periodic. The initial equation can then be represented as follows:
e.. + An-len = O.
(2.11.62)
en
We will expand the function in a series in 1I n as n do this, we first transform the function
_{en,
0,
--+ 00.
In order to
<e ~ 1 e> 1
0
using the Laplace transformation [152]
= 8(e -
l)(n
+ 1)-1 -
8(e - l)(n
+ l)-I(n + 2)-1 + ... ,
where 8(.) is the delta function. We will now make the change of variable t =
rlw
in (2.11.62).
2.11 Nontraditional Asymptotic Approaches
129
On retaining just the principal term in the sum and putting w 2 = An -
(since 0 < lei
1
j(n
~
+ 1)
(2.11.63)
1), we have the equation
e = -8(eo -
d2 _0 dr 2
1)
(2.11.64)
for determining the periodic function eo. Integration of (2.11.64) taking into account the skew symmetry with respect to the origin of the coordinates yields in the initial variables
(2.11.65)
Xo = Awt.
Expression (2.11.63), which can be treated as an amplitude--frequency dependence, and the solutions over a quarter of the period argee with those obtained by another method in [55d]. Solution (2.11.65) does not satisfy the boundary conditions deo = 0 for r = 1. dr The additional solution is the boundary layer y, and we represent eo as e
= eo +y.
(2.11.66)
Here y« eo. One has from the boundary condition for r = 1 dy _ deo _ 1 dr - - dr - - . This condition leads to the following asymptotical estimation eo dy -d "" ny. n r Substituting (2.11.66) into the governing equation and taking into account the asymptotic estimations, in the first approximation one has 1J "" - ;
2
y w dr 2 2d
en
+ r.,0
= O.
A solution of this equation in the initial variables is -(Aw)nt n+ 2 y= (n+1)(n+2) and coincides with the solution obtained in [55d] by another approach. The approaches proposed in 2.10.2 and 2.10.3 give the possibility of obtaining a solution of the nonlinear differential equation containing the term X1 +6 for 8 --+ 0 and 8 --+ 00. Matching these limiting asymptotics by, for example, two-point Pade approximants leads to the solution for any value of the parameter 8. It is worth noting that the asymptotic approach based on the distributionals now has many applications [57, 76, 161, 34d, 35d].
130
2. Discrete Systems
2.11.4 Square-Well Problem of Quantum Theory The problems of strong interactions and the nonperturbative approach are the main ones in field theory. In particular, solving the SchrOdinger equation for N --+ 00 is very interesting from this point of view. In [54] such a solution was obtained on the basis of 6-expansions in connection with the matching asymptotic procedure. Unfortunately, a substantial number of expansion terms has to be engaged in time-consuming ,calculations. In this work we propose new techniques which give us the possibility constructing expansions in the degree of c = 1IN and to obtain good results using only lower terms of the asymptotic expansions. We consider the equation t/Jxx
+ x 2N t/J - Et/J =
0,
(2.11.67)
accompanied by the boundary conditions (2.11.68)
t/J( ±oo) = O.
We seek an expansion of the eigenvalue E(N) as the series in the power liN for N --+ 00. If we suppose N = 00, we have from (2.11.67), (2.11.68) t/Jxx + Et/J = 0, t/J(±1) = O.
(2.11.69) (2.11. 70)
The eigenvalues of the problem (2.11.69), (2.11.70) are
En = 0.257l'2(n + 1)2,
n
= 0,1,2, ..
(2.11.71)
1, ••
where n labels the energy level. A comparsion of these results with the exact values of Eo(N) obtained numerically for n = 0 (see table 2.2) reveals that an acceptable accuracy can be reached only for high N I therefore a construction of a more exact solution is needed. Real minimal solutions of this equation for various N are listed in the Table 2.2. Table 2.2. Comparison of exact and approximate values of eigenvalues E(N) N
Exact value E
Eo
Error %
N
1 2 4 10 50
1.ססoo
0.9100 1.0422 1.2385 1.5731 2.1074
9.0 1.72 1.04 0.81 0.10
200 500 1500 3500
1.0604 1.2258 1.5605 2.1052
Exact value E 2.3379 2.4058 2.4431 2.4558
Eo
Error %
2.3383 2.4032 2.4428 2.4555
0.02 0.01 0.01 0.01
2.11 Nontraditional Asymptotic Approaches
131
Let us consider the function
,
0 ~ x < 1, x > 1.
Now we split the function
--+
p-N-l""((N
+ 1,p) ,
where ""((... ) is the incomplete ""(-function, and p are parameter of the Laplace transform. After splitting ""((N + 1,p) on liN and inverting the Laplace transform term by term one obtains
+ 1)-1 + 6(1)(X - 1)(2N + 1)-1(2N + 2)-1 + ... 1)(2N + 1)-1(2N + 2)-1 ... (2N + i)-1 + .... (2.11.72)
Thus, in the interval 0 1Plxx
+
<x
~
1, (2.11.67) may be represented by
- E1Pl = 0 .
(2.11.73)
Seeking a solution of equation (2.11.73) in the form of the expansion 00
1Pl =
00
L 1Pa (2N + 1) -1 ;
E = LE(k)(2N + 1)-1 ,
k=O
k=O
after splitting into (2N
+ 1) -1, one obtains a
recurrent system of equations
-1PI0xx - E(O)1PlO = 0,
(2.11.74)
-1Pllxx - E(O)1Pll - E(l)1PlO
+ 6(x -
1)1P1O = 0,
(2.11.75)
Solving equation (2.11.74) for the case of symmetry with respect to the line x = 0 (the antisymmetric case may be considered similarly), one obtains 1Pl
= C cos AX,
A = (E(O))1/2.
(2.11.76)
Now let us consider the zone x > 1. For the zero-order approximation one can neglect the term E1P2 ~/.(O) _ x2N ~/.(O) = 0/2xx 0/2
0
A boundary condition for x
1P~O)
--+
0
for
x
(2.11.77)
.
--+ 00
can be formulated as follows:
--+ 00.
(2.11.78)
The solution of the boundary value problem (2.11.77), (2.11.78) is the following
1P~O) = CI X1 / 2 K II (VX N+ 1), where K II is the Bessel function, v = 0.5/(N
(2.11.79)
+ 1).
132
2. Discrete Systems For x for
=
1 the solutions
x =1
1/Jl and 1/J2 must be matched. Then one obtains
1/Jii) = 1/J~i), .I.(i) 'f/lx -
.I.(i) 'f/2x'
(2.11.80) .
~
= 0 , 1 , 2, ...
From (2.11.80), assuming i = 1, we obtain a transomdental equation for
.oX: -ctg'\ = 2Kv (v) - KI-v(V) - Kl+v(V) Now, we consider the equation of_a higher-order approximation for x > 1. If we suppose 1/J2 in the form 1/J2 = 1/J(x), where x = X N + 1 j(N + 1), we have the following equation for the function t/;(x):
1fi;tx + Nx- N - 1 if;;t + Ex- 2N if; - if; = O.
(2.11.81)
After expanding the function x- 2N and x- N - 1 in a series of the power 1j(2N + 1) and 1j(N + 2) as described above, one can obtain 00
x- 2N =
I)-1)i O(i)(1 - 1jx)(2N + 1)-1 ... (2N
+ 1 + i)-I,
(2.11.82)
i=O 00
X- N - 1
=
I)-1)i O(i)(1-1jx)(N+2)-I ... (N+1+i)-I.
(2.11.83)
i=O
Substituting expressions (2.11.82), (2.11.83) into (2.11.81) and splitting it into 1 j N, we have a recurrent sequence pf equations, whose solution gives us the possibility formulating the boundar~ conditions for 1/J~i). In conclusion, we may say that the approach proposed above is the natural asymptotic method for solving the differential equations which contain the term x 1+ e5 for 0 ---. 00. A similar asymptotical approach for the case of small o was proposed in [30d]. Matching solutions for 0 ---. 0 and 0 ---. 00 by means of a two-point Parle approximant one can obtain a solution for any value of
O.
2.12 Pade Approximants 2.12.1 .One-Point Pade Approximants: General Definitions and Properties
The principal shortcoming of the perturbation methods is the local nature of solutions based on them. As the technique of asymptotic integration is well developed and widely used, such problems as elimination of the locality of expansion, evaluation of the convergence domain, construction of uniformly suitable solutions, are very urgent.
2.12 Pade Approximants
133
There exist a lot of approaches to these problems [94, 154]. The method of analytic continuation (for example, the Euler transformation t = c(1 +e)-I) requires a knowledge of the positions of the singularities' of the sought function of the parameter e [154, 70d]. It is useful to apply those methods in cases when a great number of expansion components is known. It is then possible, using, for example, the Domb-Sikes diagram [40, 70d] , to determine the positions of the singularities' and to perform analytic continuation. A significant number of expansion components is also necessary to apply the methods of generalized summation. Not diminishing the merits of these techniques, let us, however, note that in practice only a few of the first components of the perturbation theory are usually known. Lately, the situation has indeed changed a little due to the application of computers. However, up till now, there are usually 3-5 components available of the perturbation series, and exactly from this segment of the series we have to extract all available information. For this purpose the method of Pade approximants (PA) may be very useful [4, 5, 123, 144, 64d, 67d]. Let us consider PAs which allow us to perform the most natural, to some extent, continuation of the power series. Let us formulate the definition. Let 00
F(e) =
L Cie i , i=O
",m
Fmn (e) =
L...i=O ",n
L...i=O
i
aie b. i '
Ie
where the coefficients ai, bi are determined from the following condition: the first (m + n) components of the expansion of the rational function Fmn(e) into the Maclaurin series coincide with the first (m + n + 1) components of the series for F(e). Then Fmn is called the [min] Pade approximant. The set of Fmn functions for different m and n forms the Pade table. The diagonal PAs (m = n) are the mostly widely used in practice. Let us notice that the PAis unique when m and n are specified. To construct the PAs, it is necessary to solve a system of linear algebraic equations (for optimal methods for the determination of PA coefficients see [42, 43, 99]). The PAs have found wide utilization in a series of branches of mathematics and physics, and particulary for enlarging the domain of applicability of series of perturbation methods. The PA performs meromorphic continuation of the function given in the form of power series, and for this reason it allow us to achieve success in the cases where analytic continuatioin cannot be applied. If the PA sequence converges to a given function, then the roots of its denominators tend to singular points. It allows us to determine the singularities with a sufficiently great number of series components, and then to perform the analytic continuation. The data concerning convergence of the PA could have applications in practice only as options which would enhance the reliability of the results. Indeed, in practice it is possible to construct only a limited number of PAs, while all convergence theorems require information about an infinite number of them.
134
2. Discrete Systems
Gonchar's theorem [39d] states that if none of the diagonal PAs ([n/n]) has any pole in a circle of radius R, then the s:equence [n/n] converges uniformly in the circle to the initial function f. Futher, the lack of poles in the sequence [n/n] in the circle of radius R implies the convergence of an initial Taylor series in the circle. As the diagonal PAs are invariant with respect to the fractional-linear transformations z ---. z/( CtZ + (3), then the theorem is valid only for the open circle containing the expansion point and for any domain being a union of such circles. The theorem has one important consequence for continuous fractions, namely: the holomorphity of all suitable fractions of an initial continuous fraction inside a domain n implies uniform convergence of the fraction inside
n.
An essential disadvantage in practice is: the necessity of verifying all diagonal PAs. The point is that if inside a circle of radius R only some subsequence of the diagonal sequence PA has no pole, then its uniform convergence to the initial holomorphic function, in the given circle, is guaranteed only for r < ro, where 0.583R < ro < 0.584R. There exists a counterexample showing that in general r < 0.8R. Since in practice only a finite number of components of the series of the perturbation theory is known and there are no estimations of the convergence rate, then the above theorems could only increase the likelihood of the results obtained. This likelihood is also augmented by known "experimental results" since the practice of PA application shows that the convergence of PA series is usually wider than the convergence do~ain of the initial series. Let us note that widely applied continued fractions form a particular case of PAs. In fact, the suitable fractions, representing the sequence of approximations of the continued fraction, coincide with the following PA sequence: [0/0]' [1/0]' [1/1], [2/1], [2/2]' _.. _Therefore we shall not separate the case of the application of continued fractions. The following circumst8J}.ces are essential. In the perturbation theory asymptotic series, divergent for all values of the parameter c =f:. 0, are very often obtained. This does not permit us to evaluate the value of the sought function with arbitrary precision for any c. At the same time a transformation with a PA (or into a continuous fraction) gives an expression suitable over in a wide range of problems. The approach is strictly mathematically proved for those series where (-l)nCn (Cn is the n-th coefficient of the series) is the n-th moment of some mass distribution, but numerous applications of similar approaches also show their applicability to more general cases. 2.12.2 Using One-Point Parle Approximants in Dynamics
We shall consider the Duffing equation which can be studied with different methods, which allows us to compare their efficiency. We shall apply the perturbation method combined with a PA to the problem.
2.12 Pade Approximants
135
The equation is written in the form
it. + u + u 3 = O.
(2.12.1)
The vibration frequency w has its exact value W=
1rv'1 + A2 2V2K(B) ,
(2.12.2)
where B
7r/2
jA2
= arctg V2+"A"2'
K(B) =
J VI -
o
d1/J
,
A2(2 + A2)-1 sin 1/J 2
K (B) is an elliptic integral of type I. The asymptotic expansion of w in terms of small A 2 (where A is the amplitude of vibrations) has the form w = 1+
3
2
SA -
21 4 256 A
81
6
6549
+ 2048 A - 262144 A
37737 A 10 _ 9636183 A 12 67108864
+ 2094152
+ ....
8
(2.12.3)
First, we shall restrict ourselves to the first three components of the series (2.12.3) and we shall construct the PA [2/2] 32 + 19A2 W2 = 37 + 7 A2 .
(2.12.4)
Taking into consideration the components""' A6 of the frequency expansion, we have W4
1 + 1.13A2 = 1 + 0.756A2
+ 0.261A4 + 0.0599A4·
(2.12.5)
Continuing the process we obtain a sequence of diagonal PAs in the form W2n =
2n
~ (tiC
2i
( 2n
~ Pic
2i
)-1
Along with the diagonal PAs we shall study an element of the Pade table of order [2/4] 1 + 0.513A2 w = 1 + 0.138A2 + 0.030A4'
constructed with the first three components of the series (2.12.3). The result of the frequency calculations according to (2.12.3)-(2.12.5) are graphically presented in Fig. 2.22. The curves 1, 2, 3 correspond to the sums of three, seven and eleven components of the series (2.12.3). Curves 4, 5, 6, correspond to the Pade approximants of order [2/3], [3/3], [4/4]. The exact
136
2. Discrete Systems
s....----r--------,-----,-----,----,
-
w 4
3
4 2 I----J--~~---+----_+--~
.----1~-
I
------+----1
3 2 1 20
40 A 2 50
30
Fig. 2.22. Frequency of amplitude dependence for the Duffing equation constructed with the perturbation method and Pade approximants
solution is represented by the dashed curve. The nondiagonal approximant is represented by curve 7. It can be seen that the best approximation is achieved with the diagonal PAs. Lately solitons and solutions close to them have been widely used in mechanics. These are essentially nonlinear soh1tions which cannot be constructed using the quasi-linear approach when any nttmber of components is conserved. It is still more interesting to note that the PA allows the construction of solutions of that type, beginning from local (quasi-linear) expansions. Moreover, the term "padeon" has appeared. A model example is presented by the boundary problem
y" - y + 2y 3 = 0,
y(O) ,- 1,
y(oo) = 0,
which has the exact solution ("soliton") y = cosh-1(x).
(2.12.6)
A solution in the form of the Dirichlet series y =·Ce- x [1 - 0.25C 2 e- 2x
+ 0.0625C4 e- 4x + ... J,
C = const,
after rearrangement into the PA and determination of C becomes the exact solution (2.12.6). PAs often give a good result even for a small' number of components of the perturbation series. Obviously, however, the efficiency of the PA increases when the number of approximations increases. So, in [4, 5] many components of the expansion series of the amplitude e 2 of the period of the Van-der-Pol equation have been constructed by PAs which has led to the discovery of the
2.12 Parle Approximants
137
singularities of the ,sought period as a function of e 2 and then, using analytic continuation, the construct of a solution applicable throughout the range of e 2 . At present there is the a possibility of obtaining the approximations of a higher-order with computers. It can be imagined that in the case where a complicated problem of the construction of the approximation of a higherorder in the perturbation methods is solved, then it is desirable to try to apply PAs and other methods of convergence acceleration. At the same time it must be noticed that iterative methods are essentially simpler to realize by means of computer technology. PAs can be used to improve these methods. Let the iterative process have the form
T(uo) = 0,
Un = Tl(Un-d,
n = 1,2, ....
We introduce the function Sn(e):
Sn(e) = Uo + CUt - Uo)e + (U2 - ude 2
+ ... + (un
- Un_den.
(2.12.7)
For e = 0 we have Sn(e) ~ Uo, for e = 1 Sn(e) ~ Un' Then, we rearrange the series (2.12.7) with the PA and suppose e = 1: U
~
Sn
E:l Oi , m + p = n. = Uo +""p 1 + LJj=l {3j
(
2.12.8
)
Let us consider, as an example, the problem of big deflections of round isotropic plate of radius R, with a free opening of radius Ro and a rigidly restrained external outline on which a superficial pressure of constant intensity is acting. The problem solution was found in [50d] using the method of finite central differences for the Young modulus E = 62.4 kg/m3 and the Poisson coefficient v = 0.335, RoR- 1 = 0.1. The method of succesive approximations applied for the solution of the system of nonlinear algebraic equations, for comparatively big loads, converges for some 150-200 iterations, and the convergence to the solution has an osbillating nature. Table 2.3. Radial forces T in a round isotropic plate - iteration procedure Approxim. number
T
o
5.27286 1.09640 4.81246 1.45039 4.55120 1.67086 4.37191 1.82867 4.23735 1.94992
1 2 3 4 5 6 7 8 9
Approxim. number 10 145 146 147 148 149 150 151 152
T
4.13072 3.02320 3.11416 3.02603 3.11236 3.02680 3.11063 3.02849 3.10890
138
2. Discrete Systems
Table 2.3 gives the result of computations of the dimensionless radial force T = N r R 2 D-l for p = R- 1 , where r is the polar coordinate; q* = O.5qR4 (Dh)-1 = 35 is the intensity of the external load. Applying the method of generalized summing the situation can be improved (Table 2.4). Let us present the proposed method. The PA (2.12.8), taking into account four approximations, will have the form 5.319 - 284.883e - 27.606e 2 T = 1 _ 52.762e _ 47.992e2 (2.12.9)
Table 2.4. Radial forces T in a round isotropic plate - using of Pade approximants T
Approxim. number
Approxim. number
T
6
3.0656 3.0760 3.0791 3.0789 3.0789 3.0789
o 2.6955 3.0140 3.0941 3.0656
1
3 4
5
7 8 9
Solution [5Od)
When e = 1, the formula (2.12.9) gives T = 3.079. The boundary problem considered above demonstrates the high efficiency of Parle approximants to accelerate the convergence of iterative processes. PAs can be used for a heuristic evaluation of the domain of applicability of the perturbation theory series. The e Values, up to which the difference between calculations according to the segment of the perturbation series and its diagonal PA does not exceed a given value (e.g. 5%), can be considered as approximative values for the domain of applicability of the initial series. A transformation to a rational functional allows us to describe nontrivial behaviour at infinity and to take into consideration the singular points of the solutions. We shall consider,' as an example, the problem of the flow around a thin elliptical airfoil (Ixl < 1, Iyl ~ e, e « 1) by a plane stream of perfect liquid incoming with velocity v. The first few components of the asymptotic expansion of the relative stream velocity q* on the airfoil surface are: 1
q* =
2_ V -
1
+e
_
~ e2 x 2
2
1-x 2
_
~ e3 x 2
1-x2 + ...
The written solution diverges for x (2.12.10) by the PA, the singularity for x 2
q* =
(1 - x )(1 + c) 1 - x 2 + 0.5e2x2
+ O(e 4 ).
2
(2.12.10)
= 1. After replacing expansion = 1 disappears: (
2.12.11
)
Fig. 2.23 presents for e = 0.5: 1 - the exact solution, dashed line - the solution (2.12.10), 2 - PA (2.12.11), and the point line - the solution according to the
2.12 Pade Approximants
139
1.6 ....------oy----...,..---~---.,
q.
o
o
1.2
0.8
I 0.4
L-~
----
-----+----~
--
I 0.8
0.4
0.6
x 0.2
Fig. 2.23. Compansion of the PA approach and Lighthill method
Lighthill method [108, 154]' which gives in this case worse results than the PA. 2.12.3 Matching Limit Expansions From the physical point of view, every nontrivial asymptotic usually has an inverse. In other words, if an asymptotic for e --+ 0 (e --+ 00) exists, the asymptotic for e --+ 00 (e --+ 0) can be constructed. Then there appears one of the principal sharpest problems for the asymptotic approach - namely the construction of solutions appropriate for 0 « e « 00. This may be solved both on the level of solutions and on the level of equations. In particular, one can try to synthesize the limit equations with the purpose of obtaining a "complex" relationship allowing for a smooth transition from e --+ 0 to e --+ 00. For a synthesis of solutions one can utilize two-point PAs (TPPAs) [72, 8.1, 85, 87, 117, 127, 128]. The definition of TPPAs is given below. Let 00
F(e) ~
L aiei
for e
--+
OJ
(2.12.12)
i=O 00
F(e) ~
L
bie i for e
--+ 00.
(2.12.13)
i=O
The following function will be called the TPPA
where the coefficients Ok, (3k are defined so that the first p coefficients of the proper part of the Laurent expansion of
140
2. Discrete Systems
Let us investigate the model problem of vibration of a chain consisting of n masses m, joined with springs of rigiditya. The detailed model is a finite difference approximation of the longitudinal vibration of a rod. The deflection of the k-th particle (Yk) complies with the equation
miik = a [(Yk+l - Yk) - (Yk - Yk-d] ,
k
= 1,2, ... , n.
(2.12.14)
At the ends of the chain the boundary conditions are given by Yk = 0
for
k::; 0
and
k > n.
There are n possible proper forms of vibrations: kS1r
Yk = As sin - - cos(wst + 4>s), s = 1,2, ... ,n, n+l and the appropriate frequencies of free vibrations are given by
w. = 2ft; sin 2(:: 1)'
(2.12.15)
(2.12.16)
Let us construct the asymptotic expansions of the frequency W s in the vicinities of the points s = 0 and s = 2(n + 1). We substitute variables in the expression for w s putting
x=
x(0.51r -
X)-l,
X
= s1r[2(n
+ l)t 1 .
In the same way, instead of the segment [0, 2(n + 1)] for s, we obtain the semi-interval x E [0,00). Enumerating the expansions for x --+ ~ and x --+ 00, we obtain sin
2(1"~ x)
=;
[x - x2+ (1 - ~~)'x3
- (1 sin
1rX
2(1
+ x)
1-
=
2 1r
~2) x4 + ...J, x --+ 0, x- 2 +
8'
(1 _
- (1 - ":) x- 4
1r
2 )
12
+ ... ,
(2.12.17)
x- 3
x --+ 00.
(2.12.18)
A solution, appropriate for 0 ::; x ~ 00, can be obtained with the TPPA method 2 W -= p 1.57x + 0.81x J s 1 + 1.57x + 0.81x2 . (2.12.19)
ra [
y:;;;
The results of frequency calculations according to (2.12.16)-(2.12.19) are presented in Fig. 2.24. The exact solution (2.12.16) is designated by 1, the expansions (2.12.17) and (2.12.18) by 2 and 3. The rearranged Parle solution coincides very well with the exact solution over the considered interval. An analysis of the diagrams shows that the TPPA has enabled us to construct an approximative solution appropriate for any frequency of vibrations.
2.12 Pade Approximants
2.0
141
~---..,-----,----"""",--------,
w 1.6 --+------+.-------
1.2
-----------I
0.8 0.4
o -. -D.4l.....-.....---ll..--...L...----..J------'-------' 4 2 6 8 o
Fig. 2.24. Two-point PA in the theory of the oscillations of chain
An important TPPA application may be the inverse Laplace transform. Indeed, having a given transform, it is possible to investigate the inverse transform behaviour for t ---+ 0 and t ---+ 00. The problem is the inverse transform description for 0 « t « 00 [30d, 68d]. It is proposed in the monograph [68d] that only those components which give an asymptotic for p ---+ 0 and p ---+ 00 (whare p is the transform parameter) and also the principal singularities of the expression should be left under the integral sign of Mellin's integral. If the simplified integral can be calculated, then an approximate analytic solution is obtained. In spite of the unquestionable utility of this approach, convincingly proved in th~ above monograph, the problem on the whole remains unsolved. The PAs have been applied to the inverse Laplace transform to widen the domain of applicability of power expansions. One of the TPPA application methods for the solution of the inverse Laplace transform is the F(p) transform expansion in Taylor and Laurent series in the vicinities of the points p = 0 and p = 00, which is then followed by the replacement of F(p) by a rational function according to this scheme. Then the transition to the inverse transform is realized ac~ording to well-known rules. But the aim is achieved more quickly by applying the TPPA directly to the asymptotics of the inverse transform f(t). Here is an example:
F(p) = Ko(p)e- P , where K o is the McDonald function,
f(t) =
1 v 2t
rn; -
f(t) = t- 1
-
v7In + ...
4y 2
t- 2
+ ...
for for
t
t
---+
OJ
---+ 00.
142
2. Discrete Systems
The exact value of f(t) is f(t) = [t(t + 2)r 1/ 2 ,
t > O.
(2.12.20)
The TPPA has the form
f(t) ~ (t
+ V2t)-1.
(2.12.21)
Figure 2.25 presents the solutions (2.12.20) and (2.12.21) (curves 1 and 2 as appropriate). If the asymptotics are not of the power form, the difficulties are also surmountable. Sometimes the asymptotic for p --+ 00 may be represented as a sum of exponential functions or of sines and power series [85]. In other cases, it is necessary to introduce nonpower functions into the fractional-rational expressions and expand the latter into power series for t --+ 0 [11]. 0,5
f 0,4 0,3 0,2 0,1 0 0
4
2
6
t
8
:R.ig. 2.25. Matching limiting solution in the theory of the Laplace transform
Another interesting example is the Van der Pol equation. We give some necessary preliminary information according to [94]. The Van der Pol oscillator is governed by the equation
x
+ kx(X 2 - 1) + X = O.
The solution tends in time to an oscillation with a particular amplitude which does not depend on the initial conditions. The period of this limit oscillation T is of interest and is plotted in Fig. 2.26 as a function of the strength of the nonlinear friction, k. The continuous line gives the numerical results obtained by means of the Runge-Kutta method. The dashed curves give the second-order perturbation approximations
T = 21r(1
1
+ _k 2 ) + O(k 4 )
16 T = k(3 - 2ln 2)
as k ~ 0
+ 7.0143k- 1/ 3 + O(k-1ln k)
(2.12.22) as k
--+ 00.
(2.12.23)
The TPPA formula uses two terms of the expansion (2.12.22) and the first term of the expansion (2.12.23):
2.12 Pade Approximants
143
15 .----------,-----.----r---,----,---.,--------:-;]
T 12 - --
- - ---- --I----I----:o,.!.j'~--.._'I:~=-----=--!'~-___i
9
t---------t-----------
6 I -1-----
3
- ----
---l---------------+--- ---
OL--_ _L - -_ _L - - _ - - - - l_ _----L_ _ 2 3 4 1 5 o
---l._ _----'-_ _- - - '
6
k
7
Fig. 2.26. Various solution for the period of the Van der Pol equation
6.2832 + 1.5294k + O.3927k 2 T = ---I-+-0.-24-3-3-k--and shows good agreement with the numerical results for all values of k (curve 1 in Fig. 2.26). 2.12.4 Matching Local Expansions in Nonlinear Dynamics! Interesting results were obtained by the use of two- point Parle approximants in the theory of normal vibrations in nonlinear finite-dimensional systems [118, 56d].
Consider a conservative system
II = all ) + II = 0, Xi. = dx dt ' az' i = 1,2, ... , n, (2.12.24 II = II(x) is the potential energy, assumed to be a positive definite
'..
miXi
i
Z
Xi
where function; and X = (Xl,X2, ... ,xn )T. The power series expansion for II(x) begins with terms having a power of at least 2. Without reducing the degree of generalization, assume that mi = 1, since this can always be ensured by dilatation of the coordinates. The energy integralJor system (2.12.24) is n
~ LX~ + II(Xl,X2, ... ,xn ) =
h,
(2.12.25)
k=l
h being the system energy. Assume that within configuration space, bounded by the closed maximum equipotential surface II = h, the only equilibrium position is Xi = 0 (i = 1,2, ... , n). 1
By courtesy of Yu.V. Mikhlin
144
2. Discrete Systems
In order to determine the trajectories of normal vibrations, the following relationships can be used [135]: - II 12 + Xi1( - II) -- - II Xi (i = 1,2,3, ... ,nj X = Xi ).( 2.1 2 .26) 2x"i 1 h~n + L.Jk=2 x k These are obtained either as Euler equations for the variational principle in Jacobi form or by elimination of time from the equations of motion (2.12.24) with consideration for the energy integral (2.12.25). An analytical extension of the trajectories on the maximum isoenergy surface II = h is possible if the boundary conditions, Le. the conditions of orthogonality of a trajectory to the surface, are satisfied [135]: x~ [-IIx(X,
X2(X), ... ,xn(X)] = -IIXi (X, X2(X), ... ,Xn (X)), (2.12.27)
(X,X2(X), ...,xn(X)) being the trajectory return points lying on the II = h surface where all velocities are equal to zero. If the trajectory Xi (X) is defined, the law of motion with respect to time can be found using
x + IIx (XI,X2(X), ... ,xn(x)) =
0,
for which the periodic solution x(t) is obtained by inversion of the integral. Now consider the problem of normal vibrational behaviour in certain nonlinear systems when the amplitude (or energy) of the vibrations is varied from zero to an extremely large value. Assume that in the system
z + II
Zi
(Zl'
Z2, ... ,zn) = 0
(2.12.28)
the potential energy II(ZI' Z2, ..:, zn) is:a positive definite polynomial of Zl, "" Zn having a minimum power of 2 ~nd a maximum power of 2m. On choosing a coordinate, say Zl, substitute Zi = CXi where c = ZI(O). Obviously, XI(O) = 1. Furthermore, without loss of generality, assume XI(O) = O. Then
(2.12.29) where V = E~:OI ckV(k+~) (Xl, X2, ... ,xn ), v(r+l) contains terms of the power (r + 1) of the variables in the potential
V(c, Xl, X2, ... ,Xn ) = II(ZI (xd, Z2(X2), ... , Zn(X n )). It is assumed below that the amplitude of vibration c = z(O) is the independent parameter. At small amplitudes a homogenous linear system with a potential energy V(2) is selected as the initial one while, at large amplitudes, a homogenous nonlinear system with a potential energy V(2m) is selected. Both linear and nonlinear homogenous systems allow normal vibrations of the type Xi = kiXI, where the constants k i are determined from the algebraic equations
ki Vx(;) (1, k 2, ... ,kn ) = V:J;) (1, k 2, ... ,kn ).
2.12 Pade Approximants
145
A number of vibrations of this type can be greater than the number of degrees of freedom in the nonlinear case. In the vicinity of a linear system at small values of c, trajectories of the normal vibrations X~I)(X) can be determined as a power series of x and c (assuming that Xl = x), while in the vicinity of a homogenous nonlinear system (at large values of c), x~2)(x) can be determined as a power series of X and c- l . The construction of the series is described in [55d]. The amplitude values (at ± = Xi = 0) define the normal vibration mode completely. Therefore, for the sake of simplicity, only the expansions of p?) = xP)(I) and p~2) = xi(l) in terms of the powers of c wil be discussed below: 00
p~l) =
00
La;i)d,
(2) _ '"'" a(i) - j
Pi
-
LJlJj
C
.
(2.12.30)
j=O
j=O
In order to join together the local expansions (2.12.30) and to investigate the behaviour of the normal vibration trajectories at arbitrary values of c, fractional rational TPPAs are used: (i)
P8
=
~8 (i) ~1 L.Jj=O a j f..,-
E
8
. 0 J=
b(i)' J'
cJ
(2.12.31)
or ~8 (i) ~1-8 p(j) = L.Jj=O a j f..,8
~8
L.Jj=o
b(i) j
'-8'
(2.12.32)
cJ
Compare expressions (2.12.31) and (2.12.32) with the expansions (2.12.30). By preserving only the terms with the order of cr ( -8 < r ~ 8) and equating the coefficients at equal powers of c, n - 1 systems of 2(8 + 1) linear. algebraic equations will be obtained for the determination of a;i), b;i) (j = 0,1,2, ... ). Since the determinants of these systems L1~i) are generally not equal to zero, the systems of algebraic equations have a single exact solution, a;i) = b~i) = O. J Select a TPPA corresponding to the preserved terms in (2.12.30) having the nonzero coefficients. ay), bY). Assume that b~i) 1= 0, for otherwise as c ---+ 0 X~l) ---+ 00. Without loss of generality, it can also be assumed that bg) = 1. Now the systems of algebraic equations for the determination of a;i), by) become overdetermined. All the unknown coefficients a~l), ... , a~l),
bp), ... , b~l)
(i = 2,3, ... , n) are determined from (28 + 1) equations while the
"error" of this approximate solution can be obtained by a substitution of all coefficients in the remaining equation. Obviously, the "error" is determined
146
2. Discrete Systems
by the value of L1~i) , since at L1~i) = 0 nonzero solutions and, consequently, the exact Pade approximants will be obtained in the given approximation in terms of c. Hence, the following is the necessary condition for convergence of a succession of the TPPA (2.12.31), at s ---+ 00, to fractional rational functions: p(i)
=
E~ a~i)d
~) i b. ci =0 1
1=0
,,~ LJ 1
(b~i) = 1),
(2.12.33)
namely, lim
L1(i)
8-00
=0
(i
= 2,3, ... , n).
(2.12.34)
8
Indeed, if conditions (2.12.34) are not satisfied, nonzero values of the coefficients A~i), b~i) in (2.12.33) will obviously not be obtained. Conditions (2.12.34) are necessary but not sufficient for the convergence of the approximants (2.12.31) to the functions (2.12.33); nevertheless, the role of conditions (2.12.34) is determined by the following consideration. Since in the general case there is more than one quasilinear local expansion and essentially nonlinear local expansions are alike, the Dl.J,mbers of expansions of the respective type being not necessarily equal, it is the convergence conditions (2.12.34) that allow one to establish a relation between the quasilinear and essentially nonlinear expansions, that is, to decide which of them corresponds to the same solution and which to different ones. For a concrete analysis based on the above technique, consider a conservative system with two degrees of freedom, , whose potential energy contains the terms of the 2nd and 4th powers or the variables ZI, Z2. Substituting ZI = CX, Z2 = C1J, where c = ZI(O), (x(O) = 1), one obtains 2
V = c
X2
(
d l 2"
y2
)
+ d2"2 + d3xy + c4
x2y2
(x4
'1'1"4
+ '1'2 x3 y
4) =c2V(2) + c4V(4).
+ '1'3-2- + '1'4x y3 ,+ '1'5 ~
The equation for determining the trajectory y(x) is of the form
2y"(h - V)
+ (1 + y'2)(-y'Vx + Vy ) =
0,
(2.12.35)
while the boundary conditions (2.12.27) can be written
(~y'Vx + Vy)lh=v = O. For definiteness, let d 1 = d2 = 1 + '1'; d3 = -'1'; "II = 1; '1'2 = 0; '1'3 = 3; '1'4 = 0.2091; '1'5 = '1'. Write the equations of motion for such a system: Ii + x + '1'(x - y) + c2(X 3 + 3xy2 + 0.2091 y3) = 0, ii + Y + '1'(y - x) + c2(2 y 3 + 3x 2y + 0.6273y2x ) = O. (2.12.36) In the linear limiting case (c = 0) two rectilinear normal modes of vibrations y = kox, k~l) = 1; k~2) = -1 are obtained, while a nonlinear system
2.12 Pade Approximants
147
(where the equations of motion contain only the third power terms with respect to x, y) admits four such modes: k~3) = 1.496; k~4) = 0; k~5) = -1.279; k o(6) -- -5 . In order to determine nearly rectilinear curvilinear trajectories of normal vibrations, (2.12.35) is used along with the boundary conditions. By matching the local expansions the following Pade approximants are obtained I -IV II-V -1-1.11c2-O.275c4 1+1.202 p = 1+1.61c2+O.72C4 p = 1+1.00c2+O.215C4 "y
"y
= 0.5
1+1.062 p = 1+2.06c2+3.20c4
-1-2.76c2-1.36c4 P = 1+2.31c2+l.04c4
= 0.2
1+1.70c2 P = 1+3. 96c 2+ 13.29c4
-1-6.41c2-9.03c4 P = ~1:-+--=5:-:.3:-::::0--;c2""+--=7:-:.0:-=2---:c4'--·
(2.12.37)
The two additional modes of vibration exist only in a nonlinear system; as v increases (the amplitude c decreases), they vanish at a certain limiting point. For the analysis of these vibration modes, assume a new variable u = (p - 1.496)/(p - 5). By using the variable u, two expansions in terms of positive and negative powers were obtained; therefore, fractional rational representations can be introduced as above. By comparing these expansions, the following TPPAs are obtained III - VI 8.874u+1.126u 2 v = 1+4.300u+2.836u 2+O.549u 3 "f =
"'V
,
0.5
= 0.2
35.497 U +5.108u 2 v = 1+3.021u-0.794u 2+O.622u3
(2.12.38)
88.986u+1.470u 2 - 1-0.143u+3.747u 2+O.072u3 ·
v -
-:--~~--=-",",:,",-:-----;;:~~~
Now proceed to the determination of the limiting point. Obviously, it can be found from
8v =0 8u . From (2.12.38): at "Y = 2 the limiting point is v ~ 1.21, c ~ 0.91; at "Y = 0.5 the limiting point is v ~ 11.10, c ~ 0.30; at "Y = 0.2 the limiting point is v ~ 23.93, c ~ 0.20. Hence, as "Y -+ 0 the limiting point is characterized by the amplitude c -+ O. Therefore, the two additional vibration modes in a nonlinear system can exist at rather small amplitudes of vibrations. Note that the quasilinear analysis does not allow one to find these solutions even at small amplitudes.
148
2. Discrete Systems
In the limit, when I = 0, a linear system decomposes into two independent oscillators having identical frequencies and admits any rectilineal modes of normal vibrations. Obviously, the full system (2.12.36) at I = 0 admits four modes of vibrations (in the nonlinear case) Y2 = kY1, k = {1.496, 0, -1.279, -5}. Thus, fractional rational Pade approximants allow us to estimate the nonlocal behaviour of normal vibrations in nonlinear finite-dimensional systems. For system (2.12.36) the evolution of the modes of normal vibrations is shown in Fig. 2.27 using parameters ( = In(l + c2 h 2 ) and r.p = arctg p (the picture shows periodicity in
Evidently, the TPPA is not a panacea. For example, one of the "bottlenecks" of the TPPA method is related to the presence of logarithmic components in numerous asymptotic expansion. Van Dyke [154] writes: "A technique analogous to rational functions is needed to improve the utility of series containing logarithmic terms. No striking results have yet been achieved". This problem is most essential for the TPPA, because, as a rule, one of the limits c --+ 0 or c --+ 00 for a real m~hanical problem gives expansions with logarithmic terms or other gives complicated functions. It is worth noting that in some cases these obstacles may be overcome by using an approximate method of TPPA construction by taking as limit points not c = 0 and c = 00, but some small and large (but finite) values [69d]. On the other hand, in [61, 58d] were proposed so-called quasifractional approximants. Let us suggest that we have a perturbation approach in powers of c for c --+ 0 and the asymptotic expansions F(x), containing, for example, a logarithm for c --+ 00. By definition QA is the ratio R with the unknown coefficients aI, 131, containing both the powers of c and F(x). The coefficients ai, 131 are chosen in such a way that (a) the expansion of R in powers of c matches the corresponding perturbation expansion; and (b) the asymptotic behaviour of R for c --+ 00 coincides with F(x). The main advantage of the TPPA and QA is simplicity of algorithms and the possibility of using only a few terms of the expansions. Besides, it is possible to take into account the known singularities of the defined functions. On the other hand, one of the important problems of the TPPA and QA is to control the correctness of the realized matching. Sometimes we can use numerical methods [100] or a procedure of recalculation of the matching
2.12 Pade Approximants
149
OO...----r---.--r------,r-----,--,---,
VI
IV
V
(
III I
10
2 1
0.1 O+----I----L--.+---~t--------.::"f__--&---__I
-7r/2
-7r/4
o
7r/4
Fig. 2.27. The evolution of normal vibration modes. Local normal mode expansions are marked by I, II (quasilinear case) and III, IV, V and VI (essentially nonlinear case)
parameters [64d]. Along with the comparison of the known numerical or analytical solutions, numerical or experimental results, it is possible to verify the modified expansions by their mutual correspondence. To estimate the error of the obtained TPPA, the Newton-Kantorovich method is used, and then one Can utilize the well-developed mathematical techniques concerning the effective estimators. AlsO one may use one-point approximants for the expansion for £ -+ 0 and £ -+ 00 for comparing with the TPPA. But in general the question is open. It is known that the PA posseses the property of self-correction of the error [111, 113] and may be used for the solution of ill-posed problems. In other words, errors of the nominator and denominator mutually vanish. This effect is closely connected with the fact that errors in the coefficients of the PA don't spread arbitraly, but "mistaken" coefficients are created in the
150
2. Discrete Systems
new good approximations to the solution. But we do not know whether this property exists for the TPPA. With reference to this, we must say that many results in the theory of the one-point PA were obtained on the basis of numerical experiments. For example, many different methods for accelerating convergence of sequences and series were tested and compared in a wide range of test problems, including both linearly and logarithmically convergent series, monotonic and alternating series [147]. This paper gives detailed comparisons of all the tested methods on the basis of the number of correct digits as a function of the number of terms of the series used. Such computations would be very useful for the theory of the TPPA.
\
3. Continuous Systems
3.1 Continuous Approximation for a Nonlinear Chain Let us consider the system in which the masses, interconnected by a weightless beam, interact with nonlinearly elastic supports distributed equidistantly along the length (Fig. 3.1). The corresponding equation of the free motion in the absence of friction is written in the form ~
m
8 2w 8t 2 h(x - jl)
L
84w + EJ 8x 4
-
8 2w S 8x 2
j=-~ ~
+
L
(3.1.1)
q(w)h(x - jl) = OJ
j=-~
here m is the magnitude of each concentrated mass, w(x, t) is the transversal displacement, l is the spacing between supports (masses), h is the Dirac delta function, q(w) = aw + fnJJ3, and S is the stretching force. In a linearized system (b = 0), harmonic vibrations and waves conserving their shape in time playa fundamental role. But other types of waves, e.g., localized wave packets, inevitably "spread out" because of dispersion. In a nonlinear system, for EJ = 0 (the masses are connected by a string), the situation turns out to be reversed. Quasiharmonic waves are distorted, but there exist localized solutions of soliton type with a time frequency that exceeds the highest frequency of the natural vibrations of the linearized system. We demonstrate the possibility of constructing such solutions in the general case (EJ =I- 0). We write the equations of motion using a finite difference approximation of the elastic forces in the beam
d 2 v· dT;
+ a(6vj -
+ Vj+2 + Vj-2) Vj-I) + Vj + VJ = 0,
4Vj+l - 4Vj-l
+,8(2vj - Vj+l -
where Vj = Wi!!'
T=~t,
EJ a = al3'
S ,8 = al'
(3.1.2)
152
3. Continuous Systems
;'
Fig. 3.1. Weightless beam with discrete masses on nonlinearly elastic supports
System (3.1.2) has a stationary solution in the form of a "sawtooth" standing wave vi = (-l)iV(r), where the function V( r) satisfies the differential equation d2V dr 2
+ 4( 4a + .8)V + V + V 3 = O.
-
We shall seek localized solutions of soliton type in the form vj(r)
= (-l)iVj(r),
(3.1.3)
assuming that the functions Vj(r) vary smoothly with the index j. Substitution of (3.1.3) into (3.1.2) leads, after some transformations, to the system d 2 V· dr;
+ a(6Vj - 4Vj+l - 4Vj-l + Vj+2 + Vj-2) +(8a + .8)(Vj+l + Vj-l - 2Vj) + 4(4a + .8)Vj + Vj +
(3.1.4)
"J3 =
Using the continuum approximation to replace the set of functions the function V(e, r) of two variables, we pbtain the equation ,
a2 v
ifiv
a2v '
3 ar 2 + a ae 4 + (8a +.8) ae 2 + 4(4a + .8)V + V + V = 0,
O. VJ(r)
by
(3.1.5)
e
where = xiI. Seeking solutions of soliton type with small amplitudes, we take as the first approximation a function that is harmonic in time: V(e, r) = A(e) sinwr.
(3.1.6)
Applying the procedure of the Galerkin method to (3.1.5) and taking (3.1.6) into account, we obtain the equation 4 a A a ae
4
2 a A + (8a +.8) ae
2 2 3 3 2 + (w - wd A + 4A = 0,
(3.1.7)
where w21 = 1 + 4(4a + .8). Let w2-w21 = €2 > OJ then for the existence of the desired soliton solution it is necessary for the quantity € to be a small parameter, with A ""' c, c. Here in (3.1.7) the first term Can be discarded, so that the soliton solution corresponds to the separatrix of the second-order differential equation
t. ""'
2 a A (8a +.8) ae -
2
3 cA + 4A3 = 0
3.1 Continuous Approximation for a Nonlinear Chain
153
and has the form
Vi
() T
4( l)j
=
-
esinWT ch(eejjV4 j3 a +,8)'
(3.1.8)
For w 2 close to w~, the function (3.1.8) is evidently a spatially localized perturbation (Fig. 3.2) that performs periodic pulsations in time (an envelope soliton or "pulson"). As numerical investigations show, pulsons occur not only under special initial conditions but also as a result of "self-localization" of nonlocalized perturbations such as a sawtooth standing wave. Thus, a highfrequency vibration can precede the "soliton" stage of a dynamical process.
/
/
"\
""
\
/
/
\ ./
Fig. 3.2. Spatial localiza-
tion of oscillations The most important condition for the appearance of the soliton mode in real systems is the presence of a "quasilimiting" frequency in the spectrum of the natural vibrations of the linearized system, above which the pulson frequencies are located. Now we consider a simple formal way of constructing a continuum approximation for the problem of high-frequency oscillations of a chain of masses and we also formulate continuum equations which are able to describe rather satisfactorily both low- and high-frequency oscillations. The oscillations of a chain of masses coupled by nonlinear springs are described by the equations
d 2 uk
m dt 2
+ C(2Uk -
3
Uk+l -
uk-d
+ CIUk
= O.
(3.1.9)
The well-known continuum approximation of set (3.1.9) for the case of low-frequency oscillations has the form
154
3. Continuous Systems
m
8 2u
82u h C 8x 2 2
iJt2 -
+ Cl U
3 _
0,
-
(3.1.10)
where h is the distance between the masses. We write (3.1.9) in the form [153] 2
m
8iJt2u
. 2 ( -"2 ih 8x 8 ) + 4csm
U
+ CIU3
= O.
(3.1.11)
The low-frequency approximation is now obtained by expanding the operator sin 2 ( -1/2ih8/8x) in a Taylor: series
ih 8 ) sin ( -"2 8x = 2
2
2
h 8 -"4 8x2
+ ... ,
and the high-frequency continuum approximation is obtained by expanding that operator in the vicinity of the identity transformation: 2
2
ih 8 ) h 8 sin ( -"2 8x = 1 + "4 8x 2 2
+ ....
In the latter case the equation for the function u(x, t) is as follows: m
82 u iJt2
+ 4eu + ch
28
2
u 8x 2
+ Cl u 3 =
O.
(3.1.12)
For the displacements of the masses in the chain we obtain
u(x
+ h, t)
= -u(x, t)
h 8u(x, t) 8x
+ '2
+ :... \
For the Toda chain [153] the short-wavel~ngthcontinuum approximation is m
~~ + (4C +
::2)
e -bu = 0
The existence of simple expressions for continuum approximations for low- and high-frequency oscHlations makes it possible to construct composite equations which quite satisfactorily describe the processes for arbitrary oscillation frequencies. For instance, using (3.1.11) and (3.1.12), we can construct the composite equation 2
2
h 8 ) 8 u m, ( 1 - 48x 2 iJt2
-
28
2
u ch 8x 2
+ Cl
(
2
2
h 8 ) 3 1 - 4" 8x 2 u = O.
(3.1.13)
To illustrate the efficiency of such a combination, we show in Fig. 3.3 the results of determining the frequencies of a linear (Cl = 0) chain of n masses. The exact values can be determined from the formula Wk
=
2[£ 2(::1)· sin
3.2 Homogenization Procedure
155
1.--------r-~---~
0.5
n
7f/2
Fig. 3.3. Various approximation for
the frequencies of the chain
We plot in Fig. 3.3 the quantities Wk = 0.5Wk(mc-1 )1/2; the numbers 1, 2, 3, 4 indicate the exact solution (the discrete values of Wk are connected by a solid line) and the solutions obtained on the basis of (3.1.10), (3.1.11), and (3.1.13), respectively. The proposed method can be easily adopted to a lattice of higher dimensionality, and it allows the construction of relations with a higher degree of accuracy.
3.2 Homogenization Procedure in the Nonlinear Dynamics of Thin-Walled Structures At present, the method of homogenization is used to great advantage for solving variable-coefficient partial differential equations in such disciplines as the theory of composites [44, 50, 63, 101, 139] or the design of reinforced, corrugated, perforated, etc., shells [20, 23,60, 64, 65, 8d, 9d, 12d, 13dJ. An original nonhomogenous medium or structure is reduced to a homogeneous one (generally speaking, to an anisotropic one) with certain effective characteristics. The homogenization method allows one not only to obtain effective characteristics but also to investigate the nonhomogenous distribution of mechanical stresses in different materials and structures which is of great significance for evaluating their strength. Then, the main idea of the method is based on the separation of "fast" and "slow" variables. As a start, a certain periodic boundary problem is formulated ("cell" or "local" problem) and.its solution, assuming periodic continuation of boundary conditions, is obtained. For that purpose the local coordinates ("fast" variables, in the case of using a multiscaling method) are introduced. After that the averaging itself upon local ("fast") coordinates is performed. 3.2.1 Nonhomogeneous Rod
As an example we treat here the case of axial oscillations of a rod with periodic cross-section a(x/c) in a nonlinear resisting medium with periodic
156
3. Continuous Systems
properties b( x / c). The governing equation may be written in the following form: 2 a [ (X) au] (X) 3 (X) a u = 0, (3.2.1) ax a € ax + b € u - c € at2 where a, b, c are periodic functions,
We introduce the "fast" variable y = x/c. Then, the differential operator a/ax, applied to the function u(x, y, t), becomes
a
ax
+e
a
-1
(3.2.2)
ay'
Let us consider the following ansatz for the solution of (3.2.1):
u
= uo(x, y, t) + cUI (x, y, t) + e2u2(X, y, t) = ....
(3.2.3)
Now we substitute expressions (3.2.2) and (3.2.3) in (3.2.1) and identify the powers of e: £-2
e
-1
:y [a(y) ~:] = 0; a [au l ] ay a(y) ax
(3.2.4)
a2uo a + a(y) axay + ay
~
[
aU1] a(y) ay = 0;
(3.2.5)
12-
[a(y) aU 2] + a(y) a2uo + [a(y) aU 1] 2 ax ax ay ay ay a2Ul a2uo +a(y) axay + b(y)ug - c(y) 8t 2 = 0;
(3.2.6)
We use the technique described in [50, 139]. From (3.2.4) and the conditions of periodicity we have
Uo = uo(x, t). Then, (3.2.5) (the so-called "cell" or "local" problem) may be rewritten as
~
[a(y) aU 1] = _ duo auo . dy ax ay ay
After integration one obtains
aUl ay
-
auo ax
c(x, t) a(y)'
= - - + ......:....-,-'-
(3.2.7)
The function c(x, t) is defined from the conditions of periodicity for Ul:
Ul(X, y + 1) = Ul(X, y)
3.2 Homogenization Procedure
157
and may be written as
C( x, t) =
!
a:: '
a
I
ii =
[
] -1
a-I dy
Excluding the function BudBy from (3.2.6), one obtains
[8u
-B a -2] By By
BUI] B [a + -By Bx
+a ~ B-UO- + bUo3 2
Bx2
cB2uo - - -- 0 . &2
(3.2.8)
Now we apply the homogenization operator
1(... I
)dy
o to equation (3.2.8). The first two terms vanish due to the periodicity of the corresponding functions, and finally we have ~
B2 uO a Bx 2
- 3 _B2 uO + buo - C Bt 2 = 0,
where
I I
Ii =
I I
b(y) dy;
c=
o
c(y) dy.
0
This homogenized equation has only constant coefficients and its solution is simpler than in the case of the governing equation (3.2.1). For the function UI one may obtain from (3.2.7)
[~/Y a a-
UI =
Buo Bx
UI (x,
Y + 1, t) =
0
1
dy - Y] ,
UI (x,
y, t).
Now we must make two very important remarks. First of all, we have U
=
Uo
+ o(c),
but
Bu _ Buo Bu I () Bx - Bx + By + 0 c . So, we cannot obtain the correct expression for the derivative in the framework of the homogenized problem solution. Secondly, the solution of the local (cell) problem for a quasilinear differential operator (when the highest derivatives are linear) may be obtained from the linear boundary value problem.
158
3. Continuous Systems
3.2.2 Stringer Plate The governing object is depicted in Fig. 3.4.
l/2
w Fig. 3.4. Stringer plate
We use Berger's equation ([51J, see also Sect. 3.3.1 of this book) and add some terms to it, taking into account the rib discreteness: Ml
D.:1.:1W - N.:1W
L
+ Eclc
;
o(y -'kl)Wxxxx
k=-Ml
= - [Ph - Eelc
k~~M' b(y - kll] W".
Here a b
Nh2ala2 = 3D J J[(Wx )2 + (Wy)2] dxdy, o0 M 1 = O.5(M -I), l = 2a2M-1. The transition conditions from one part of the plate between ribs to the other may be written as
W+ -- W- ,
w+ y -- wy'
D(Wi;y - W;yy)
w+ yy -- wyy'
= EclcWxxxx + Pc FcWtt , (...)±
=
lim (...). y-+kl±O
We suppose that a typical period of the solution in the y direction (L) is much smaller than the distance between the ribs l (c = iL -1 « I). Then, we
3.2 Homogenization Procedure
use a multiscale approach. Let us introduce "fast" 1] (1] = L -ly). Then
~ 8y
1]1 (1]1 =
159
l-l y ) and "slow"
L-1~ + l-l~.
=
01]
81]1
The normal displacement W will be represented as the expansions W = W o(e,1]) + c4W1(e, 1], 1]I) + ... ,
e
where = L -IX. After substituting the above expressions into the governing relations and separating them with respect to c, one obtains -Wh7171171l711 = .1.1Wo - N oL 2D- 1.1Wo + phL 4D- l W Ott , (3.2.9)
JJ b"J
NoL2h2b1b2
= 3D
b1
[(WO€)2
+ (W071 )2] de d1],
b = L- 1a 1, b = L -l a2 ,
-b"J 0
Wll711=0=Wll711=1' W l711711 1711 =0 = Wl711711 Wl711711711
711 =0 -
1
Wl711I711=0=Wl7111711=1' 1711
Wl711711711
=1'
711 =1
1
= (Dl)-l(Ecl cWo€€€
+ pFcL 4W Ott )'
The conditions for a nontrivial solution of the boundary value problem (3.2.9) are .1.1Wo - NoL -2 D- 1.1Wo + phL 4D- l W Ott = (Dl)-l(EclcWott = (Dl)-l(EclcW071717171 + PcFcL4Wott. The exact solution of the cell problem (3.2.9) is
WI = (24DL)-1(Ecl cW o€€€€
+ PcFcWotd1]~(1]~
_1)2.
The function WI in the general case does not satisfy the boundary conditions, and leads to the appearance of boundary layers near the ends x = 0, a1. Let us suppose that the ends x = 0, a1 are clamped. To obtain the boundary layer function Wkp, we introduce the "fast" variable 1]1 = L 1x (then 8/8x = L - 18/8e1) and the expansions W kp = c4W 1kp (e,1],6,1]1,t) + .... After separating, one obtains
.11.11W 1kp = 0, where .1 1 ( •.. ) = (.. ')€1€1 + (.. ')711711; WI 1711 = W l71 1711 = 0; W l71 1711 =0 = WI 1711 =1 = 0; for 6 = 0 W 1kp = -WI, W 1kp6 = 0, for 6 -+ 00 W 1kp, W 1kp6 -+ 0; for {I = l-l a1 W 1kp = -WI, W 1kP6 = 0, for 6 -+ -00 W lkp , W 1kP6 -+ O.
160
3. Continuous Systems
This boundary value problem may be solved routinely by Kantorovich's variational procedure [48d]. Now let us compare the asymptotic solution with the exact one, which may be obtained for the static problem. In the nonlinear theory the exact solution may be constructed very rarely, and it is wonderful that we may do it for our very complicated problem. We choose the governing equation in the following form:
D\74W - N\72W
a4 w kl) ax 4
M
L
+ Eclc
8(X2 -
k=-M
= Q(Xl,X2)
1
_ ~ ~ . (1I"SXl) ~~q8psm - - cos (0.511"(2P + I)X2) . al a2 8=1 p=l
(3.2.10)
=
The plate is simply supported:
a2 w
W=--2 =0
for Xl = 0, al;
2w a W=--2 =0
for X2 = ±a2.
aXl
aX 2
(3.2.11)
One Can obtain a solution of the nonlinear boundary value problem (3.2.10), (3.2.11) in the form 00
W(Xl,X2) =
00
LL
W mn(Xl,X2),
,
m=ln=l
•
where W mn
( )-{f Xl, X2 -
+
X n cos (0.5.Bn 2) a2
211"j ,) -t-X2
oo
+L .
X j(+) n' cos (0.5.Bn 2
1=1
1
a2
j(_) (0.5.BnX2 +~ ~ nj cos . a2
211"j
- -t-X2
)}
1=1
. sin (
"'::1),
"'m = lI'm,
tin = 1I'(2n + 1).
Substi~uting into (3.2.10) and splitting it into cosines, one obtains an infinite recurrent system of nonlinear algebraic equations:
(a~ + JL2.B~)(a~ + JL2.B~ + A)en + a:n,K = Pmn ;
(3.2.12)
[a~ + JL2(a~t))2] [a~ + JL2(a~t))2 + A] e~t) + a~,K =
0;
[a~ + JL2(a~~))2] [a~ + JL2(a~~))2 + A] e~~) + a:n,K =
0;
where i = 1,2,3, ...
(3.2.13)
3.2 Homogenization Procedure
161
00
K = en
+ 2)e~j) + e~j)];
(3.2.14)
j=1
A =
3{(a~ + Jl2,6~)e~ +
00
L [a~ + Jl2(a~j))2] (e~j))
2
j=1 00
+ L [a~ + Jl2(a~j))2] (e~j))
2
};
j=1
~
ErI
= Da2 LJ cos
'Y
2
(1r(2n + l)k) M ;
k=-Ml
i"\>(1:") ~nt
=
/.l
~n
+ 21rMi'' na(~) t = /.l
~n
-
21rMi .
Then, one can rewrite the system in the form Pmn - 'Ya:nK
(a~ + Jl2,6~)(a~ + Jl2,6~ + A);
en = (+) _
eni - -
'Ya'!nK
.
][~ K
]'
[a~ + Jl2(a~~))2] [a~ + Jl2(a~~))2 + A]'
(:-) = _ nt
[a~ + Jl2(a~~))2
a~ + Jl2(a~~))2 + A
i
~ 1, 2, ....
Substituting expressions for en, e~t), e~~) (i = 1,2, ...) into (3.2.14), one can obtain K as a function of A: K = Pmn S(1
+ 'Ya~S + S(+) + S(-))-I,
(3.2.15)
where
5 = [(a~ S(+) =
+ Jl2,6~)(a~ + Jl2,6~ + A)J -1 ;
f: [a~ + Jl2(a~~))2] f: [a~ + Jl2(a~~))2]
-1
[a~ + Jl2(a~~))2 + A] -1 ;
-1
[a~ + Jl2(a~~))2 + A] -1 .
j+l
S(-) =
j+l
Taking into accoun~ formulae (3.2.13)-(3.2.15) one obtains (3.2.12) as a transcendental equation (with respect to the unknown A) that may be solved routinely by numerical methods. Then we will obtain K (using formula (3.2.15)) and the amplitudes en, e~t),e~~) (i=I,2, ... ). For the numerical investigation we choose a square plate loaded by the lateral load Q
=
Q10 sin (":'1 ) ~ (O.~:X2) .
162
Xl
3. Continuous Systems
We also suppose v = 0.2, EcIM/(Da2) = 200; = 0.5 a l, X2 = a 2lx 2, M2 = (Dh/a2)M2. The numerical results are plotted in Fig. 3.5.
qlO
ol----+--A---+--+--~~-___r-_r_:::;;;;;;;~__,
o
0.5
o
0.5
o
0.5
1
l:i 1
x2
1
Fig. 3.5. Bending moments in the Stringer plate in the perpendicular direction to ilieri~ .
3.2.3 Perforated Membrane
Consider the problem of transverse oscillations of a rectangular membrane weakened by a double periodic system of regularly spaced identical circular holes of radius a. The ratio c of the period of perforations to the characteristic size of the region n is a small quantity. The outer contour 8n of the membrane is rigidly clamped, while the edges or'the apertures 8ni are free. Let us begin from the linear case. In mathematical language we have the boundary value problem
2 2 2 (8 U 8 U) _ 8 u c 8x + 8 y 2
2
2
-
iJt2
in
n,
(3.2.16)
3.2 Homogenization Procedure U
= 0
(3.2.17)
an,
on
163
au an
(3.2.18)
is the transverse displacement of the points of the membrane, c2 = pi P, p is the tension in the membrane, and p is the density. We take a solution for the characteristic oscillations of the membrane in the form u(x, y, t) = u(x, y)eiwt , where A = w 2/c 2 and w is the circular frequency. Then, instead of (3.2.16) we obtain where
.
U
a2 u a2 u
ax 2 + a y 2 + AU = O.
(3.2.19)
We have presented the solution of the problem (3.2.19), (3.2.17), (3.2.18) just posed as an asymptotic series in the powers of a small parameter
u = uo(x, y) + e: (UlO(X, y) + Ul (x, y, c;, 1])) +e: 2 (U20(X, y) + U2(X, y, C;, 1])) + "', where c; = xle: and 1] = yle: are the "fast" variables. The functions Uo, UlO, U20, .. · depend only on the "slow" variables, and the other variables ui(i = 1,2... ) are periodic together with their derivatives with respect to the "fast" variables and have a period equal to that of the structure. Similarly, we expand the frequency A = Ao
+ e:AI + e: 2A2 + ....
After separating e: we obtain from (3.2.19) an infinite system e:- l
a2Ul
a2Ul
+ 87]2
0,
(3.2.20)
a2uo a2Ul a2Ul + 8 y 2 + 2 8x8c; + 2 8y81] a2U2 a2U2 + ac;2 + a1]2 + AoUo = 0,
(3.2.21)
ac;2
=
a2uo 8x2
a2Ul
8x 2
a2U2 a y 2 + 2 8x8c;
a2Ul
+
a2UlO
+ ax 2
a2U2 + 2 8y81]
+
a2U3 ac;2
+
a2U3 87]2
a2UlO
+ a y 2 + Ao(Ul + UlO) + AlUO =
O.
(3.2.22)
The corresponding boundary conditions (3.2.18) assume the form aUl e:o
an +
auo an = 0;
e: l
aU2
+ aU1 + aUlO
where
an
an
on ani -
an -
0
,
(3.2.23)
on an.
.
81 an is the derivative with respect to the
"fast" variables.
164
3. Continuous Systems
The solution of the boundary problem for a complex multiply connected region now breaks up into three stages. The first stage is the solution of the "cell problem" (3.2.20), (3.2.23). On the opposite sides of the "cell" the function Ul must satisfy the periodicity conditions
uIIE=b = uIIE=-b, 8Ul
_ 8Ul
8(, E=b -
ull71=b = ull71=-b, 8Ul _ 8 Ul
8(, E=-b' 81] 71=b -
(3.2.24)
81] 71=-b'
Using Galerkin's variational method to solve problems (3.2.20), (3.2.23) and (3.2.24), we represent Ul as Ul
~ ~ (
. m1r(, n1r1] m1r(, . n1r1]) Almnsm-b-cos-b- +A2mnCOS-b-sm-b- .
= 2of=:'o
After performing the necessary operations, we obtain A Imn
=a
8uoA* 8x mn'
A
2mn
* = a OO° 8y A mn,
A:n
where n is a constant determined from the vanishing of the variation of Galerkin functional. The second stage of the solution of the problem is the construction of the averaged relations. Applying the averaging operator to (3.2.21) i>(x,y) =
I~;I
JJ
q»(x,y,(,,1])d(,d1]\
{li
n;
where is a "cell" without holes, we obtain the averaged equation with a boundary condition on the outer contour of the membrane: 8 2uo 82uo 8x 2 + 8 2 + BAouo = 0 in n*, Uo = 0 on 8n, y where n* is a membrane without perforations, B =
(1 _1r (1 _1r 2 4 b2
a )
2 4 b2
a _
1r a 2 ~ ~ A* 2 b2
20 f=:'o
mn
J
1(%1rvm + n 2))-1 Vm+ n 2
2
2
'
and J 1 , is the Bessel function of order 1. At the third stage we find the first correction to the frequency AI. To do this we must determine the function U2 as a solution of the boundary value problem 82u2 8(,2 8U2
+
an +
a2U2 82uo 82uo 82ul 82ul 81]2 = - &:2 - 8y2 - 2 8x8(, - 2 8y81] -
8Ul 8n
+
OO lO
8n = 0
on
8ni ,
AOUO
in
ni,
3.2 Homogenization Procedure
165
with the periodicity conditions similar to (3.2.24). Proceeding as in the determination of the function Ul, we can obtain 00 00 ( . m1rE n1r1] U2 = ~~ Clmnsm-b-cos-b-
m1rE . n1r1]) + C 2mn cos -bsm -b- + cp(E, 1]), where
G1mn
= Al mn (~o
.,. a.;~o),
C 2mn
= A 2mn ( : ;
.,.
a.;~o),
and cp( E, 1]) is a function satisfying the condition cp( -E, -1]) = cp( E, 1]). The form of the function c.p(E, 1]) is unimportant, since it makes no contribution to the averaged equation and, consequently, is not used in the determination of AI. After averaging (3.2.22) we obtain
82ulO 82ulO 8x 2 + 8 2 + B(AOUlO + Al uo) = y
o.
To determine Al we multiply the equation just obtained termwise by Uo and integrate over the region r}* [120, 122]. If UlO = ill = 0 on 8n, then the differential operator
82ulO L(UlO) = 8x 2
+
82ulO 8y 2
is self-adjoint. Then Al = 0 and the expansion of the characteristic frequencies begins with A2 - a term of order c 2 . In that case, if ill does not satisfy the boundary condition on the contour of the membrane and consequently UlO ;;J 0 on 8n, we obtain a nonzero first correction to the characteristic frequt:ncy. Now let us investigate the nonlinear but quasilinear (the terms with derivatives in the governing equation are linear) case - a membrane on nonlinear support. The governing equation is 2
2
2
(8 U 8 U) 2 3 8 u C 8x 2 + 8 y 2 +ClU = 8t 2 ' 2
and the boundary conditions are given by (3.2.17), (3.2.18). Here c? is the rigidity of the nonlinear support. After splitting into c, one obtains the cell problem in the form (3.2.20), (3.2.23). Then its solution coincides with the solution of the linear problem, and the homogenized nonlinear equation may be written in the form 2 ( 82UO 82uo) 2 3 82uo C 8x 2 + 8 y 2 + BclUo = B &t 2 in 8n·,
Uo = 0 on 8n.
166
3. Continuous Systems
3.2.4 Perforated Plate
We use the averaging method for the computation of densely perforated plates. As has been mentioned above, having the solution of the static local problem, this approach gives the possibility to obtain, without basic difficulties, the solution of dynamical and quasilinear problems. We consider the problem of the bending of a rectangular plate, weakened by a doubly periodic system of holes. Let n be the domain occupied by the pl~te, let the exterior contour be 8n and let 8ni be the boundary of the hole. The periodic e of the structure is the same in both directions and small in comparison with the characteristic dimension of the domain n (e « 1). The boundaries 8ni of the holes are free, and the exterior contour 8n of the domain is fastened in a definite manner. We have the boundary value problem 8 4u 8 4u 8 4u . (3.2.25) ax4 + 2 8x 28 y 2 + a y 4 = fInn· M r = 0,
~ =
°
(3.2.26)
on 8ni ,
where 2 82u 2 8 u M r = vLlu + (1 - v) ( cos a 8x 2 + sin a 8 y 2 2
~
8 2u ) + sin 2a 8x8y ;
a
8
= cos a axLlu + sina 8yLlu 2
+(1- v)~ [cos2a 8 u as
8x8y
+ ~sik2a (8
2
2
U_ 8 U)] '. a y2 ax 2
2
u is the normal deflection, and a is the angle between the exterior normal n to the contour and the x axis. We represent the solution of the problem in the form of a series of powers of a small parameter e: _ U - Uo
' + eUI + e2 U2 + .. "
(3.2.27)
where Ui = Ui(X, y, E, TJ) (i = 0,1,2, ...), variables. Taking into account the relations
e
x/e,TJ
y/e are the "fast"
88188818
-'=-+--' ---+-ax ax e 8e' ay - ay e 8r} the initial equation and each of the boundary conditions splits with respect to e into an infinite system of equations
84uo 8e4
84uo
+ 28e28TJ2 +
8 4uo 8TJ4
=
0,
in n i ;
3.2 Homogenization Procedure
167
(3.2.28)
[( I - v )
sin
20 cos 0
2
] B3UQ
+coso - -3+
ae
+ sin a] a;;;.0 + [(1 - v) ( + sin a] :;2~
[(1- v) (
3 . ] a uQ +smo aea1]2
= 0,
a4Ul
B4Ul
[(
cos2asina -
cos 2a cos a -
on
ani;
0,
in
a4Ul
ae4 + 2 ae 2a1]2 + a1]4
=
)
ni ;
sin 20 cos 0 2
I-v - - - -
~ sin2a cos a )
~ sin2asina)
(3.2.29)
(3.2.30)
(3.2.31 )
168
3. Continuous Systems
(3.2.32)
(3.2.33)
(3.2.34)
(3.2.35)
3.2 Homogenization Procedure
169
where {}i is a characteristic cell of the structure. Thus, the solution of the formulated problem (3.2.25), (3.2.26) for a composite multiply connected domain splits into a series of steps in domains with a simpler geometry, from which we can distinguish two fundamental problems: a local problem ("a problem on the cellU ) which consists in the solving of the biharmonic equation in the domain (}i with the given boundary conditions on the contour of the hole and the periodic continuation conditions on the opposite sides of the "cells" ull€=b = ull€=-b 8Ul
8~
8Ul
I€=b
= 8~
82ul
8~2
82ul I€=b =
83ul
8~3
I€=-b
ull 11 =b = ulI 11 =-b 8Ul 8Ul 81] 111=b = 8T] 111=-b
8~2
I€=-b
82ul 82ul 81]2 111=b = 81]2 111=-b
I€=-b
83ul 83ul 11=b 81]3 1 = 81]3 111=-b;
aJUl I€=b =
8~3
(3.2.37)
and a global problem which consists in the solving of an averaged equation of the form (3.2.36) in the domain {}* without perforations and with the initial boundary conditions on the contour of the plate. As follows from relations (3.2.28), (3.2.37) and (3.2.29), (3.2.37), the functions UQ, Ul do not depend on the fast variables, Le., UQ
= uQ(x, y);
Ul
= Ul(X, y).
(3.2.38)
Consequently, the solution of the problem is represented in the form of a sum of a certain smooth function and a small fast oscillating correction; moreover, the expansion starts with the U2 O(e 2 ) term. After the successive solving of the "cell problems" (3.2.30)-(3.2.32), (3.2.37) and (3.2.33)-(3.2.35), (3.2.37) and the determination, of the functions U2, U3, we determine the principal part of the solution, Le., the function UQ. Applying to (3.2.36) the averaging operator (...) f'.I
(.
~.) = I~:I
JJ(...)d~
d1],
{li
we obtain the averaged equation in the form
170
3. Continuous Systems
84uo ( 8x 4
84uo
+ 2 8x 28 y 2 +
If ( + In; I
84uo 8y4 -
4 8 u3 8x8€3
1
D.
8 4u 2 + 8y 28€2
I
)
Ini I In;1
(3.2.39) 4u
84u3 84u3 (J4u3 3 8 2 + 8y8€2817 + 8x8e817 2 + 8y8173 + 8x 28€2
3 84u2 + 8y28172
+
84u2 8x28172
+
4
4 8 u2 ) de d - 0 8x8y8€817 c., 17 .
We obtain the solution of the "cell problems" (3.2.30)-(3.2.32), (3.2.37) and (3.2.33)-(3.2.35), (3.2.37) for a plate with a square net of perforation holes of radius a by making use of the Bubnov-Galerkin method, modified for the case of natural boundary conditions. We consider them successively. We represent
~ ~(
U2 = ~o
f='o
. m1r€ . n1r17 A 1mn sm -b- sm -b-
. m1r€ n1r17 + A 3mn sm -b- cos -b-
m1r€
n1r17
+ A 2mn cos -b- cos -bm1r€ . n1r )
+ A4mn cos -b- sm - b17 - '
(3.2.40)
where A 1mn , A 2mn , A 3mn , A4mn are constants, defined from the conditions of the vanishing of the variation of the Galerkin functional. The selection of the function U2 in the form (3.2.40) allows us to satisfy the periodic continuation conditions (3.2.37); then the variation of the Galerkin functional becomes
where by M r - I, ~ we have denoted expressions (3.2.31), (3.2.32), respectively. As one can see from (3.2.41), by virtue of the symmetry of the considered domain the constants, A 3mn , A4mn are equal to zero. The unknowns A 1mn , A 2mn are determined after carrying out the standard procedure of the Galerkin method: 282uo2 82uo= 2 82uo * 1mn 2 A*2mn + b - 8 A = b 8x8y A 1mn ; A 2mn = b - X8 y 2 A*2mn,(3.2.42) . where Aimn' A*2mn, A*2mn are numerical coefficients. By means of the same scheme, after transforming the right-hand sides of the equations and the boundary conditions, we obtain the solution of problem (3.2.33)-(3.2.35), (3.2.37). From similar considerations, we represent
~~( . m1re n1r17 U3 = ~~ B1mnsm -b- cos-b +U2(UO
-+
ud;
m1re . n1r17 ) + B 2mncos-b-sm -b-
(3.2.43)
3.3 Averaging Procedure
171
and with the aid of the Galerkin method we find 3 (-
b
B 1mn =
B2mn
B"'lmn
eJ3uo 8x 3
eJ3uo B"'2mn 8 3 y -
3 (-
=b
-
u
eJ3 o ) + =B"'lmn 8x8 y2
;
3
=
8 uo ) + B"'2mn 8x 2 8y ;
being numerical coefficients. This approach to solving local problems with the aid of the modified Bl1bnov-Galerkin method turns out to be especially efficient for the determination of the global characteristics, displacements and averaged coefficients, since for the determination of the latter one Can use integral representations. As an example we consider a plate for which alb = 1/3. If in expansions (3.2.40), (3.2.43) we restrict ourselves to one-term approximations, then for the coefficients we obtain B"'lmn, B"'lmn, B"'2mn, B"'2mn
Aill = 0.0102; A"'210 B"'110 B"'lll
A "'210 = A "'201 = -0.0156;
= A"'201 = -0.0090; = B"'201 = 0.0074; = B"'211 = 0.0042;
B"'110 B"'lll
= =
B"'201 B"'211
= 0.0014; = 0.0055.
Then, after some transformations of (3.2.39), we obtain the averaged equation in the form
- 84uo A 8x4 where
A, 2B
A=
- 84uo
- &4uo 8 y 4 = f,
+ 2B 8x 2 8 y 2 + A
are the averaged coefficients
0.860;
2B = 1.690.
In the case of one-term approximations, comparison with the known values [40d] shows the satisfactory accuracy of the results.
3.3 Averaging Procedure in the Nonlinear Dynamics of Thin-Walled Structures 3.3.1 Berger and Berger-Like Equations for Plates and Shells In 1955 Berger proposed approximate nonlinear equations for the deformation of rectangular and circular plates, neglecting the second invariant of the strain tensor in the potential energy expression (the "Berger hypothesis" [51]). Berger's equations have become widely used due to their simplicity and visualization. Later Berger's results were generalized for shallow shell and sandwich plate problems.
172
3. Continuous Systems
Similar equations were applied to dynamic problems. The adequacy and applicability of the "Berger hypothesis" were frequently and widely discussed in scientific papers. It has been shown that the "Berger hypothesis" leads to insufficient results when applied to orthotropic plates; there is no obvious pattern of generalization to shallow shell equations (for example the direct application to dynamic equations of a shallow shell was shown in [57] to be erroneous) . Various approaches were proposed to verify the "Berger hypothesis", including extravagant ones (propositions to regard the (1 - v) term as a small parameter, and to neglect the second invariant of the stress tensor instead of the strain tensor in the potential energy terms). Here we describe a noncontradictory derivation procedure of Berger-type equations in the application to rectangular and circular isotropic plates, and isotropic and sandwich shallow shells. It is shown that the second invariant of the strain tensor is small in a random way and this takes place only for isotropic single layered and transversally-isotropic three-layered plates; logically sequential procedures for the composition of Berger-type simplified theories require us to apply the homogenization approach. First of all, let us consider several intuitive considerations. The applicability of the "Berger hypothesis" to isotropic rectangular plates was justified by considerable amount of numerical analysis and appears to be beyond doubt. In other words, the contribution of the second invariant J 2 of the strain tensor to the potential is undoubtedly smaller than that of the first invariant J 1 . Taking into account '. J1
= Cl + C2;
Cl = U x
+ O.5w;;
J2
= CIC2 C2 =
O.25c~2;
v y + O.5w;;
C12 =
uy + Vx
+ wxw y
the corresponding inequality for a rectangular plate 0 < x < a, 0 < Y < b may be written as a
a
!
!(A+B 1 +C)dXd Y »(I-v) j j(A-B 2 )dXd Y
o
0
b
b
(3.3.1)
0 0
A = 2ux v y + w;u x + w;v y ,
+ B 12 , B 2 = O.5B ll + B 22 , B ll = u; + u~, B 12 = uxw; + VyW~, B 22 = (uy + vx)wxwy, C = O.25(w; + W~)2. B 1 ,= B ll
The main difference between the left and right hand parts of (3.3.1) is connected with the C-term. Let us consider the eigenvalue problem assuming that the displacements and bending moments are equal to zero along the plate boundary. Applying Galerkin's procedure for the one-term approximation (u, v, w) = Ai(t) x sin(m1rx/a) . sin(n1ry/b) one can see that the (A + Bd
3.3 Averaging Procedure
173
and (A - B 2 ) terms contribute equally (at least, by order of magnitude) to the potential energy, except for the special case a = b, m = n. Hence, the Cterm contribution to the potential energy must prevail, as for the m, n ~ 1 case. Then, due to differentiations, the magnitude of the C-term becomes significant. Moreover, the C-term contains a slowly varying part instead of the rapidly varying B 12 , B 22 terms, and the integrals of the former ones become small. These considerations have led us to the decision to use the homogenization method (the nonlinear WKB-method [160]), based on the high variability of the solution along spatial coordinates, for the purpose of composition of Berger-type equations. The nondimensional equations of motion of a rectangular plate may be written as (12(1 - V2))-I£\72\72w + £(FEEWT]T] - 2FET]WET] + FT]T]WEE) + WTT = 0, \72\72F + £( WEEWT]T] - W~T]) = 0, FT]T] = (1 - V2)-I(11.E + 0.5£w~ + V(VT] + 0.5w~)), FEE = (1 - V2)-I(YT]
+ 0.5Ew~ + v(11.E + 0.5w~)), FET] = -0.5(1 + V)-I(UT] + vE + £wT]wT]). where £ = h a; (E,1]) = (x/y)a; P = P/ Eha; (11., V, w) = (11" r = dt p(1- n11, 2)/E; \72 = 8 2/8E 2 + 8 2/81]2.
v, w)/h;
The most natural way of introducting "rapid variability" into the nonlinear system requires one to include the "rapid" variable £OO(E, 1]), regarding it as an independent variable. The value of a would be specified during the limiting (£ -+ 0) system derivation process. Now, following the multiple scale method, we obtain (the notation E, 1] describes the "slow" variables, as before)
8 8 00 8 8~ = 8E + £ T] 80 ;
8 81]
8
= 81] + £
00 8 T] BO .
We suggest that the functions P, w, 11., v are sums of "slow" (Le. depending upon the "slow" variables only) and "rapid" periodic components of the unknown period Oo(E, 1]) [12, 3d, 4d]: F = pO(E, 1]) + £{3t p1 (E, 1], £00),
+ £{32 W1 (E, 1], £00), 11. = 11,O(E, 1]) + £{33U1(E, 1], £00), v = vO(E, 1]) +£,B4V1(E,1],£OO).
W = wO(E, 1])
The following relations are to be used, too: pO ,..., £/'I WO ;
w0,..., £/'2;
11,0,..., £/'3;
V°,..., £/'4;
8/8r( ... ),..., £6( ... ).
There are asymptotic integration parameters ,Bi, /'1, /'2, /'3, /'4, fJ describing the relative orders of magnitude of the "slow" and "rapid" components: pO and wO, 11,0, vO and £. The noncontradictory choice of its values, being routine
3. Continuous Systems
174
work, has to be managed while satisfying the conditions of the noncontradictive character of the limiting (e ---+ 0) systems. The nontrivial limiting systems may be obtained from (3.3.1), assuming
a
= -0.5,
11 = 1,
(31
12
= 0,
= 0,
(32
< 0,
/33, /34
13, 14
> 0,
8
> -0.5,
=0
and may be written as (12(1 - v2))-IWJooo(lJ~
+ 0~)2 +
+(F~€O~ - 2F€110€011 + F~110VWJO + W~T = 0, FJooo(O~ + O~)
=
0,
(3.3.2) (3.3.3)
e- 1 FJoO~
+ F~l1 = 0.5(1 - v2)-I(WJ)2(0~ + vO~), e- 1 FJoO~ + F~€ = 0.5(1 - v2)-I(WJ)2(O~ + vOV, e- 1 FJOO€Ol1 + F~l1 = -0.5(1 + V)-I(WJ)20 xi0l1 ·
(3.3.4) (3.3.5) (3.3.6)
The underlined term in (3.3.2) may be derived using a O-averaging pro-
o
cedure, (... ) = 00 1 J( ... )dO --
0
1
-0
Foo = 0,
U
r..o
-0
0
0
0
(F€€, .l'ryl1' F€l1) = (F€€, F1J1J , F€l1) ,
= 0.5(1 - v2)-I(WJ)2(0~
+ O~).
Using the previously introduced variables we get a
b
:
(wm9~ + 9~) = :b I I (w~ + w~)~ dy + O(E); o 0 equation (3.3.2) becomes the Berger equation D\1 2\1 2w + N\1 2w + phwtt = 0; \12
2
=
8 8x 2
fj2
+ 8 y2;
II o
B =
Eh 1 _ v2'
(3.3.7)
b
a
- B N2ab
2
D' Eh B = 12;
2 2 (wx+wy)dxdy.
0
The strain compability equation becomes linear:
\14F = O.
(3.3.8)
One could easily obtain Berger's equations fOT the viscoelastic plate,
h2
12 r\12\12w - r N\1 2w T
where
+ phwtt
= 0
r'l/J = 'l/J J R(t - rI)drl, and R is the relaxation kernel. o
3.3 Averaging Procedure
175
Circural plates are to be considered separately for centre-holed and continuous plates. In the first case, Cartesian coordinates may be used, regarding (3.3.7), (3.3.8). In the second (ro = 0) case, r- 1 varying coefficients are to be taken into account. Finally, one obtains (3.3.7), where
82
1 fJ 'l = 8r2 + ;: 80 2
+
1 82 r 2 8e2 '
2n R
N =
2:R2 !J(w; + w;) dr de.
o0 Let us consider a shallow shell, the curvatures of which are k 1 , k 2 , and the in-plane dimensions are a and b. Assuming a ""' b, ki ""' 1 and following the procedure described herein, we obtain Berger's equations
4 + h\1kF + N\1 2w + ~ (W xx
D\1 w
a
! !(k + 1
o a
+W••
b
vk2)w
dydx
0
b
! !(k + 2
Vk1)WdYd Y)
+ phwtt
= 0,
o 0
'l4p
+ E'lkw = 82
'l k
0,
(3.3.9)
82
= k 1 8x 2 + k 2 8 y 2.
Two points of special value may be outlined. Firstly, equations (3.3.9) allow all the possible natural limiting passags: a Berger plate; a Kirchhoff nonlinear bar, a shallow arc and, finally (the problem which the Berger hypothesis approach failed to overcome), linear shallow shell equations. Secondly, the "Berger's hypothesis u applied to (3.3.9) appears to be invalid (the second invariant of the strain tensor energy term is not smaller in order of magnitude compared with the first invariant term). Let us look for approximate equations of transversally-isotropic sandwich shells. Introducing Do = Deo;
where the parameters e, jL depend on the bending stiffness of the load-carrying layers, and on the sandwich shear-resistance (e, eo, /3), can be calculated using the formulas given in [2] (Pk' h k denote the density and thickness of the k-th layer).
176
3. Continuous Systems
1, J.L ""' hI R, 0 ""' 1, approximate equations can written as Do(1 - OJ.LR 2V'2)V' 2V'2 X + hV' kP + NV'2 w
Assuming 00
""'
a b
+ ~ (W xx /
/
o
+ IIk 2)w dydx
0
b
a
+W yy / o
(k,
/
(k 2 + Vk1)WdYdX)
+ PIWtt =
0,
(3.3.10)
0
V' 4p + EV'kW = 0, 0.5(1 - v)J.LR 2V'21/J = 1/J } W = (1 - J.LR 2V'2)X .
(3.3.11)
Considering the case 0 < 1, the underlined term in (3.3.10) must be omitted as well. Limiting systems, governing static and dynamic behaviour of nonlinear bars, and linear plates are of no interest to us and are omitted. This investigation can be concluded as follows: 1. "Berger's hypothesis" in its initial formulation appears to be true for
isotropic single-layered and transversally isotropic multi-layered plates only. 2. As a matter of fact, Berger's equations represent the first approximation a of homogenization procedure (the nonlinear WKB method) when the rapid variability of the solution witH respect to spatial coordinates is assumed.
3.3.2 "Method of Freezing" in the Nonlinear Theory of Viscoelasticity The classical averaging method (in the form of the "method of freezing" [37d]) is a very usefull approach for solving the integro-differential equations of nonlinear dynamics. Let us consider for example the equation of the nonlinear oscillation of the viscoelastic rectangular plate (0 ~ x ~ a, 0 < Y ~ b) rV'4 w - 12rh- 2JV'2 w + phD-1Wtt = O. (3.3.12)
r
t
J R(t -
ab
td
Here
=
w(x, y, t)
. m7rX . n7rY = A(t) sm -a- sm -b-'
3.4 Bolotin-Like Approach for Nonlinear Dynamics
177
Then, for the amplitude A(t) one obtains a very complicated integro-differential equation
Dlr(A
+ 3h- 2 A 3 ) + A tt
= 0.
(3.3.13)
Here
D7r 4
[ (': )
2+ (~) 2] 2
D 1 = --=--------=--
ph
For low-frequency oscillations the function A(tI) is changed slowly with respect to the relation kernel R, so we may "freeze" A(tI) at the point t = tl [37d] and replace the integral t
J1
= / A(tI)R(t - tI) dtl o
by the following one
J1
~
A(t)J; t
J = / R(t - tI) dtl.
o Then (3.3.13) turns out to be the ordinary nonlinear differential equation with variable coefficients D 1 J(A
+ 3h- 2 A J ) + A tt = 0,
which can be solved by using the averaging procedure [141]. It is possible also to use the second procedure of freezing, applying to J the averaging operator II [37d]
-
1/ T
(...) - II( ... ) = lim T--+oo T
(...) dt.
o
As a result, we obtain the following equation with constant coefficients:
D 1-J(A
+ 3h- 2 A J ) + A tt
= 0,
which may be solved exactly or by the perturbation or averaging procedure [141].
3.4 Bolotin-Like Approach for Nonlinear Dynamics 3.4.1 Straightforward Bolotin Approach Bolotin [56] proposed an effective asymptotic method for the investigation of linear continuous elastic system oscillations with complicated boundary
178
3. Continuous Systems
conditions. Bolotin's method is also called the dynamic edge effect method. The main idea of this approach is to separate the continuous elastic system into two parts. In one of them - the so-called interior zone - solutions may be expressed by trigonometric functions with unknown constants. In the second part - the dynamic edge effect zone - Bolotin used exponential functions. The matching procedure (along the edges or unknown interior lines) permits one to obtain the unknown constants, and the complete solution of the dynamics problem may be written in a relatively simple form. This approximate solution is very good for high-frequency oscillations, but even for low-frequency oscillation cases the error is not excessive (see references [62, 73, 74] and the references quoted therein). These considerations are devoted to nonlinear oscillations of shallow cylindrical shells and rectangular plates. As the governing equations we use the approximate nonlinear equations obtained in Sect. 3.3.1: 8 2F 8 2w B2w 8 2w D'V 4 w - hR- 1 - 2 - N'V 2w + N l - 2 + N 2- 2 + ph ~2 = 0, (3.4.1) 8 Xl 8 Xl 8X 2 V" t"'7 4
v
l B2w 8 Xl2
=0
F
+ ER-
=
h2~a2]] [(::.)' + (:J] dx, dx o
'
where N
0
12D 2 N = h2ala2R
2,
•
2
1
Ja a w dXl dX2, J o 0
'.
N l = N2V.
System (3.4.1) may be rewritten in the mixed form 82ul -8 2 Xl 82U2
82ul X2 82U2
a2U2 1 8w - vR- -a = 0, Xl X2 Xl 82Ul -1 8w 8x~ + 0.5(1 - v) ax~ + 0.5(1 + v) aXlaX2 - vR aX2 = 0,
+ 0.5(1 - v)-8 2 + 0.5(1 + v) a a
2 2 'V w - D- l (N'V 2 N l 8 w _ N2 8 _ ax~ 8x~ 8U -12h- 2R- l ( 2 + V 8Ul _ R-lw) 8X2 8Xl
4
W_
(3.4.2) (3.4.3)
W) 2 w = O. 8t 2
+ ph 2D- l 8
(3.4.4)
Let us consider the boundary conditions as follows (clamped edges) Ul = U2
8w
= W = aXl = 0
Ul = U2 =
W
=
8w aX2
=0
for Xl = 0, al, for X2
= 0,
a2.
(3.4.5) (3.4.6)
3.4 Bolotin-Like Approach for Nonlinear Dynamics
179
Here we shall investigate normal modes of nonlinear" oscillations. For continuous systems this means that the dependences on the spatial and time variables may be separated in an exact or in an approximate way [159]. Let us represent the interior solution of (3.4.1) in the form
W(Xl,X2,t) = Wo = fIcoskl(Xl -xlO)sink 2(x2 -x2o)6(t), F(xll X2, t) = Fo = h cos k l (Xl - XlO) sin k 2(X2 - x2o)6(t),
(3.4.7) (3.4.8)
where k l (k2) and XlO(X20) are unknown constants; k l (k 2) is the wavelength and XlO(X20) is the phase shift in the Xl(X2) direction. Substituting expressions (3.4.7) and (3.4.8) into the initial relations, one can obtain an ordinary differential equation for the time function 6 and a relation between the functions Eland 6: 2 8iJt2 6 + W 2 ( 1 + 116 + 12E 2) 6 = 0, (3.4.9) l
6
= EfIk~(Rh)-l(k~ + k~)-26,
where w 2 = Dp- l h- 2n,
n
= (k~
(3.4.10)
+ k~)2 + 12(1- v2)(hR)-2kt(k~ + k~)-2,
11 = -12A 3 fI (nRh2ala2)-1(vk~ + k~), 12 = 1.5/;(nh2ala2)-1(k~ + k~) [k~(al - A l )(a2 - A2)
+ k~(al + Ad(a2 + A 2)]
, Al = 0.5kil [sin2k l (al - XlO) +sin2k l x lO], A 2 = 0.5k2"1 [sin 2k2(a2 - X20) + sin 2k2X20] , A 3 = -(k l k2)-1 [sinkl(al - XlO) +sinklxlOl . [cos k2(a2 - X20) - cosk2X20]. Let us designate the solution of (3.4.9) satisfying the initial conditions E = 0, dE/dt = 1 for t = 0 as
Using (3.4.2), (3.4.3) one finds UlO and UlO = U20 =
U20
in the interior zone to be
f3 sin k l (Xl - XlO) sin k2(X2 - X20)
where
f4
(3.4.11)
- -fIk2 (k~ + (2 + v)kD R(k~ + k~)2
(3.4.12) (3.4.13)
180
3. Continuous Systems
The constants kl, k2, XlO and X20 are unknown, and the boundary conditions are not yet satisfied. Consequently, one proceeds to construct corrective solutions in the narrow zone near the edges. Let US introduce the new variables Ulb, U2b and Wb - the components of the corrective solutions localized near the boundaries. The shell displacements can thus be expressed in the forms
(3.4.14) Substitution of expressions (3.4.14) into (3.4.2)-(3.4.4) yields 2 8 2(UlO + Ulb) 05(1 _ ) 8 (UlO + Ulb)
+.
8 Xl2
v
82 X2
82 (U20 + U2b) _ !:... 8(Wo + Wb) _ 0 8 8 R 8 -, Xl X2 Xl 2 8 2(U20 + U2b) 0 5(1 _ ) 8 (U20 + U2b) +. v 82 8 X22 Xl 2 8 (UlO + Ulb) _..!:.. 8(wo + Wb) - 0 R 8 X2 -, +0.5 (1 + v ) 8 Xl 8 X2
(3.4.15)
+0.5 (1 + v )
V'(wo + Wb) - h 2:
+(
8(WO + 8
a2 1
2
+
8
+ Wb)
2] d d Xl X2
1/(
8x~
Wb)) 2
0
2 _ 2 [ 8 (wo + v 8 2 Xl
a2 al
(WO+Wb)]
77[(a(w~:. o
Wb))
X2
{V 2(Wo
(3.4.16)
Wb)
l
)d\ d } Wo + Wb Xl X2
-
12 [8(UlO+Ulb) h2 R v 8XI
o 0
+ 8(U20 +
U2b) _
8X2
2 2 Wo + Wb] ph 8 (wo + R + D 8t2
Wb)
=0 .
(3.4.17)
Equations (3.4.15) and (3.4.17) are very complicated and cannot be solved in an explicit way without ~ymptotic simplifications. First of all, we must separate the interior and the corrective solutions. For this purpose one can use energy estimations. Thus, let us estimate the integral coefficients in (3.4.17) for large parameters k l k2 » 1 . f'V
a2 a l
II (~:~) II (:~)
2
dXl dX2
o0
a2 a l
o0
2
dx 1 d x 2
~ k~.
~ k..
2
II (::) II (~:) a2al
dXl dX2
~ k~.
dXl dx2
~ k2.
00
a2 a l
00
2
3.4 Bolotin-Like Approach for Nonlinear Dynamics
77(::) (~:)
dXl dX2
o0
181
~k
2·
If we eliminate all terms of lower order in (3.4.17), we obtain the simplified equation
V'(wo + Wb)
a2 at
/I
o _
h2 :
1
{v (Wo + 2
a2
Wb)
77[( :~)
Wo
2
o 0
2 aw o ) 2] d d _ 2 [ a (wo + Wb) Xl X2 v a 2 ( a X2 Xl
+ .
-
+
12 [ a(UlO + Ulb) d Xl d X2} - h 2R V aXI
2 a (wo + Wb)] a 2 X2
+
a(U20 + U2b) aX2
0
+ Wb]
Wo
ph
2
+ D
R
a2 (WOat+ Wb) =
0
2
(
'
3.4.18
)
Substituting equations (3.4.11)-(3.4.13) for the interior solution into system (3.4.15) and (3.4.18), one obtains approximate equations for the corrective solutions a2Ulb a2U2b v aWb v) a 2 + 0.5(1 + v) a a - R-a = 0, Xl X2 Xl X2 Xl a2U2b a2U2b a2UI b v aWb - R-a = 0, a 2 + 0.5(1- v) a 2 + 0.5(1 + v) a a x2 Xl Xl X2 X2 a2W b 4 'YIn ( a2Wb) 'Y2 n 2 2 V' Wb - k 2 k 2 cp(t) - a 2 + v-a 2 - k2 k 2 cp (t)V' Wb a2Ulb
a 2
+ 0.5(1 -
2
12
+V
I
[aUlb
- h2R v aXI
Xl
X2
+
aU2b Wb] a X2 -
I
+
(3.4.20)
2
2
ph a2Wb _ at 2 - O.
If + D
(3.4.19)
(3.4.21)
Equations (3.4.19)-(3.4.10), describing the corrective solutions, are linear differential equations with time-dependent coefficients. Spatial and time variables cannot be separated exactly in (3.4.19)(3.4.21), but one can use the variational Kantorovitch method [48d]. Let . us briefly describe this method. First of all, let uS represent the solution of (3.4.19)~(3.4.21) in the form satisfying the condition of periodicity: Ulb(XI,x2,t) ~ U I (XI,X2)CP(t),
t) ~ U2(XI, X2)cp(t), Wb(XI, X2, t) ~ W(XI' X2)CP(t). U2b(XI, X2,
(3.4.22)
Now, one can substitute expressions (3.4.22) into (3.4.19)-(3.4.21), multiply these equations by cp(t) and integrate over a period. Then one obtains
182
3. Continuous Systems
d u Ul d3lUl
+ d12U2 + d 13 W = 0, + d32U2 + d33W = 0,
(3.4.23)
12 CO = R2'
C2
=
2 '"Y2>..flh k? + k~
f
T
T
cp3 (t) dt,
>.. -:-1
= / cp(t) dt.
o
0
Now the partial differential equations (3.4.23) have constant coefficients, and one can use the operational method of the solution of differential equations with constant coefficients. In accordance with the main idea of this method one can operate with the deriv~tives 8/ 8Xi as with the constants and use the methods of linear algebra [82]. Then (3.4.23) may be reduced to a single equation for the function ifJ, D*ifJ = 0,
(3.4.24)
where U l = 0.5(1 - V)Dr3ifJ,
W
= 0.5(1 -
(3.4.25)
v)Di3ifJ.
(3.4.26)
Here D* is the determinant of system (3.4.23), and Di3 (i = 1,2,3) are the minors of the determinant D*. Let us now consider the edge effect at the Xl = zone edge. In this zone we represent Ul , U2 and W in the form
°
Ul = 8 1 (xd sin k 2(X2 - X20), W = 8(xdsink 2(x2 -X20)'
U2 = 82 (Xl) cos k 2(X2 -
X20),
(3.4.27) (3.4.28)
For 8 1 ,8 2 and 8 one can obtain a system similar to system (3.4.23) (where
8/8x 2
--+
k2 ).
The characteristic equation for system (3.4.23) is
(p2
+ k~)(h2p6 + aup4 + a12p2 + a13)
= 0,
(3.4.29)
3.4 Bolotin-Like Approach for Nonlinear Dynamics
183
where
an = -h 2 (ki - 4k~)
+ 01,
2 _ k2 [h2(2k2 + 5k 2) + CO(l - v )(2k? + a12 - 2 1 2 (k? + k~)2
k~) _
20 ] 1 ,
2
4 [2
2 2 CO (1 - v ) ] a13 = -k2 h (kl + 2k2) + (k~ + k~)2 - 01 , 01
= VCl
-
C2·
Equation (3.4.29) has two imaginary roots (P7 = +ikl and PS = -ik l , i 2 = -1) belonging to the interior solutions (3.4.11)-(3.4.13) and one must eliminate them from (3.4.29). The next six roots are WI) all )0.5 P4,l = ± ( -2rcos ( 3" +"""'3 '
±
P5,2 =
±
P6,3 =
1r -
( 2r cos (
3 1r
( 2r cos (
WI
) +
all
(3.4.30) 0.5
"""'3 )
+ WI all 3 ) + """'3 )
'
(3.4.31)
'
(3.4.32)
0.5
where WI
= arccos(qr- 3 ),
q
=
a~l all a12 a13 27 6 + 2'
. ()(3a 12 - a?1)0.5 3 r = sIgn q Then, near the boundary Xl = 0, one has 6
tP =
L Clk exp(Pkxd, k=l
(where Clk are arbitrary constants), and the corrective solution displacement may be written in the form
u~~
6
= coR
L C1kPk(Vp~ + k~) exp(Pkxd sin k 2(X2 -
X20)CP(t),
k=l
'U~~
6
= CO Rk 2
L Clk[k~ -
(2 +
v)p~l exp(Pkxd sin k 2(X2 - X20)cp(t),
k=l 6
W~l)
=
L Clk (p~ -
k~)2 exp(Pkxd sin k2(X2 - X20)CP(t).
k=l As described above, one can thus easily deal with the edge effect at the X2 = 0 boundary.
184
3. Continuous Systems
Let us rewrite boundary conditions (3.4.5) and (3.4.6) in the form UlO
+ u~~
Wo
+ W~l) = 0,
awo -aXl
UlO
+ U(2) lb
U20 + u~~
U20
= 0,
=
0,
+ u~~ +
awo aX2
-- +
(2)
Wo + w b = 0,
= 0,
(3.4.33)
aw~l) _ 0 aXl =
-
for Xl
(3.4.34) (3.4.35)
0,
aw~2) _ 0 aX2
= 0,
-
for X2 =
o.
(3.4.36)
These conditions must be supplemented by U(l) w(l) -+ 0 for Xl -+ 00, (3.4.37) U(l) lb' 2b' b u(2) w(2) -+ 0 for X2 -+ 00. (3.4.38) U(2) lb' 2b' b Then, the arbitrary constants may be determined from conditions (3.4.33), (3.4.35), (3.4.37) and (3.4.38). Using conditions (3.4.34) and (3.4.36), one has XlO = k1larctg {-
t C1kPk(P~ k=l
X20
= k,1 arctg { k 2
i; C2k(S~
-
k~)2 [kl t Clk(p~ _ k~)2] -I}" k=l
2 - kn [k1
i; C2k(S~
- k n2 ] -1 }. (3.4.39)
The oscillation forms can be separated1.into symmetry types. For the type symmetric in both directions one has
awo awo - a = UlO = 0 for Xl = 0.5a l, - a = U20 = 0 for X2 = 0. 5a 2. (3.4.40) Xl x2 For the type antisymmetric in both directions one has Wo = U20 = 0 for Xl = 0'.5al'
Wo = UlO = 0 for X2 = 0. 5a 2·
(3.4.41 )
Substituting the displacement into (3.4.40) and (3.4.41) and taking into account formulas (3.4.11)-(3.4.13), one obtains the transcendental equations kl(al - 2XlO)
= m1T',
k 2(a2 - 2X20) = n1T',
m, n = 1,2,... .
(3.4.42)
For m = 2k, n = 2k + 1, one has antisymmetric (in both directions) modes, and for n = 2k, m = 2k + 1 one has symmetric (in both directions) modes. Equations (3.4.39) and (3.4.42) may be solved routinely. In the limiting case 1/ R -+ 0 one obtains the solution for the nonlinear oscillations of a rectangular plate from equations (3.4.1), (3.4.4)-(3.4.7), (3.4.11), (3.4.17), (3.4.18), (3.4.26), (3.4.28)-(3.4.30), (3.4.34) and (3.4.36)(3.4.42). Now we examine the accuracy of Bolotin's method for the nonlinear case. Let us consider a simply supported square plate and introduce the notation
3.4 Bolotin-Like Approach for Nonlinear Dynamics
185
f* = /l/h, w* = w/wQ (w being the natural frequency of the square plate clamped along its edges and WQ = 1r(D/pha 4 ) being the square of the fundamental frequency of linear oscillations of the simply supported square plate). Amplitude-frequency dependencies for the nondimensional amplitude and frequency obtained by the present method (continuous lines) and by the method of approximate variables separation [158] (dashed lines) are shown in Fig. 3.6. The corresponding curves show satisfactory agreement. The discrepancy is not excessive, which confirms the acceptable accuracy of this method. 1.6
~
f*
~
1.2
l I
0.8
I
i
0.4
_ _....J....I._ __'______<._____'L..__
OL....-----"----'L.-o..-~--'-
o
2
4
6
8
10
12
_ _ ' __ _____'
14 w* 16
Fig. 3.6. Curves of the dimensionless amplitude f· versus the dimensionless frequency w· for the first five modes of the simply supported square plate
3.4.2 Modified Bolotin Approach Unfortunately, the above approach may be used only for rectangular regions. Here we propose a modification of the dynamic edge effect method (DEEM) that can be used to find the natural vibration frequencies and modes of plates and shells of a nonrectangular form at high amplitudes. For example, we deal with sector plates. To describe the motion of the plate, we proceed from the simplified equation proposed in Sect. 2.3.1:
2 ..::1 w - N..::1w
2
8 w + 0- 2 8t 2 = 0
82 8 82 ..::1=-+-+- 8r 2 r8r 8cp2
N=
~2 /
1
(J [
/
(:)\(:a:)}drd~
(3.4.43)
186
3. Continuous Systems
where r, 4> are the polar coordinates: r E [0,1], 4> E [0,0], R is the sector radius, w(r, 4>, t) is the deflection function as a fraction of r, and (J = R2(ph/ D)I/2. For the sake of argument, we shall assume elastic-restraint conditions on the contour of the plate: 2 8w 8 W] (3.4.44) wlrl,r2 = [ u r8
°
{Q8W ~
wlr = 3
Iw~i) (0,
<
-(I-
Q)[LlW-(V-l)8W]}lr r~
3
=0
(3.4.45)
(i = 0,1),
00
rl,r2,r3
where are the straight and circular parts of the contour, respectively, and u and Q are the reduced elastic parameters of the constraint (u, Q E [0,1]). Let the initial conditions be w(r,
w~(r,
(3.4.46)
We present the solution of (3.4.43) in the form (3.4.47)
w(r,
where A is the amplitude as a fraction of h. For 17(t) we first choose an approximation that satisfies the initial conditions (3.4.46): 17(t) = coswt.
(3.4.48)
We substitute expressions (3.4.47) and (3.4.48) into (3.4.43) and separate variables by Kantorovich's approximate method [48d] integrating over time on the segment [0, 21r/w]: (Ll
2
H =
-
H Ll -
.x2 )z(r,
°
e: Ii[(::y +V::" y]
(3.4.49)
rdrd",
), = (fW.
(3.4.50)
We shall assume below that .x is large. We present the solution of (3.4.49) far frQm the edges 4> = 0, in the form
°
z(r,
= W(r)!ii(
!iiI
= sin k(
(3.4.51)
e
where k, are the unknown wave number and phase. Substituting (3.4.51) into (3.4.49), we obtain
(~2 - H~ -
),2)W = 0,
~ =:1:2 + r~r -
eY
(3.4.52)
The solution of (3.4.52) that satisfies the bound (3.4.45) has the form (Jk and h are Bessel functions of the first kind for real and imaginary arguments)
3.4 Bolotin-Like Approach for Nonlinear Dynamics
+ C 2h({3r) ,
W(r) = C1Jk;(ar)
a=
187
{-~ +[(~r +Aff,
(3.4.53)
~= {~ + [(~r Hoff From (3.4.53), using the representations (3.4.47) and (3.4.51) and the boundary conditions (3.4.44), we obtain the transcendental equation
aJk+l(a) Jk(a)
+ J3h+I({3) Ik({3)
_ P(a 2 + (32) ,
1- Q
(3.4.54)
P = 1 - v(1 - Q)'
The first term in expression (3.4.53) corresponds to the contour state, while the second one desribes the dynamic edge effect (DEE) near the circular edge. Using the asymptotic formulas for the Bessel functions [93] (which are different for the cases k 2 = o('x) and k 2 = o('x)), we can show that the energy contribution of the DEE tends to zero as ,X -+ 00. To find a DEE-type solution at the straight boundaries, we present the solution of (3.4.49) in the form
z(r,ip)
=
W1(r)!Ii(cp),
WI = Jk(ar).
(3.4.55)
Substituting (3.4.55) into (3.4.49), applying (3.4.52) and the Bessel equation, and retaining only the terms of o(,X2), we obtain
2 82 2) [ 8 8cp2 + k 2 ( 8cp2 + k
r
2(2a 2 + H) ] W1!li
= O.
(3.4.56)
The expression for H that retains only the term corresponding to the ground state and is derived using the Green's second formula has the form
H =
~A2[J~(a) - J~_l (a)J~+1 (a)]Ba 2
B = 1 for k = 0,
B = 0.5(1 -
(3.4.57)
sink~ke)2
for k
i
O.
Since the variables in (3.4.56) cannot be exactly separated, we use Kantorovich's method [48d]:
2 82 2) [ 8 8cp2 + k 2 - s(a 2 + H) ] !Ii = 0, ( ocp2 + k
s=
I
W[r 3
(I
W[rdr) -1
dr;
s
~~
as A
(3.4.58)
~
00
k2 = O(A).
188
3. Continuous Systems
The general solution of (3.4.58) has the form
Wl(CP) + W2(CP), = Cll exp(gcp) + C 21 exp( -gcp) ,
w(cp) W2
where
W2
(3.4.59) (3.4.60)
=
is the DEE-type solution at the straight boundaries and
+ H)
g = [s(02
- k2
1
J2 .
(3.4.61)
The asymptotic equation follows from (3.4.55), (3.4.59), (3.4.60) and boundary conditions (3.4.44):
+ m1r,
kB = 2ke
m
= 1,2, ... ;
(3.4.62)
ke = arctg [Ug + (1 _:;(g2 + k2)] ,
(3.4.63)
From (3.4.62) and (3.4.63) as A -----+ 0 kB = m1r,
m = 1,2,....
(3.4.64)
The same result can be obtained directly from (3.4.56), using r as a parameter. This means that expression (3.4.62) becomes more nearly exact as A increases. Let us obtain the DEE-type solution at the straight boundaries under the assumption that k 2 = O(A). Substituting expression (3.4.55) into (3.4.49), eliminating A2 W 1 with the aid of (3.4.64), and retaining only the terms of 0(k 4 ), we obtain
2 82 2) (8 k2 - Hr 2) w(ep) ( +k 8
8
= O.
(3.4.65)
Separating the variables in (3.4.65) by Kantorovich's method, we obtain in the same way as above an equation that agrees superficially with (3.4.63) except that g has the form g = (k 2
+ sH)1.
(3.4.66)
The DEE-type solution is expressed by (3.4.60). If k 2 = O(A), we use the asymptotic representations for the Bessel functions to obtain from (3.4.66)
a = arctg ,
'Y
=0
(~P - ~) +" (~ + ~ + n)
;
{32 +~,
n = l, 1,2,... .
(3.4.67)
The quantity l can assume the values 0 or 1, depending on the value of p. For example, l = 0 if P = 1 and l = 1 if P =. O. We can express {3 and A in terms of 0 : {3 = (0 2 + H)1/2, A = 0{3. We now substitute expression (3.4.67) into (3.4.63) and, assuming the constant to be unknown, it is requred that (3.4.61) and (3.4.66) agree, retaining the terms of orders k and k 2 in the expression for o. We then obtain for s
3.4 Bolotin-Like Approach for Nonlinear Dynamics
S
8k) ( 1 P_Jl ) k + 4arctg 1r + 4n + 1 ( 1r 2
=
-1
Q
189
(3.4.68)
The unknowns k and 0 are found from the system of transcendental equations (3.4.54), (3.4.51) and are used to express the constants G i and Gil' The equations in which s is given by (3.4.68) are evidently to be preferred because they simplify the solution greatly without changing the essentials of the asymptotic method. Thus, (3.4.54) and (3.4.51) become independent if p = O. It also becomes unnecessary to evaluate the integrals in (3.4.58). To obtain an improved expression for 7](t), we represent the solution of (3.4.43) in the form w(r, cp, 0) = AJk(or) sin k(cp - e)7](t),
(3.4.69)
where k and 0 are defined above. Applying the Bessel equation, we substitute (3.4.69) into (3.4.43) to obtain 4H x=-. (3.4.70) 30 2 The initial conditions are
7](0) = 1,
7)(0) = O.
(3.4.71)
The solution of (3.4.70) that satisfies conditions (3.4.71) is
7](t)
= cn (O'lt, a)
(3.4.72) 1
0'1
=
Wo (l
+ x) t ,
a_ [ x ]} - 2(1 + x) ,
where the period of the solution T = 4K(a)/O'l' K(a) is a complete elliptic integral of the first kind; cn is the Jacobi elliptic cosine function. The dimensionless circular frequency of the natural vibrations is * W
=
21rO'l ---r-'
(3.4.73)
It would be more consistent to use expression (3.4.72) to approximate the time function in finding solutions of DEE-type, but the replacement of the elliptic cosine by the ordinary cosine (the first term of the Fourier-series expansion) makes it possible to simplify the notation considerably in this case without changing the esSence of the asymptotic method. Equation (3.4.70) can be solved approximately by the Bubnov-Galerkin method. The resulting dimensionless frequency agrees with the constant A determined earlier. The order of the procedures used to find the DEE can be changed: the solution of (3.4.43) can first be represented in the form (3.4.69), where k, and 0 are unknown constants; then 7](t) can be found as a function that depends on the parameters k and 0, following the application of the method described above to determine the DEE.
e
190
3. Continuous Systems
If u = 0 and () = 21r, formula (3.4.73) gives asymptotic values of the natural frequencies of the circular plate supported elastically around its contour. For a circular region with hinged boundary, (3.4.43) does not admit of an exact solution (the case A = 0 is an exception), in contrast to the case of the similarly restrained rectangular plate. The frequency calculation can be ma.de more accurate by including terms corresponding to the DEE in the expression for H. Thus, the DEE at a circular boundary is taken into account in the numerical example given below. Below we present calculated results for the ratio wNlwt for a contourrestrained circular plate; wN' wt are the first natural frequencies of the linear and nonlinear vibrations, respectively,as found by the asymptotic method (the first row) and by the finite element method (the second row) [102]: A = 0.2 1.0074 1.0072
A = 0.4 1.0291 1.0284
A = 0.6 1.0634 1.0624
A = 0.8 1.1083 1.1075
A = 1.0 1.1621 1.1619
Tables 3.1 and 3.2 give values of the dimensionless natural frequencies w* of a contour-restrained circular plate and similarly restrained circular sector, respectively; ml is the number of half-waves on the circumference and m2 is the number along the radius. In either case, the values of the derivative dw* I dA increase with increasing frequency, and more strongly with the derivative dw* IdA with m2 (ml fixed). For the sector plates, the analysis indicates that the influence of nonlinearity becomes weaker with decreasing angle e. Table 3.1 also includes the data from [148] (marked with asterisks), where the natural frequencies were foun~ by an integral-equation method, for plates with the angle () = 1r 12 and A = O. The comparison indicates satisfactory agreement of the calculated results for the first five frequencies. The largest difference does not exceed 2%. Table 3.1. Comparison of various approximate solutions A
0 0.5 1 1.5 2
ml -0 m2 = 1 10.22 10.67 11.87 13.56 15.61
0 2 39.77 41.16 44.98 50.57 57.39
0 3 89,10 91.47 98.12 108.1 120.6
1 1 21.26 21.73 23.06 25.02 27.45
1 2 60.83 61.78 64.52 68.77 74.20
1 3 120.1 121.5 125.7 132.3 141.0
2 1 34.88 35.57 37.53 40.48 44.12
2 2 84.58 85.78 89.22 94.59 101.5
2 3 153.8 155.5 160.4 168.2 178.4
191
3.5 Regular and Singular Asymptotics Table 3.2. Comparison of various approximate solutions
26.22 26.65 27.87 29.69 31.95
1 3 128.6 129.9 133.8 140.0 148.2
2 1 41.59 42.29 44.25 47.20 50.86
2 2 94.08 95.27 98.70 104.1 111.0
2 3 165.4 167.1 172.0 179.8 190.1
3 1 59.24 60.18 62.83 66.85 71.85
3 2 122.5 124.0 128.2 134.9 143.5
104.8 105.1* 105.9 109.0 113.8 120.1
178.4
167.5 165.3* 169.2 173.9 181.4 191.0
265.0 271.2 281.2 294.4
139.2 138.3* 140.7 145.1 151.8 160.4
239.6 356.1
180.0 184.4 191.5 200.8
89.57 88.13* 90.66 93.80 98.61 104.7
262.8
0.5 1 1.5 2
49.18 48.70* 49.82 51.62 54.34 57.72
241.8 359.1 248.0 367.0 257.8 379.6 270.6 396.3
0 0.5 1 1.5 2
77.95 78.78 81.14 84.75 89.28
148.6 149.8 153.5 159.4 166.9
236.0 237.7 242.8 250.8 261.4
152.7 154.1 158.3 164.9 173.3
257.6 259.6 265.6 275.1 287.5
378.1 380.7 388.3 400.4 416.4
248.0 250.0 255.8 265.1 277.1
387.9 390.6 398.7 411.4 428.3
A
1'r
0 0.5 1 1.5 2 0
'It"
2"
'It"
"3
=1 =2
1 2 67.85 68.72 71.22 75.15 80.24
(J
ml m2
3 3 204.3 206.3 212.1 221.4 233.5
541.4 544.8 554.6 570.4 591.4
3.5 Regular and Singular Asymptotics in the Nonlinear Dynamics of Thin-Walled Structures 3.5.1 Circular Rings and Axisymmetric Cylindrical Shells Normal modes for nonlinear spatial systems usually cannot be obtained on the basis of full equations. Below we show the use of straightforward asymptotic simplification for constructing simplified nonlinear differential equations (see also [70, 71, 75, 78, 103, 104]). The equations of free axisymmetric vibrations of an orthotropic cylindrical shell in projections onto the axes of an undeformed coordinate system have the form 2
- - -8M sinO-) - p 8u = o· -8 ( T I cosO 2 fu
m
fu
8 ( 8M -) 8x T I sin () - 8x cos 0
T2 +R
'
2 8 w + P EJt2
+ O.
(3.5.1)
Here p is the mass per unit area. We will write the geometric and elasticity relations in the form
8u
1 (8W) 2
+"2
CI
= 8x
TI
= BIcI;
W
-
8w
= R; 0 = arcsin 8x ; T2 = B2c2; M = DR. 8x
;
C2
R
80
= 8x;
Here, for the sake of simplicity, it is assumed that the Poisson ratio is equal to zero.
192
3. Continuous Systems
1 2;02
2 1;03
We will introduce the notation 01 = DIB R = B 1B = Hoi BI R2 (Ho is the initial energy level) and examine affine transformations of the coordinates leading to various limiting systems. U
= 03 RU ;
W
= 03 RW ; t
=
(
R4)1 2r.
PD
The transformed system has the form
2u a ( -1 a"".) a ae c cos 0 - 0 3 01 ae sm 0 - 01 ar 2 = 0; 2w a (. -1 a",,) a W ae csm 0 + 0 3 01 ae cos 0 + 02 + Ctl ar 2 = o.
Here £
~~ + ~<>3 (aa~) 2;
=
8
= arcsin ( <>3 aa~);
It
=
:~.
(3.5.2)
In the case 03 --+ 0, the limiting system describes nonlinear vibrations of a rod. When 02 --+ 0, it describes linear axisymmetric vibrations of a cylindrical shell. With 01 --+ 0,02 --+ 0, we arrive at the system
:e
(co cos ( 0 ) = 0;
:e
(co sin ( 0 ) = 0,
(3.5.3)
from which it follows that co = o. We will represent the sought functions with series of the small parameter 01 and write out the equations of the first approximation: ~.
2
a ( co cos 00 - 0 31 a""o ae ae sin 00 ) - aarU20 = 0; 2 a ( . -1 a""o ) a ae co sm 00 - 0 3 ae cos 00 - arw2o = o. Excluding the function
C;.
(3.5.4)
from (3.5.4), we obtain
a2uo 0 a2wo . ae ae - ar 2 cos 0 - ar 2 smOo 2wo 2 2uo . ) (ao o) -1] _ a [( -1 a ",,0 a a - - ae 0 3 ae 2 + ar2 cos 00 + 8r 2 sm 00 ae ,(3.5.5)
-1 0 3
a""o ao o
while from (3.5.3) we find
auo ae + "21 03 (aWo 7if )2 =
o.
(3.5.6)
Equations (3.5.5) and (3.5.6) constitute the limiting system corresponding to nonlinear vibrations of a flexible rod with its edges free in the axial direction. This system was obtained by using equations of the first approximation, which is an interesting feature of this case:
3.5 Regular and Singular Asymptotics 4
PR- ) t= ( D
193
~ r.
The transformed system has the form
8 ( 8e e cos () -
1/2
Q'1
_18K,.) 8e sm () -
Q'3
1/2 82Uo 8r 2 = OJ
Q'1
8 (. 1/2 _18K,) 82Wo 8e esm () + i l l Q'3 8e cos(} + Q'2 W + 8r 2 = O.
(3.5.7)
Here
_ 8U e - 8e
+
Q'
11Q'3 2
8(}
(8W) 2 • 8e '
K, = 8e;
With Q'1 --+ 0, Q'3 '" Q'1, we obtain 8eo 84Wo 82Wo 8e = 0; 8e 4 + co 8e 2 +
Q'2 WO
() =
. ( Q'3 Q' 11/2 8W arcsm 8e ) .
82Wo
+ 8r 2 + O.
(3.5.8)
The problem is linearized if constant axial forces are applied to the ends, and nonlinear effects can be revealed in the equations of the first approximation. Let the edge of a shell be fixed so as to prevent displacements in the axial direction. Then, the nonlinear effects in system (3.5.7) are preserved, since it follows from the first equation that
£0 = <>1 t f (8::;0) 1
l
3
2
2
d{,
o where l = l/R; l is the shell length. At the same time, with hinged ends in the second equation, the variables can be separated: .
tc
m1rX
= W m SIn -l-l3m(r),
m
= 1,2, ....
We obtain for the time function an ordinary differential equation containing a "rigid" nonlinearity which has the following form because of the initial variables ..
13m +
1 (B2
p
R2 +
4 4 ) B 1m 41r4 2 3 l4 13m + 4pl 4 w m f3m = O.
Dm 1r
(3.5.9)
Each of the functions (3.5.9) is integrated in elliptic functions. For example, with 13m (0) = 1,f3m(O)' 0, we have
13m = sp (Kmt, 8 m ),
(3.5.10)
where = m 2 1r 2D3/2 A 1/2 ( 1
K m
A=
4l2p1/2
B 1W m ) + A
4D (1+ D:'~:1r4).
2
•
'
2( 1 B w 2)-1 2A + A
= B 1W m
S m
1
m
194
3. Continuous Systems
Thus, the limiting system (3.5.9) has exact solutions corresponding to normal vibration modes. When 02 ---+ 0, (3.5.10) converts to the familiar equation for a rod with fixed ends: u
= 0I3RU;
W
= 0I3RW;
x
= R{;
t
= (p:;'2)
1 T.
The transformed equations of motion have the form
2 a ( -1 a",. ) a u ae c cos 0 - a 10 3 ae sm 0 - ar 2
:e ("Sino + 01101 1~~ coso) + 3
= 0;
0I2W +
a~ = 0,
where c, x, 0 are defined by (3.5.2). With 01 ---+ 0 we arrive at the limiting system
a a2 u ae (ccosO) - ar2
= OJ
a. a 2w W ae (csmO) + 02 + ar 2
= 0,
(3.5.11)
corresponding to vibrational tension. However, by virtue of the condition c « 1, system (3.5.11) turns out to be consistent only in the quasilinear case (03 ---+ 0, 03 ,...., ad. Let us examine plane nonlinear vibrations of a circular ring (a cylindrical shell of infinite length). We will write the equations of motion in a local coordinate system, the axes of which correspond to the tangential and radial directions at a point on the undeformed axis of the ring:
aT 1 aM ay - R ay -
a2v p 8t 2 = 0;
1
a 2M i -. a 2w a y 2 + R T + 8t 2 = 0,
(3.5.12)
where
T
=
- aM TcosO - -sinO' ay ,
aM ay
=
- aM TsinO + ay cosO;
and p is the running mass. We will write the physical and geometric relations in the form T
= En·, M
=
E..jXj
,arcsm. (aw - - -V) ay R '
ae
K,= - j
ay
o=
Let 01 = JI F R 2 ; 03 = HoiEFR. Subjecting the variables V, w, y, and t to affine transformations, we obtain the following classes of nonequivalent systems: v= 0I3RV;
W= 0I3RW ;
y=R''1;
t=
(~)' T.
3.5 Regular and Singular Asymptotics
195
The transformed equations have the form
8 (
-
a.,., 8 (
a.,.,
2
_ 81'\,.). _181'\, 8 V cCOS(}-0103 1- sm (} -csm(}-01 0 3 -8 COS(}-012 =0; 8r 8T1 TI 2
.
181'\,) _181'\, . 8 W c sm () + 01°3 8T1 cos () + c cos () - °1°3 8rJ sm () + 01 8r 2 = O.
Here c
= 8V + W + ~03 (8W _ V) 2; 8r
0= arcsin
I'\,
8T1
2
=
8(};
8T1
(0;: - v) .
(3.5.13)
With 03 --+ 0, we obtain the limiting system corresponding to the linear theory. The second limiting system is obtained as 01 --+ 0:
~ (co cos ()) -
co sin ()
~ (co sin ()) -
= 0;
co cos ()
= O.
(3.5.14)
The condition of nontensionability follows from (3.5.14):
8Vo 8T1
1
+ Wo + 2°3
(8Wo 8T1 - 1fo
)2 = o.
(3.5.15 )
We will write out the equations of the first approximation:
8 a.,.,
( c1 cos ()0 _1 -°3
:f/
1 81'\,0
. ())
° 3 8T1 sm
. ()
c - 1 sm
0
+ "3 1 ~~ cos 00 ) + 01 cosO -
0
-
0
-
8 "'0 8 2 1fo 8T1 cos (}o - 8r 2 = 0;
(0 1 sin 00
_18"'0 . 8 2 Wo -°3 8T1 sm(}o+ 8r 2 =0.
(3.5.16)
After making the obvious transformations, we may exclude the functions C1 from (3.5.16). Considering condition (3.5.15), we obtain the following relations connecting the functions Vo and W o: 1 (8Wo -8Vo + W o + -03 ._- - Vo)2 = 8T1 2 8T1
O· '
2
_181'\,0 ( 8(}0) 8 Vo 8 2 Wo . °3 8T1 1 + 8T1 + 8r 2 cos (}o - 8r 2 sm (}o
=-
~ [("31~~ +~: sin 00+8;:'0 cos 00) (1+:
(3.5.17)
r1
The nonlinear system (3.5.17) is fairly complex, so it is expedient to take the following for the limiting linearized system (01 --+ 0, 03 --+ 0)
196
3. Continuous Systems
4 4 2 2 6 8 Vo + 2 8 Vo + 8 VO, _ 8 VO + 8 VO = 0, (3.5.18) 817 6 817 4 817 2 8r 2 8r 28172
8Vo + Wo = OJ 81]
and to evaluate the nonlinear effects by using the equations of the first approximation (W = W o = 0'3 WI = ... j V = \'0 = 0'3 VI"')
:
+ W,
=
-H&~o
-Vo) ; 2
8 6 VI 8 4 VI 8 2 VI 8 2 VI 84Vl -+2+ + 2 2 2 2 4 81]6 817 817 8r 8r 8." 2 __ 8 2W O 0 + 8"'0 800 _ ~ (8 /\,0 800 8r 2 0 817 817 817 8rp 817
(3.5.19)
+ 8 2W O 800 _ 8r 2 817
2 8 VO ( ) . 0 8r 2
The n-th mode of free vibrations is a particular solution of system (3.5.19): 2
iW n7", TJ' _ 1 . iW n 7". YYIOn=wOncosnne ,von--wonsmn17e , Wn ' ,'f 'n'
UT
_
-
n (n
2
-
n+ 1
1)2
Then, the solution of system (3.5.19) satisfying the periodicity conditions has the form V;
l,n
2 =.!!7" • 2n wO,n sin 2nne2iwn 'f ,
W 1,n = [ '14 ( n - n1 ) 2 -
B]
2 cos 2n17e 2iw n 7"
(3.5.20)
wO,n
where B =
~ [(n 2 + 1)(4n2 -1) _ 4n 2 + 1]-1 2
(n 2 - 1) 2
4n 2 - 1
Thus, in the case of nonlinearr vibrations, the n-th harmonic with respect to the coordinate 17 is accompanied by a zero harmonic W
= 0'3RWj
The limiting (0'1
v = 0'11/2 0'3 Vj --+
0,0'3
rv
O'd
Y = 0'11/2 R17j
t = (PR EJ
4 )
! r.
system has the form
1 2 2 8V + W 0'1 0'3 (8W)2 = O' 8 W 8 W 817 + 2 817 '8172 + 872 = O.
(3.5.21)
The second equation of (3.5.21) is linear, and the nonlinearity of the problem is determined by the first equation of (3.5.21). With certain additive terms, this system was used earlier to investigate nonlinear vibrations of a ring [70, 71, 78]. It turns out that these additive terms playa minor role in investigating the given type of vibrations and are of the same order as the discarded terms.
3.5 Regular and Singular Asymptotics
197
The particular solutions of system (3.5.21), which satisfy the periodicity conditions, have the form Wn
={
Vn = -1 {
2 WUl,n c?snT] } eiwn'T _ W20',n sm nT]
, sin nT] } e iw n 'T
-WlO n
n
W20,n
cos nT]
n
WI0,n } e2iwn'T W20,n
01103 {
4
+ -n 0 -1 03 { 8
1
W~o2n 'sin 2nT] .
-W20,n
sm 2nT]
The axisymmetric component of the radial displacement coincides with the nonlinear correction for Wo determined from (3.5.21): W
= Q3RW;
v = Q3RV;
Y = R'I;
t
=
(i.:~)
1 T.
With 01 -+ 0, this transformation makes it possible to obtain the following limiting system: 8(ccosB) _ 8T]
2
. B _ 8 V _ O. c sm 8r 2 - ,
2
8(csinB) _ ccosB _ 8 W 8T]
=0
8r2'
where €, K, and 0 are defined by (3.5.13). Similarly to the case of axisymmetric vibrations, this system is consistent at low amplitudes (03 -+ 0, 01 "" (3)' Nonlinear effects may be revealed while analysing higher approximations.
3.5.2 Reinforced and Isotropic Cylindrical Shells Asymptotic methods are extensively used in the theory of thin shells. The asymptotic analysis of the basic equations of the theory of isotropic cylindrical shells has been carried out in, for example [65, 68, 69, 82, 124, 131, 132, 138, 29d]. The results of these studies have been extended to the dynamic case [83, 136], to nonlinear shells [69, 157] and to orthotropic shells [2]. Finally, it is shown in [56d] that in the case of a structurally orthotropic shell the presence of a large number of geometrical rigidity parameters leads to additional possibilities of asymptotic integration even though it complicates the analysis. As a result a number of new approximate equations have been obtained for the main types of reinforced shells, which have no analogy in the isotropic case. In the present book simplified boundary value problems are formulated for new types of approximate equations of the theory of nonlinear dynamics of eccentrically reinforced cylindrical shells. It is worth noting that ribbed shells were investigated in many papers [49, 162, Id] etc. The nonlinear dynamic boundary value problems of the theory of closed circular cylindrical shells eccentrically reinforced in the two principal directions are investigated within the framework of the structurally orthotropic scheme. The middle surface of the shell is chosen as the main one. Detailed discussions of the basic relations of the linear theory of shells can be found
198
3. Continuous Systems
in many monographs and papers [51d, 55d] a.nd therefore we discuss only the final results here. The governing equations of motion are written in the form proposed by Sanders [140] (in comparison with the original equations from [140], some dynamical terms are added): 2 8Nll 8N12 ....!..- 8M12 _ ~ ~ [p(N N)] _ R 28 ul = 8Xl + 8X2 + 2R aX2 2 aX2 II + 22 P 8t 2 '
°
8N 8X2 1 8 +--8 [p(Nll
1 8M12 2 Xl
8N 2 8Xl
22 1 -+ -- Q2 - -R a 2 Xl
8Ql 8Xl
+ N22)]- pR
+ 8Q2 + N22 - a8
8X2 8 - 8X2 (Pl N12
+ (P l N12 + P2 N 22)
Xl
(PlNll
282u 2 !U2
U~
= 0,
+ P2 N 12)
(3.5.22)
2a2w
+ P2 N 22) - pR 8t 2 = 0,
°
°
8Mll - + 8M12 - RQ 1_- , 8M12 + lJM22 _ RQ 2- - ·
aXl aX2 aXl aX2 Here N ll , N 22 , N 12 are the membrane stresses; M ll , M 22 , M 12 are the bending and torsion moments; Ql, Q2 are shearing forces; PI =
_.!.. 8w
P _
R aXl'
2 -
-
1 (8W u ) R aX2 - 2 ,
P _ 1 (8U2 8U l ) - 2R aXl - aX2 '
U2, ware the tangential and normal di{;placements; p = Po + If- + 'l;; Po, PI (P2) are the density of the shell material, and the stringer (ring~ material. It is worth noting that Sanders [140] defined the variant of "moderately small rotation" by setting restrictions on the components of the linearized rotation vector to the effect that the squares of these components can be at most of the order of magnitude of the strains. The components of the elasticity tensor are defined by [114] Ul,
= B llcll + B12c22 + KllKU, N22 = B 2lcU + B22c22 + K 22 K22' M u = DUKU + D12 K22 + Kucu, M22 = D 2l K U + D 22 K22 + K 22 c22,
N ll
N 12
M 12
EsF ErFr = B + --' II ' l2' B 2l = B 12 = V2l B U = V12 B 22;
' B Bu =
Ku D
=
~~
~~
K22 = -l2-;
Eh 3 = 12(1 - v 2 );
D
D33
=
Du
=
EaJks
= D33 K12,
~~
D + -ll-;
"2 + h +
D 2l = D 12 = V4 l D u = V14D22 = D;
(3.5.23)
Eh B . - 1 - v 2'
+ - -s' B22
-ll-;
= B33c12,
ErJkr 12
;
Eh B 33 = - - , '
D22 = D +
1-
V
~~ T;
3.5 Regular and Singular Asymptotics
199
Fs(Fr), J s( J r), Jks (Jkr), Ss(Sr) are the transvese section area, inertia moment, rotation inertia moment, torsion inertia moment and static moment of the stringer (rib); E, Es(Er) are the elasticity modulus of the shell material. The following geometrical relations are accepted here [140]: 1 aUl
1
C12 =
2
1 (aU2 2R aXl
1 ( aU 2 8 X2 -
2
+ 2tPl + tP;
= R 8Xl
cll
= R
C22
R aX2'
2R
aXl
aX2
= 0;
Ul
(3) w = 0;
2
2 ;
(3.5.24)
(Xl
= 0.1, e = L/ R, L is the
aw = Wxl
(4) -
= 0;
Xl
(6) N ll = OJ
(5) N12 = 0;
1
+ 2tP2 + tP
- tP) •
We assume that at the end faces of the shell length of the shell) the following are specified: (2)
)
au l ) 1 1 8
~22 = ~ atP2 . ~12 = ~ (8
(1) U2 = 0;
W
(7)Ql
= 0; (8) M ll = O.
(3.5.25)
Below we shall denote any variant of the boundary conditions by the symbol Gijn, where the set of indices corresponds to the numbers of the specified boundary conditions. Let us introduce the small parameter
Dl
Cl
= ( B 2R2
)1
This parameter will be used for the estimation of orders of magnitude of various terms and parameters. Let us also introduce the dimensionless geometrical rigidity parameters
Dl
C2
= D
2
D3
;
C3
B3
=D
l
B2
;
C4
=B
K ll
l
;
K 22
= B l ; C6 = BlR; C7 = B 2R; B l = Bll(l- V12 V2l); B 2 = B 22 (1 - V12 V2d; 1 1 B2l
C5
B3
D1
-B 33
= Du
D 3 = D 12
-
B ll B 22 (1 K?l B ; l
D2
.
V12 V2l)'
= D22 -
K~2 B ; 2
K ll K 22
+ D 33 + B l B2 B 12(1 - V12 V2d·
Depending on the assumed estimates of the value of these parameters, three types of reinforced shells can be distinguished:
200
3. Continuous Systems
- stringer shells (SS) Cl « 1, C2 ,...., c~, C3 - ring stiffened shells (RS) Cl « 1, C2
,....,
ell C4 ,...., c5
< 1, C6
,...., GIl, C3 ,...., Cl, C4
,...., Cl, C7"""
< 1,
C6 ,....,
0; 0,
1/2 c7 ,...., cl ;
- "wafer" shells (WS)
Cl
«: 1,
C2 ,....,
1,
C3
< 1,
c4 ,....,
1,
c5
<
C6 ,...., c7 ,...., Cl·
Derivatives of various components are estimated in the following manner
a aXl
-
-at
,...., Cl
;
a 8X2
-a2
-,...., cl
;
We also introduce the parameters
w ,....,
R
Ul""" cr sw ;
cr4;
a at
-03
-,...., cl Ok
(k = 4 - 6) through the relations
U2""" crew.
It is known that two-dimensional equations of the theory of shells are valid provided that the following estimates are satisfied
o < Ok <
1,
k
= 1,2,3.
Now let us briefly describe the asymptotic procedure (for details we refer readers to [83, 29d]). We pose an expansion for any components of the desired stress-str.ain shell state U (3.5.26) Substituting ansatz (3.5.26) into the governing boundary value problems and comparing the coefficient of c~, we conclude that the limiting (cl --+ 0) systems are strongly dependent on the vah~ of the parameters Ok. Now we must scan all possible Ok and search all s~nsible values of these parameters, for which the limiting systems make mathematical (well-posed) and physical sense. It is remarkable that as a result of this routine but very laborious procedure we obtain only a few limiting systems which are analysed below. Now let us consider the possible simplifications of the general relations of the reinforced shells, which r~sult from the previous assumptions. As a result of the asymptotic procedure, one obtains the following limiting systems (1) 02* < 2' 1 * * * 0 , 04= * 202' * 01=02, 03= Here for WS and SS shells ok = Ok and for RS
* * * 05=06=02'
(3.5.27)
3.5 Regular and Singular Asymptotics
201
and
(3.5.28)
Equations (3.5.27) describe the nonlinear membrane motion. 1 * 03* = - 1 + 202' * 04* = 202' * Os* -- 2' 1 (2) 02* < 2' 01* -- -21 + 2 02' £ WS·,U;I ",* - 21 + '" 06* = 02*( ok* - Ok* lor u;I, 02* = 02, 03* = - 1 + 03, 04 = 04, = os, 06 = 06 for SS; 0i = -~ + 202 + 01, 02 = 202, 03 = -~ + 202 + 03, ok = 20 k , k = 4,5,6 for RS).
Os
(3.5.29)
=0,
Nll
= BUell + B I2 e 22,
0
= B 2I e ll + B2 2e 22 + K221'\,22 ,
N I 2 = B 33 eI2,
(3.5.30) 1 aUI ell = R OXI
+ K 22e 22,
M 22 = D 22 1'\,22
U 2 _ w) + ~4>~; 0=2.R (aaX2 2 1 aw
4>1
= - R aXI '
0=
1 aw
4>2
1
2
+ 24>1;
-2.... ( aU 2 + aUI) + ~4>I4>2; 2R
OXI
U2
= - R aX2 - R'
OX2
1
4>
(8u 2
2
aUI)
= 2R aXI + OX2 .
Equations (3.5.29)-(3.5.30) are the nonlinear quasimembrane shell motion equations (only bending moments in the circumferential direction are considered) without tangential inertia. These equations may be obtained as a result of the reduction of general relations if it is assumed that the following relations are satisfied e22
= 0,
eI2
= O.
These relations denote physical conditions of extension in the circumferential direction and the absence of shear in the middle shell surface.
202
3. Continuous Systems
For ok > 0 it is possible to omit the term in brackets in relations (3.5.27)(3.5.30). The equations of membrane and quasimembrane vibrations are of fourth order with respect to the axial coordinate Xl, and they can be satisfied by two boundary conditions on every shell edge only while integrating the corresponding limiting systems. From the point of view of singular perturba.tion theory we deal with outer solutions and we must construct inner solutions (boundary layers) [83, 120, 122, 29d]. The boundary layer solution has a large variability index in the Xl direction, and its variability in the circumferential direction and in time is the same as with the inner solution. Now let us present all the stress-strain state components U as follows =
U
U(O)
+ U(k) ,
(3.5.31)
where the indexes (0) and (k) indicate the components of the outer solution and the boundary layer respectively. It is necessary also to introduce the parameter v characterizing the order of w(O) with respect to w(k): w(k) '" crw(O).
The value of the parameter v and the boundary layer variability in the Xl direction depend on the boundary conditions and are defined in an asymptotic, splitting process. As an illustration of the method used we consider the boundary conditions for the variant G~~. First of all, we write asymptotic orders of the components of the boundary conditions for the inner solutions u(O) '" 1
c- 1/
2W(0)'
aw(O)
1/2-2a;w~)
' a Xl '" c 1
'" 1
3/2-a;
N(O)
;
12
'" C 1
(0). W,
(3.5.32) U(k) '" 1
N(k) 12
c 1/ 2W(k)
'" c 1/ 2 + v W(0).
1 c- 1/
'" '" 1
1
2- a ;
W
(k)
aw(k)
'aXl
1/2-a;+v (0) '" c 1 W;
-1/2 (k) '" c1 w
W
(k)
'"
-1/2+v c1
(0). W,
v (0) '" c 1 W .
We choose the value of the parameter v from the condition of the absence of a contradiction in the limiting boundary value problems (in other words, the number of the boundary conditions for the limiting system must coincide with the order of the differential equation with respect to xd. In the case under consideration the unique possible value of v is v = 1 - 20 2 > O.
(3.5.33)
Let us emphasize that the boundary layer nonlinearity order estimation results immediately from (3.5.33) w(k) '"
cl R .
3.5 Regular and Singular Asymptotics
203
Estimating (3.5.32) and taking into account (3.5.33), we find w(O) '" c~(1-2a;)w(k) (0)
Ul
»
w(k),
-(1-2ai) (k) --..
'" cl
u l
8w(0) '" 8w(k) 8Xl
(k)
~ u l
,
N(k)
N(O) 12 '"
8Xl'
12'
Then, the splitting of the boundary conditions may be represented in the following form W (O) 12:1=0,l --
u(O) 1 12:1=0,l --
8W(k) 8Xl
o·,
8w(0) 12:1=O,l
=
8Xl
Izl=o,l;
(3.5.34) (k) 1 _ (0) I N 12 zl=O,l - - N12 Zl =O,l·
(3.5.35)
Therefore, for the outer and inner solutions, boundary conditions (3.5.34) and (3.5.35) must be, respectively, given. In the same way other boundary conditions are split. The results are presented in Table 3.3. To obtain the boundary layer equations, let us take (3.5.31) into the initial equations and take into account the outer solution equations (3.5.33)(3.5.34). Table 3.3. Splitting of the boundary conditions for the half-membrane state and boundary effect
c
Splitting boundary conditions w(O)
12
C34
,
8u~k)
U(l)
w(k)
'Xl'
+ ...l.. (8UJ(k») 2 = L(k) 2R 8X2 1 w(O) w(O) L (k) M(O) + M(k) , Xl' l' 11 11 W(O) N(O) L(k) w(O) + w(k) , 11' l' Xl Xl W(O) N(O) L(k) M(O) + M(k) , 11' 1'11 11 W(O) u(O) w(k) N(O) + N(k) , l' Xl' 12 12 W(O) N(O) w(O) + w(k) N(O) + N(k) , 11' Xl xl' 12 12 w(O) w(O) N(O) + N(k) M(O) + M(k) , Xl' 12 12' 11 W(O) N(O) N(O) + N(k) M(k) + M(k) , 11' 12 12' 11 11 w(O) u(O) Q(k) w(k) + w(O) , l' l' Xl Xl w(O) N(O) w(O) + w(k) Q(k) , 11' Xl xl' 1 w(O) w(O) M(O) + M(k) Q(k) , Xl' 11 11' 1 w(O) N(O) M(O) + M(k) Q(k) , 11' 11 11' 1 u(O) N(O) w(O) + w(k) Q(k) l' 12' Xl Xl' 1 N(O) N(O) w(O) + w(k) Q(k) 11' 12' Xl Xl' 1 u(O) N(O) M(O) + M(k) Q(k) l' 12' 11 11' 1 N(O) N(O) M(O) + M(k) Q(k) 11' 12' 11 11' 1 _
w(k)
8X2
C 38 12 13 C46 13 C68 23 C 45 34 C 56 23 C 58 35 C68 f2
C 47 14 C67 12 C 78 16 C 78 24 C 57 45 C 67 25 C 78 56 C 78
1~
204
3. Continuous Systems As a result, we get the following limiting systems *
02
*
1 = 2'
0 1
1 = 0 ,v * = 1 ,05* = 2'
0 3
*
06
= 1-
ok = Ok,
(k = 1 - 6), v* = v for WS and SS; = Ok, (k = 2,3,4,6), v* = v - 1/2 for RS. The equations of motion have the form Here
ok
*
1 < 2'
N(k) 8 _..;;..;11 ..... 8Xl
+
8 2 M~~) 2 8 Xl
8N(k) 12 = 8X2
+
0 '
8N(k) 12 8Xl
2 (0) 8 w(k) N ll 8 2 =
(k) RN22 -
0,
0;
(3.5.36)
N~~) = B2lC~~) + B22c~~), M~~) = Dll~~~) + Knc~~) j
N~~) = B33C~~) ,
1
8u(k) 1 (8 2W(k)) 2 + _ 1 8w(0) 8w(k) ; l_+_ R 8Xl 2R 8xr R2 8Xl 8Xl
1
(k)
1 8w(0)
w(k)
c22 = -
R + R2
aw(k) .
8X2
8X2'
(k 1 (8U~k) 8U~k)) c )-- --+-12 -
oi = 20 1, Os = 20 5,
Xl
o = Bl1c~~) + B12C~~) + Kll~~~) c(k) 11 -
8N(k) 22 = 8X2
+
*
02'
2R
8X2
8Xl
1 +_
(8W(0) 8w(k)
R2
8w(k) 8w(k)
+ 8w(0) 8w(k)) I .
8X2
8Xl
1
+ ---R2 8Xl 8X2 8X2
8i l
'
(k) _ 1 8 2 w(k) ~ll - - R2 8x~ .
We add equations (3.5.36) to conditions (3.5.35) and get a well-posed approximate boundary value problem to satisfy the boundary conditions. In (3.5.36) ihe variable coefficients, which are obtained by the outer solution term, can be "frozen" with respect to Xl on the shell edges. This is valid because the outer solution variability index in the axial direction is much smaller than the boundary layer index, and in the zone localized near the shell edges, the inner solution may be assumed constant with respect to the X 1 variables. We can limit the equations for the boundary layers, and thus give the possibility of satisfying the boundary conditions, which differ from (3.5.36) by the presence of the inertial term in the third equation of motion. The above derived sets of equations are sufficiently accurate to describe the bending state in a shell. Let us formulate simplified WS boundary value problems: 01
=
1
02
= "2'
03
= 0,
04
= 1,
05
= 02
=
1
"2'
3.5 Regular and Singular Asymptotics
205
In this case the limiting equations are equal to well-known equations for the shallow shell theory. One has to extract the parameter cs for these shells of this class from geometrical-rigidity parameters, which are the tangent rigidity and the extension-compression ratio. If reinforcement is strong enough, these values are small. Let us introduce the parameters of asymptotic integration Qk (k = 1- 6): a
-01
-
cs
f'V
a aX2
;
aXl W
R
Q4
f'V
cs
.
U6
Cs W,
f'V
a
j
-
-03 cs ;
f'V
at
05
til
;
-02
Cs
f'V
06
f'V
Cs
w.
As a result of the asymptotic procedure we get the following limiting systems, which do not have an analogue in the isotropic case
(a)
Q1
aN(l) _....::.:ll~ aXl
= Q2,
+
= Q4 = 2Q2,
Q3
aN(l) 12 aX2
a 2 M(l)
_~ll=- + 2 a Xl2
aN(l) 12 aXl
= 0,
a 2 M(l) 12 a Xl a X2
+
D
a
(1) 1~ (1) _ _ cll - R aXl (1) _ c12 -
+2
-.!... (aU~l) 2R
aXl
aN(l) 22 aX2
(3.5.37) a 2 w(1)
( )
D
pR 2 _......,...._ = 0', a t2
-
aU~l)) aX2
(1) 22 K 22
Ng) = B33C~~),
(1). + K 22c22 ,
M(l) 12 -
(a (1)) R +
2 (1) _ (ifJ(l)). 1 ,c22 -
+
= 0,
+ RN221
M(l) 22 -
I
~
= Q6 = Q2;
0 = B22C~~) + K22K~~,
(1). + K llcll
(1) llK ll
+
a 2 M(l) 22 a X2 2
o = Bllcg) + KllKg) , (1) M II -
as
~
~ aX2
_
(1)
W
~ 2
D
(1). 33 K 12 I
2. (ifJ(l)) 2 ,
~A;,(1)A;,(1) + 2 '.i'1 '.i'2 .
The stress-strain state of a WS is approximately described by an equation of three types. U nUke in the case considered above, in this case it is necessary to introduce not one, but two parameters VI and V2 characterizing the ratio of the order of magnitudes of the quantities defining each of the three states W(l), w(2), w(3): w(2)
f'V
c~lw(l),
w(3)
c~2W(1).
f'V
(3.5.38)
The VI and V2 values are defined in every case by the boundary condition splitting process. Moreover, it is possible to have the following VI and V2 values 3 3 VI
= 2j
V2
= 2;
VI
= "2 j 5
= "2;
V2
=
VI V2
2;
= 2;
VI
= 2;
V2
= "2'
5
The corresponding limiting systems are:
(b)
Q1
= Q2 -
!
I
Q3
= 2a 2,
VI
= ~ (VI =
2), Qs
= Q2 - ~,
a6
= a2;
206
3. Continuous Systems
(3.5.39)
Ng) = B 11 c Ng)
= B 33 cg),
"'(121) _
2.. aui2) .
<;.
Q1
R
-
(2) ~22
(c)
l;) ,
= - R2
= Q2 + ~,
+
Mg) (2) _ c22 -
aXl'
1
aN(3) --:-..::.:11:.... aXl
NJ~) = B 2l cg) + B22c~~),
= D22~~~);
2.. (au~2) R
aX2
_ W
(2)).
,
a2w (2) . aX~
Q3
aN(3) 12 aX2
,
= 2&2, v2 = 2 (V2 = ~), =0 '
aN(3) 12 aXl
+
aN,(3) 22 a X2
Q5
= 0,
= Q2 - ~, a 2 M(3)
~l
&6
= 0;
= Q2 -
2;
(3.5.40)
aX l
The final results for splitting boundary conditions are given in Table 3.4a. It is worth noting that (3.5.39), (3.5.40) are everywhere linear, and this is sufficient simplification for the solution of practical problems. For stringer shells we have the following limiting systems ,
1
02
= 2'
01
= 0,
03
= -1,
04
= 1,
05
= 1,
06
=
1
'2; (3.5.41)
3.5 Regular and Singular Asymptotics
207
Table 3.4(a). Splitting of boundary conditions "Wafer" shells
C 12 C 34
13 C 46 12
C 38 13 C 68 23 C 45 C 34 56 23 C 58 C 35 68 12 C 47 14 C 67 C 12 78 16 C 78 24 C 57 45 C 67 25 C 78 56 C 78
w(l)
+
W(l)
+
3
U(l) u(2) u(l) u(3) 1 l' 2 2 w(l) w(l) u(l) u(2) N(l) N(2) N(3) , Xl I 2 2' 11 11 11 w(l) M(l) u(l) U(2) U(l) u(3) , 11' 1 l' 2 2 (1) M(l) N(l) N(l) N(3) U (1) u (3) W , 11" 12 12 11' 2 2 (1) N(l) N(2) N(3) U(l) U(2) W(l) W • 12 12 12' 1 l' Xl (1) w(1) N(l) N(3) N(l) N(2) Xl W , '12 12' 11 11 W(l) M(l) N(l) N(2) N(3) U(l) U(2) , 11' 12 12 12' 1 1 (1) M(l) N(l) N(3) N(l) N(2) W , 11' 12 12' 11 11 Q(l) W(l) U(l) U(2) U(l) u(3) l' Xl' 1 l' 2 2 Q(l) w(l) u(l) u(3) N(l) N(2) N(3) l' Xl' 2 2' 11 11 11 Q(l) M(l) U(l) u(2) U(l) u(3) 1'11' 1 l' 2 2 0(1) M(l) N(l) N(l) N(3) u(l) u(3) l' 11' 11 11 11' 2 2 Q(l) N(l) N(2) N(3) u(l) u(2) w(l) l' 12 12 12' 1 l' Xl (1) 0(1) N(l) N(3) N(l) N(2) W XI , l' 12 12' 11 11 Q(l) M(l) N(l) N(2) N(3) U(l) U(2) l' 11' 12 12 12' 1 1 0(1) M(l) N(l) N(3) N(l) N(2) l' 11' 12 12' 11 11 ,
Xl'
+
+
+
+ + + + + + + + + + + + +
+
+
"2
+
+
+
2 3
"2
+
2
+
3
"2
+
+
+ +
+
+
+
2
+
3
"2
2 3
"2
+
2 3
"2
+
2
+
3
"2
+
+
2 2 2 2 5
"2
2
"2
3
+
5
"2 5 "2 5 "2 5 "2
"2 5 "2 5 "2
2
+
2 2 2 2
5
(1) (1). _ B (1) B (1). + BI2c22 + K 111'\,11' 0 - 2Icll + 22C22 , D (1) K (1) (1) (1) N I2 = M ll = 111'\,11 + llcll; (1) (1) (1) M(I) . D (1) M 22 = D21~1l + D 22 1'\,22 ; 12 = 331'\,12; _1_ (aw(I))2 (1) _ ~ au~I) cll - R aXI + 2R2 aXI (1)
N ll =
_ 1 0-R
(1) B UCll (1) B 33 c I2'
(1 )
(au~I) 1 ---w ) +aX2
2R2
(1) au(I)) 0 - 1- (au _2_+_1_
-
2R
(8w(I))2
aXI
1 aw(I) aw(1) . +_
aX2
2
(1) _ 1 a w(1). 1'\,11 - - R2 a 2 ' Xl
aX2
2R2 aXI
aX2'
1 a 2w(I).
(1) _
1'\,22 - - R2
(1) _ 1 a 2w(1) 1'\,12 - - R2 a a .
a 2 , X2
Xl
X2
Equations (3.5.41) describe the dynamic shell state which varies rapidly in the circumferential direction. The boundary layer varies rapidly both along the guide and the generatrix and it is defined by the system (1) 02 = 01 = ~, 03 = -1, v = 1 (or v = ~), 05 = 06 = ~;
!'
aNi~)
_...:.:.... + aXI
aN(2) 12
aX2
= 0,
aN(2) 12
aXI
+
aN(2) 22
aX2
=0
'
(3 5 42) ..
208
3. Continuous Systems a 2M(2) axi 1
Ng) (2)
N 12
a 2W(1)
(2)
+ RN22 + aX~
(2) N 22
= 0;
= Bl1c~;) + B12c~~ + Kl1~~;); (2) = B 33c12;
~ aUl2) .
(2) _ cll -
R a Xl
I
M(2) 11
(2) _ C22 -
N~~} =
B 2l cg)
+ B22c~;);
(2) (2) = D ll~ll(2) + D 22~22 + K llcll j
2.. (au~2)
_ (2)) Ra W X2
~ aW(l)
+ R2
a X2
aW(2) . a X2 I
~ (aU~2) + aUl2)) + _1_ aw(2) aw(l) .
c(2) _
2R
12 -
aXl
aX2
(2) _ 1 a 2w(2). ~ll - - R2 ax~ I
2R2 aXl
aX2
I
(2) _ 1 a 2w(2) ~22 - - R2 ax~ .
The final separations of the boundary conditions are shown in Table 3Ab. Table 3.4(b). Splitting of the boundary conditions C 12 C 34
13 C46 12 C 38 13 C68 23 C 45 34 C 56 23 C 58 35 C 68 12 C 47 14 C67 12 C 78 16 C 78 24 C 57 45 C 67 25 C 78 56 C 78
Stringer shells
i;) _ w(k) = L(2) 2 W(1) W(1) L (2) N(1~ + N(2) l' 11 11: W(1) U(1) L(2) M(1) + M 2) • l' l' 11 11 w(1) M(1) L(2) N(1) + N(2) , 111 1'11 11 w(1) L (2) w(1) N(1) + N(2) , 2' 12 12 W(1) N(2) w(1) N(1) + N(2) , 12 1 1 11 11 W(1) U(1) N(2) M(1) + M(2) 1 l' 12' 11 11 W(1) M(1) N(1) + N(2) N(2) 1 11' 11 11 1 12 W(1) L (2) W(1) Q(1) + Q(2) 1 2' 1 1 W(1) Q(2) W(1) N(1) + N(2) , l' 11 11 w(1)
, ,
L(2) w(k) 8u 1 1 Xl' 8x
Xl
Xl'
Xl' Xl
Xl'
Xl'
M(1) 1- M(2). Q(2) 1 Xl 1 11 11' 1 w(1) M(1) N(1) N(2) Q(2) , 11' 11 11. 1 u(1) N(1) N(2) W(1) Q(1) l' 12 12 I Xl 1 1 N(1) N(2) w(1) Q(1) Q(2) 11 1 12 I Xl' 1 1 u(1) M(1) N(1) 1- N(2) Q(1) l' 11' 12 12' 1 M(1) N(1) N(1) N(2) Q(1) 11' 11' 12 12' 1 w(1)
1
-
W(1)
+
+
+
+
+ Q(2) 1
.
(
1 1 1 ~
2
1 1 1 3
"2 1 1 1 1 3
+ Q(2) 1 + Q(2) 1
"2 3 "2 ~
2
(3.5.43)
3.5 Regular and Singular Asymptotics
o=
N ll = KllK.ll' N 12 = B33c12, M22 = D 22 K.22;
+ B22c22,
B2lcll
M ll =
209
DllK.ll;
M 12 =
D 33 K.12.
The stress-strain relations are defined in this case by (3.5.41). Equations (3.5.43) describe mainly the bending vibrations of the shell. (3)
02
~,
>
aNll
01 = 02, 03
+ aN12 = 0,
8N12
8X2
8Xl 2 8 M ll 8 Xl2
8Xl
(8W -a NIl
8 + -a Xl
28
Xl
1
= 1,
05
= 06 = 1- 02·
+ 8N22 = 0,
(3.5.44)
8X2
8w)
8 + -a N 12 + -a X2 X2
(8W 12 -a N Xl
8W ) + -a N 22 X2
= OJ
1
8Ul
(8W ) aXl
= R aXl + 2R2 1 (8U2 2R 8Xl
C12 =
04
2
w -pR 8t 2
cll
= -1 + 202,
1 8U2
2 c22
j
1
( 8w )
= R aX2 + 2R2 aX2
2
1 8w 8w 8X2'
8Ul)
+ 8X2 + 2R2 8Xl
The formulae for the components of the elasticity tensor do not change in comparison with the governing equations. Equations (3.5.44) correspond to the SS vibrations with higher frequencies than in the previous case. Here we have large order variability both in the circumferential and axial directions of the outer solution (w(l)) and the boundary layer (w(2)). Let us introduce the parameter v: w(2)
rv
crw(l).
The asymptotic investigation shows that both the states are dynamic, therefore the splitting of the boundary conditions will not be unique. There are two correct splittings of the boundary conditions represented in Table 3.4c. For example let us represent the limiting systems for the variant of the boundary conditions To solve this problem we begin by calculating the outer solution, described by the following equations:
GU.
1
01
= 02 = 2'
8N(1) _....::.;11"-0 8Xl
+
8 2 M.(1) 22 ---:::8-x~~"-0
03
8N(1) 12
8X2
= 0,
04
8N(1)
0 -, -
+ R".r(l) .LV 22
_
= 1,
12 8Xl
R2
P
82
5
05
+
= 06 = 4'
8N(1) 22
8X2
(1)
W 8t 2
= 0;
= 0,
(3.5.45)
210
3. Continuous Systems
Table S.4(c). Splitting of the boundary conditions
c 12 C 34 13 C 46
12 C 38 13 C 68 23 C 45 34 C 56 23
C 58 C 35 68 12 C 47 14 C 67 12 C 78
16 C 78 24 C 57 C 45 67 25 C 78 C 56 78
Ring reinforced shells
+
(1) U(1) W(1) W(2) W(2) U1 , 2 , , xl (1) N(1) W(1) W(2) W(2) U 2' 11' , xl U(1) U(1) W(1) W(2) M(2) 1'2' '11 (1) N(1) W(1) W(2) M(2) U 2' 11' , 11 U(1) W(2) W(1) W(2) N(1) l' Xl' , 12 (1) N(1) W(1) W(2) W(2) N 11' 12' , Xl U(1) N(1) W(1) W(2) M(2) l' 12' '11 N(1) N(1) W(1) W(2) M(2) 11' 12' , 11 U(1) U(1) W(1) W(2) Q(2) 1 , 2' xl xl' 1 U(1) N(1) W(1) W(2) Q(2) 2' 11' xl xl' 1 u(1) U(1) M(1) M(2) Q(2) l' 2' 11 11' 1 u(1) N(1) M(1) M(2) Q(2) 2' 11' 11 11' 1 N(1) N(1) W(1) W(2) Q(2) 11' 12' Xl Xl' 1 U (1) N(1) M(1) M(2) Q(2) 1, 12' 11 11' 1 u(1) N(1) W(1) W(2) Q(2) l' 12' Xl Xl' 1 N(1) N(1) M(1) M(2) Q(2) 11' 12' 11 11' 1
+ + + +
+
+ + + + +
+ + + + +
The boundary layers are defined by following limiting system:
1
02
1
= 4'
8f.l(2)
_~ll=--
+
8X1 8 2 M(2) 11 -8---;";2=-Xl
°1
= 2'
°3
= 0,
8N(2) 12 _ 0 8X2 - ,
+
82M(2) 22
8 2 X2
=
B 33 £g) ,
= 0,
8N(2) 12 8X1
+ RN22(2) +
o = B11£~;) + B12£~~)' Ng)
v
+
1
05
= 2'
8N,(2) 22 = 8X2
8 2 W (2) 8x~
(1)
3
°6
= 4'
0
(3.5.46)
' 2
82w(2)
NIl - pR - -2- = 0;
8t
= B21£~;) + B22£~~) + K22K~~), Mi~) = D11K~;); MJ~) = D 22 Kg) + K22£~~); Ng)
3.5 Regular and Singular Asymptotics
(2) _ C 11
.!. 8U~2)
_1_ (8W(2)) 2 R 8Xl + 2R2 8Xl
-
cg) =
211
(2) _ w(2). c22 - - R '
...!... (au~2) + 8U~2)) + _1_ 8w(2) 8w(2) . 2R
8X2
8Xl
2R2 8Xl
8X2
If at first one calculates the boundary layer, asymptotic analyses of the governing equations show that v = -0.5 and the limiting system for the inner state coincides with system (3.5.45), if in the third equation of motion the term (8 2w(2) /8x~)Nn) is introduced. ' Let us consider the amplitude-frequency dependencies for nonlinear vibrations of the simply supported stringer shell. The governing relations may be chosen in the form (3.5.41). Let us pose [70, 71] W(Xl, X2, t) = iI (t) sin SlXl cos S2X2 + h(t) sin 2 SlXl. (3.5.47) Here Sl = trml- l , S2 = n; and m and n are the wave numbers in the axial (circumferential) direction. It should be pointed out that the time functions 11 and h are not independent. The connection between them should be taken from the condition of continuity of the displacement U2 in the circumferential direction [70, 71]: Ii
J~~~
a = 2trR- 1 .
d X 2 = 0,
(3.5.48)
o Using geometrical relations and (3.5.47), one obtains from (3.5.48)
h
= 0.25R- 1 s~lr
The Airy stress function we get from relations (3.5.47) is in the following form: B 1-1", '.I'
=
P2s2-2(1
-
. 5 2~c 2 cos 2S2 X2 ca s 22)C~ smSlXl COSS2X2 - 16 P
+0.5s~e sin SlXl cos 2S l Xl cos s2x2;
P = sls2"l,
e= R- l II.
Now we use the Galerkin procedure for the governing equations Ii l
II II
L l (w) sin SlXl cos S2 X2 dXl dX2,
o 0
Ii l
Ll(w)sin2s1xldxldx2,
o 0 where
L 1 (w) = V'tw - R
(:::~ - V'~) 4> -
L(w,4»
+ PR2~~.
212
3. Continuous Systems
As a result we obtain the following ordinary differential equation with constant coefficients for the time function e(t):
~:; + o{ [ G¥; Y+e~:n + A,O A e' + A 2
3{5
= O.
(3.5.49)
JB
Here tl = l (pR2)-1 t; Al = e~e4+2e~e3e4P-2+e~e2e4P-4+s24(1-e~s~)2; A 2 = 116 + !s~de4 - ~(1- e~s~); A3 = ~s~; a = 332s~. Let us consider a practically important case of steady-state periodical vibrations and use the method of strained coordinates [119, 120, 122] for solving (3.5.49). We change the independent variable tl to a new one r = wt l , where w is an unknown frequency of the periodic solution. Then (3.5.49) must be replaced by . d (3.5.50) dr'
()=
The initial time point may be chosen in any way because of the periodicity of the solution.. Without loss of generality let r =
e= I, e= O.
0,
. (3.5.51)
Let us introduce a formal small parameter and pose
e(r) = e6(r) + e 26(r) + e 36(r) + ... W = Wo + eWl + e2W2 + e3W3 + ...
(3.5.52) (3.5.53)
with the constraint that expansion (3.5.53~ is uniformly asymptotic [119, 120, 122]. Now substituting (3.5.52), (3.5.53) into (3.5.50) and comparing the coefficients of en in the usual way, we find 1
e
2
e 3 e
e
4
2"
wo6 + AIel = 0,
(3.5.54)
2
"
wo6 + A l e2 = 2WO,W16, 2"
(3.5.55)
2 . . . .
wo6 + A le3 = -(WI + 2WOW2)6 - 2WOW16 2 '2 .. 2 3 - awO(6el + 6el) - A 2 l ,
e
(3.5.56)
W5e4 + A le4 = -2(WOW3 + WOW2)el - (w~ + 2WOW2)e2 - 2WOWle3 - a [W5(26ele2 + 26e16 +
+2WOWl(6e~
+
e~e'l)]
-
3A3e~6,
e~e2 + e~e2) (3.5.57)
The initial conditions (3.5.54) give us 6(0) =
I,
ei(O) = 0,
el(O) = 0, ei(O) = 0, i = 1,2, ...
(3.5.58) (3.5.59)
3.5 Regular and Singular Asymptotics
213
The solution of the initial value problem (3.5.57), (3.5.58) is Wo =~;
6
= fCOST.
(3.5.60)
Let us rewrite (3.5.55), taking into account (3.5.60): ..
6 +6
=
2f
Jjf;Wl cos T.
(3.5.61)
From the conditions of the absence of secular terms, it follows that WI = O. Then initial value problem (3.5.59), (3.5.61) has the solution 6 = O. In the same way one obtains W2 = 0.125~(3Cl - 20)f2; C l = A 2A 1l ; 6 = -0.03125Cl(COS T - cos 3T)f3 + 0.06250 (cos T + COS3T)f3; W3
C2
= 0.039062~ h'1(-Y2 - 6CI) - 2')'2h'2 + 80) + 80C2] f4; = A 3 A 1 , 11 = C l - 20, "1'3 = 3Cl - 20.
= 0;
W4
l
As a result, we have approximate expressions for the frequency of the periodic nonlinear vibrations:
w=~n;
n = 1 + 0.125 {')'2 + 0.03125 [')'2(')'1 -
2')'2 - 160)
- 6C l ')'1 + 80C2] f 2c 2 } f 2c 2.
For the clamped edges of the shell the displacement w may be approximated as W(Xl' X2, t) = II (t) sin 2 SlXl cos S2X2 + h(t) sin 2 SlXl, where h = l36s~R-l ff. The coefficients of time equation (3.5.49) in this case are
[2
A 1 ="38 2ClC4 A 2 = 12 1
2 c 4P -2 + gClC2P 3 2 -4 + 2s -4( 1 + elc3 2
2 4] ; s 2 -1) + 0.125clC4S2 +"23 [(c62
21 4 A 3 = 128S2;
0
9 4 = 64 s2'
n
are obtained for the following
= 0.012;
C4 = 0.6; C5 = 0.75; = 500; 7 = f Rh- l ; c = 1;
The numerical results for the frequency values of the parameters Cl C6
c6 s 2)2] 2 ;
= 0.015; = 0.005;
C2 C7
= 0.00016;
= 0;
LR- l
C3
= 2;
Rh- l
m = 1,n = 8.
The typical amplitude-frequency dependencies are represented in Fig. 3.7. From these results it follows that the vibration frequency n decreases with increasing amplitude. Consequently, for the stringer shell we have the weak nonlinearity of the soft type.
3. Continuous Systems
214
1.0...-------r------,-----,
I
\
::: ~.~~~..~~-----l_e\. -+c~~~~---
,
0.4 -- --- -~- ------~0.2
-----~-
----~--
0.5
e--------
1
n
1.5
Fig. 3.1. Frequency-amplitude dependence for the stringer shell
As can be seen from the above simplified boundary value problem, after asymptotic decomposition the computation of the shell can be done in several stages; at each stage all equations not higher than the fourth order in Xl must be considered with the corresponding boundary conditions. Simplified boundary value problems, which are novel in the literature concerning shells, may serve as a basis for calculations of a very wide class of problems for elastic shells. They can also be used as a starting point when seeking further reduced equations under additional simplifying assumptions. We note that the results obtained with the use of this separation are in good agreement with the results of numerical computations carried out without separation of the boundary conditions. '. 3.5.3 Nonlinear Oscillations of a Cylindrical Panel The equations of the nonlinear vibrations of an elastic rectangular cylindrical panel [151] together with the initial and boundary conditions can be written in the dimensionless form L1~w
+ 64 a;w -
k62a~p' = 64 q(x, t)
1 L1~P + 2a62L(w,w) + ak6 2a;w
=
+ 62 L(w, p),
(3.5.62)
0,
[w, atw]t=o = 0, [a;p, aXaypa~W + v6 2a;w, a~w + (2 - v)6 2a;ayw]Y=±1 [w; a;w, 62a;p - va~p + (2 + v)axa~p]Y=±1 = 0, L1 I -- ay ':}.2
(3.5.63) = 0,
(3.5.64) (3.5.65)
+ 62 ax2 ,6 = -a2. al
Here Xl
2
= alx,
c =
X2 = a2X,
pha4I D- I ,
a
=
W = alW, F = Dp, T Eh al2D- I , k = a - lR- 1 ,
= ct, Q = qD a -3 ., l . (3566)
3.5 Regular and Singular Asymptotics
215
It is assumed that the transverse load Q is a function of the longitudinal coordinate Xl and the time 'T. The panel platform occupies the rectangle Ixpl < ap, /3 = 1,2. The boundary conditions (3.5.64) correspond to a free edge, and (3.5.65) to a fixed hinge support. Besides problem (3.5.62)-(3.5.65), the nonlinear integro-differential equation of the vibrations of a circular arch, written below in dimensionless form
2 (1- v )a:w + 8,w -
~(k + &,w)
1
J[~(axW)2
- kW] dx = q
(3.5.67)
-1
[w, atw]t=o = 0,
[w, a;W]X±l = 0
is considered.
A natural small parameter 6 occurs in the system of (3.5.62)-(3.5.65). Therefore, there is a problem of constructing an asymptotic form as 6 --+ O. Asymptotic expansions are constructed in the form
~ 6m [ wm(x, Y,t) W = ~o ~=
l; 6
m
00
[
1+x 1+X] , + um(-6-,y,t) + vm(-6-,y,t) 1+x
(3.5.68)
1+x
~m(x, Y, t) + CPm(-6-' Y, t) + '¢m(-6-' Y, t) ] .
The functions Wo m , ~m are found by using a first iteration process. For this the solution is sought in the form 00
{w,~} = L
6m{wm,~m}.
(3.5.69)
m=O
We substitute (3.5.69) into (3.5.62)-(3.5.65) and collect coefficients of identical powers of 6. Equating the coefficients of 60 and 6 1 to zero, to determine Wo, ~o and WI, ~1 we obtain
a~Wm = OJ
[a~Wm, a;Wm]Y=±l = 0;
[w m , a;Wm]X=±l = 0;
(3.5.70)
m =0,1
a~~m = 0; Seeking W m
, ~m
[a;~m, axay~m]Y=±l = 0;
[a;~m, axa;~m]x=±l = OJ
in the form 3
{Wm,~m} = Lyj{Wm,j,~m,j} j=O
we have from (3.5.70)
Wm = wm,o(x, t)+YWm,l(X, t), [wm,o, a;wm,O]X=±l = 0, ~m = 0.(3.5.71) The function WO,O is still uknown and will be determined below. The function ~o is taken to be equal to zero since it follows from the formulation of
3. Continuous Systems
216
problem (3.5.62)-(3.5.65) that the function g> is determined to the accuracy of the linear components in x and y. Continuing the iteration process it is found that the functions Wm,j, tPm,j vanish for odd values of m and j. Consequently, henceforth in this chapter we speak only about evaluating the function Wm,j, tPm,j for even m and j. Equating the expression for 6 2 to zero and taking (3.5.71) into account, we deduce
W2 = W2,O(X, t) + y2 w2 ,2(X, t); tP = C2(t)y2.
2W2,2 = -v8;wo,o;
(3.5.72)
The function W2 ,0 is still unknown and will be determined below. At this stage of the first iteration process, the conditions on the boundary x = ± 1 are not satisfied. The discrepancies occurring here are later compensated by using boundary layer functions. To determine C 2 (t) we will use the well-known identity connecting the functions tP and W for a fixed reinforcement of the boundary x = ± 1 in the longitudinal direction 1
1
f(a~
. (3.5.73)
-1
Using (3.5.69) and (3.5.72), we deduce from (3.5.73) that 1
C2(t)
~ f U(axwo,o)2 - kWO,o] d~.
=
-1
(3.5.74)
•
Equating the expression for 6 4 to zero, we obtain the system of equations
8~W4 + 28;8~W2
+ 8;wo - (k + 8;wo,o)8;tP2 = q, [8~W4 + V8;W2' 8;W4 + (2 - v)8;8yW2]y=±1 = 0, 8~tP4 + a( k + 8;wo,o)8;W2 = 0, [8;tP4 ,8x 8ytP 4]Y=±1 =
(3.5.75)
0.
We find from (3.5.75) 2
W4
=
L
y2m W4 ,2m(X, t),
W4 ,4-
v - 2 ~
2
8x W2,2
(3.5.76)
m=O
W4:2 ,
V 2 = ( 1 - v ) 8x2 W22, - -8 2 x W20 ,.
Here W4,O is also an uknown function. Taking account of (3.5.72) and (3.5.74) to determine the principal term of the expansion (3.5.69) from (3.5.75) and (3.5.76), we obtain the integro-differrential equation (3.5.67) for which the zeroth initial and boundary conditions are derived from (3.5.63) and (3.5.65) by using (3.5.71). Changing to dimensional variables in (3.5.67) by means of (3.5.66), we arrive at the well-known equation of arch vibrations. Furthermore. we find
3.5 Regular and Singular Asymptotics
217
the principal term of the expansion (3.5.69) for the function ifJ from (3.5.72) and (3.5.74). Let us now construct the next terms of the asymptotic form. It can be shown that W2m, tP 2m are determined in the form
ifJ2,
m
{W2m, ifJ 2m } = Ly2j {W2m,2j, ifJ2m,2j(X, j=O
tn·
In particular, we have from (3.5.72), (3.5.73) and (3.5.75) 1
ifJ 4 ,2 = -2ifJ 4 , 4 ! ifJ4,4
dx +
~.
-1
To determine
W2,O
we derive 1
(1 -
2 v )8:W2,O
+ 8;W2,O -
2ifJ2,28;W2,O - /g =
2/
!
ifJ4,4
dx
-1
V
2
+68xq
2 2 + 3v (1 -
(I = + a~qwO,O; k
9
6
v)8xwo,0,
= -;
j
w2,olx=±1 = 0
(3.5.77)
W2,O/dX) .
We note that, unlike (3.5.67), equation (3.5.77) is linear. Equating the expressions for 62m + 2 , 62m + 4 (m = 2, 3, ... ) to determine the functions W2m and ifJ2m , we obtain
(1 -
2 v )8:W2m,O
+ 8;W2m,O -
2/ifJ 2m + 2 - 2ifJ2,28;w2m,O
m+l ifJ2m+2,2
= C2m + 2 (t) -
L jifJ 2m + 2,2j; j=2
= l2m,O
(3.5.78)
m+l
ifJ 2m +2,O
=
L j=2
(j -
1)ifJ2m+ 2,2j.
The functions l2m,0, ifJ2m + 2,2j (j = 2, ... , m + 1) are found in the previous stages of the first iteration process, while the functions W2m,2j are calculated in terms of derivatives of the functions wo,o, W2,O, ... , W2m-4,0. The functions C 2m + 2 (t) (m > 1) are determined from identity (3.5.73) on substituting expansion (3.5.68). The boundary layer functions Um,
2), needed to close (3.5.77) and (3.5.78) are obtained here simultaneously. We substitute (3.5.68) into (3.5.62)-(3.5.65), we take account of the results of the first iteration process, we make a change of the variables x = -1 + 6e (x = 1 + 6() and we collect coefficients of identical powers of 6. Equating the coefficients for 60 to zero, we find a system of nonlinear equations with zero
218
3. Continuous Systems
right-hand side for Uo, CPo from which we obtain uo coefficients for 61, 6 2, 63 to zero we deduce
+ CPo =
O. Equating the
L1~CP2 = 0, [a~CP2' aeayCP2]y+±1 = 0 (3.5.79) Acp21e=0 = 2VC2(t), Bcp21e=0 = 0, [Acp2' Bcp2]e=l-ooo --+ 0 L1~u4 = ka~CP2, [U4, alU4]e=l-ooo --+ 0 (3.5.80)
Ul = CPl = U2 = U3 = 0,
[a~U4 + ValU4' ~U4
+ (2 - v)a~ayU4]y=±1 = 0
= -w4Ix=-l,
alU4le=0 = -o;w2Ix=-1
u41e=0
(L1 2 = al +a;,
A
= al- va;,
B
2
= al + (2 + v)aea~, 1=6'
We note that the boundary value problems for u m , CPm are linear for m > 1. The functions v m , 1/Jm are found analogously. We will illustrate the calculation of the boundary layer function U4 for the case of a rectangular plate (k = 0). We construct the solution in the form 00
U4 = aoeSOeFo(Y)
+ 2Re
L
ame-Sont; Fm(y).
m=l The Papkovich functions Fm(y) are determined from the boundary value problem (the prime denotes the derivative with respect to y):
+ 2s~F~ + s~Fm = 0 [F~ + vs~Fm, F::: + (2 - v)s~F:n]y=±l = 0
F/nv
(3.5.81)
(so, Sm are, respectively, the real and co~plex roots of the equation !P(s) = (3 + v) sin 2s -- (1 - v)2s = 0). . To calculate am from boundary conditions (3.5.80), the problem is posed of representing the two real functions /I = -W4( -1, y, t) and h = -a;W2( -1, y, t) in the form of the series 00
{/I, f2} =
L {I, s~}amFm(Y).
(3.5.82)
m=l Here the time t plays the role of a parameter. To obtain the initial conditions for t = 0 for the function W2m ,0, we substitute (3.5.68) into (3.5.63), and we collect the coefficients of identical powers of 6 and equate them to zero. In particular, the coefficient for 6° yields the initial conditions written in (3.5.67) for wo,o, The consistency conditions
q(±I,O)
= a;q(±I, 0) = at q(±I, 0) = 0
should be satisfied here. The coefficients of 62 and 64 are reduced, respectively, to the zero-initial conditions for the functions W2,0, atW2,O and W4,0, atW4,0. Analogous consistency conditions on the higher derivatives of q are added to construct the next terms of the expansion.
3.5 Regular and Singular Asymptotics
219
After evaluating the principal terms of expansion (3.5.68), the process of constructing the next terms of the asymptotic form is continued analogously: functions of the first and second iteration processes are determined alternately. The boundary values of the functions of the first iteration process W2m,O, 8;W2m,o are determined simultaneously in the solution of the boundary layer problems. In the case of rigid clamping of the panel edges Xl = al ([w, 8x W]x=±1 = 0) the principal term of the expansion is also determined from the equation of arch vibrations, but with the boundary conditions [w, 8x W]x=±1 = O. In the case of the hinge of supports or rigid clamping of the boundaries X2 = ±a2 there is no passage to the limit from the equations of the vibrations of a cylindrical panel to the equations of the vibrations of an arch.
3.5.4 Stability of Thin Spherical Shells Under Dynamic Loading 1 As shown below, the theory of two-point Pade approximants gives the possibility of obtaining a solution of very complicated problems. The very interesting example of using this technique is matching by TPPA coefficients of limiting equations and constructing on this basis the constituting equation, which may be exploited for any values of the parameters. In the course of solving the problem of the stability of shells under dynamic loads, it becomes necessary to describe the motion of its mid-plane with deflections that are large when compared to the thickness. Characteristic forms from the linear theory are usually chosen as the approximating functions when approximate analytical methods are used to reduce the initial system of partial differential equations to a Cauchy problem for ordinary diferential equations. However, the effectiveness of such approaches is limited to the region of small values of the deflection amplitude. Here an asymptotic method is used to obtain the corresponding differential equation describing the motion of a shell with significant deflections. Since one may use as a small parameter a quantity which is proportional to the ratio of the thickness of the shell to the amplitude of its deflection Wo, the resulting equation will be more accurate, the greater the deflection of the shell. To describe the motion of the structure throughout the entire range of displacements, we obtain an ordinary differential equation whose coefficients are determined by combining the corresponding expansions for large and small deflections. The thus-formulated Cauchy problem is solved numerically by the Runge-Kutta method. The efficiency of the proposed approach is evaluated by comparing the results of calculations with known experimental data. As the initial equations, we will examine the equations of motion of an orthotropic spherical shell written in terms of the stress function ifJ and w for the case of the axisymmetric deformation 1
By courtesy of A.Yu. Evkin
220
3. Continuous Systems
D1l
WIll) -D22 ( -wI - -wI ) = h a (aw aifJ) IV +2----2 3 r r r r 8r ar ar 2 h a ( raifJ) aw +-- +q+-ph-; rR ar 8r 8t 2
(w
u
B
( ifJI
V
(>II 1 )
+ 2-r-
( (>1
-
~ +.!. a 2w aw + ~~ (>1 )
B22 ?'" -
r ar 2 ar
B U B 22 - B~2
r
R ar
(3.5.83)
(r
aw) = ar
o.
After the substitution of variables r2
hifJ F=-
z---
- woR'
w W=-, Wo
Wo
qR2
_
Bu
Bu
q=4
we obtain c: 2[z 2W IV
+ 4zW III + W II (9 -
= 2(zw I FI)I + (zFI{q -
Ao)/4] 2w a C fh2 '
(3.5.84)
c: 2[z2 FIV + 4zF III + F II (9 - a)/4]
= WI (2zW I + WI) + (zWI)I,
(3.5.85)
where
Ao = D 22 , Du w Wo
Bu
a _ B 22 - Bu '
= h'
= wot,
and Wo is the natural frequency of linear vibrations of the shell. In the case of an isotropic sph~re
T
c:
2
C
=
2h woV3 (1 - v 2 ) ,
= pR2
w5V3(1 2E
Ao
= 1,
a
= 1,
v 2)
.
When the amplitudes of the deflection wo are sufficiently great when compared to the thickness of the shell, the parameter c: 2 becomes small and can be used in an asymptotic integration of system (3.5.84)-(3.5.85). In [36d], the corresponding procedure was performed for the case of a static load on an isotropic spherical shell. It was established that the main approximation of the asymptote yields good results when wo/h > 4. In the case of an orthotropic sphere, the parameter c: 2 can also be regarded as small for the corresponding deflection if we exclude from consideration shells in which there is a substantial increase in flexural rigidity in the meridional direction D 11 .
3.5 Regular and Singular Asymptotics
221
When c = 0, (3.5.85) has two solutions. The first, WI = 0, corresponds to the momentless state of the shell. The second solution, WI = -1, corresponds to the mirror reflection of the part of the shell relative to the plane whose intersection with the sphere gives a circle of radius rl = v'woR. Thus, in the case of large deflections, the form of the shell becomes determinated and in the initial variables it is described by the function
In the neighbourhood of r = rl (z = 1), discontinuities in the derivatives are compensated for by rapidly changing functions of the internal boundary layer. We obtain the following equations for them in the main approximation of the asymptote: v IV
+ 2(v I uI)I
- u lI = 0,
u IV
+ vI (1 -
2v I ) = 0,
(3.5.86)
these equations coinciding with the corresponding equations of the static problem [36d]. Here, the functions u and v are differentiated with respect to the variable x = (1 - z)/c, W = c(r)v(x), F = c(r)u(x). The boundary conditions fo.r the sphere which is rigidly fixed at its boundary (at r = ro) take the form vI
= 0,
vI
= 1,
u lI uI
= 0,
= 0,
= -xo; X -+ +00, X
(3.5.87) (3.5.88)
where
xo =
Zo -1 c
r5
Zo = woR'
The relations of boundary value problem (3.5.86)-(3.5.87) were obtained with the assumption that
q «1,
a2w C ar 2 '" 1.
(3.5.89)
In this case, neither the load nor the inertial term enters into the equations or the boundary conditions for the functions u and v describing the stress state of the internal boundary layer. It should be noted that the satisfaction of relations (3.5.89) is necessary only for large deflections. Thus, it is satisfied in all cases of practical importance. In particular, it is possible to study the action of shock loads in which the parameter q can rearch large values. However, due to the instantaneous nature of these loads, the given parameter remains small even when the deflections are substantial. The equation needed to determine the function wo(t) or c(r) can be obtained by the variational method. As in [36d], we obtain the following equation for the total potential energy of the system when a uniformly distributed radial pressure acts on the surface:
222
3. Continuous Systems
where D1 =
321rD ll hYb R '
I [( 00
Jo =
2] dx -_ 0.56 + 0.3/22 . u II) 2+ (II) II Xo
-:to
An expression for Jo(xo) was obtained after the numerical solution of boundary value problem (3.5.86)-(3.5.87) for different values of Xo and a subsequent approximation of the corresponding function in the interval 0, 5 < Xo < 00. The last term is connected with the effect of the edge on the deformation of the shell. One obtains the following expression in the main approximation for the kinetic energy of the system:
K=
1rh4 Rpj 12w~ , 2
where 1 = wo/h. The corresponding equation of motion has the form
j 1 + j2 + w~V1 = 4/Yb (qiT) + qo) .
(3.5.90)
Here, one has isolated the dynamic q(T) and static qo components of the load:
A=
w~pR2Bll
3 2
B ll B 22 - B?2 '
w 2 = 6J ob / * A
0 h/lb3 / 4
-
J 0 = 0.56 + ' (y'2H/h -
,v1)
5a
(I < 2H),
where H is the camber of the shell. One may describe the motion of a shell with small deflections by the Ritz method. Here, we make use of the following approximation of the deflection function:
w(;, t) =
{
f(t)h
o
[1- (:.rr 0
r.
(3.5.91)
r* ~ r < ro,
where ro is the radius of the circumference of the shell in the plane. Such a function was used in [96] for the case when r* = ro, in the study of the stability of an isotropic medium under dynamic loads. Here, let us examine an orthotropic shell. It is also assumed that the quantity r * is arbitrary within the interval 0 ~ r* < roo The stress function for which the
3.5 Regular and Singular Asymptotics
223
strain-compatibility equation (3.5.83) is satisfied and continuity of the displacements at r. = ro is assured has the form 8 = B 4r.f(1 - V12 V2I) (B d _ ~ 8r 22 R 1 9- a
-B 8f2(1 - VuV2I) (B 2d + ~ 22 r. 9- a where
B 12 V12 = - , Bu
B 12 V21 = - , B22
5
) d 25 - a
_
d
5
+
25 - a
1 (V12 - 5 1 - V12 25 - a
B1=
d
7 ),
49 - a
3)
V12 9- a
- ;
(VI2-3 2V12-10 1 - 1112 9- a 25 - a 1
B2 =
+
+ VI2-7) , 49 - a
r
d= - . r. One obtains the equation of motion in the form
j + f + (3f2 + 1]f3
41'Vbq(r) , A
=
(3.5.92)
where
>. =
:~ [b(9 - >'0) + k
2
(B
1 -
2(9
__ 80 [ -B _ 9 (3 2kA B 1 2 10(9 - a) r2
k= R~' 1]
~ a) + 10(2: _ a)) - qOkJb] ; 4
+ 5(25 -
_
a)
1 ]. 5(49 - a) ,
5 1'="3'
= _ 160
3Ak 2
[B
2
+
2 _ 2 + 4 ] . 5(9 - a) 5(25 - a) 35(49 - a)
An equation describing the motion of the shell with both large and small deflections can be written in the form Aoi + A 1 j2
+ A 2f
= 4A3 V;7f(r)
+ 4O:f '/!7fo,
(3.5.93)
where the coefficients Ai are obtained by combining the corresponding asymptotic representations of the coefficients of (3.5.93) in the form (3.5.90) and (3.5.92). Using the Pade approximation we find
A o = 1 + af, A3
A 1 -- -!!.L 1 + f'
A _ 2 -
1 + a{3w~v7 1 + a{3w~v7 + (3f'
1r Rh 3 f2
= l' + af,
a=
II w 2dF' F
where w (r) is the deflection function describing the form of the shell with small deflections. In accordance with (3.5.91), 0: = 5/k.
224
3. Continuous Systems
1--+0 we obtain (3.5.92), since Ao --+ 1, Al --+ 0, A 2 = 1 + (31 + 0(/ 2 ), A 3 --+ r. Similarly, if we pass to the limit with 1/1--+ 0, we obtain It is easily shown that as
A o '"
ai,
aw 2 A2
'"
vi'
which, to within the constant factor 0, corresponds to (3.5.90). In accordance with the above mathematical model describing the motion of a shell with large deflections, Fig. 3.8 shows three qualitatively different states of shells corresponding to different levels of deflection. In the case of relatively small deflections (I '" 1), the shell undergoes bending in the region of the vertex with radius r •. With an increase in deflection (I » 1), the form of the shell becomes close to a mirror reflection of its surface perpendicular to the axis of rotation. The reflection region (in which the shell undergoes bending) increases in size with an increase in deflection. The radius of its circumference is rl = J]TiJl. Such a shape is energetically advantageous for the shell, since the membrane strains are concentrated within a narrow zone of the internal boundary layer (at r :::::: rl). The third state of the shell in the case of snap-through is characterized by the effect of the fastened edge. The latter prevents a further increase in deflection without membrane strains, which in turn leads to a sharp increase in the stiffness of the structure.
4.-------------,----------,--------:<J'I
2 --.--
oO~----2"------l------"-----------' 4
6
maxi
8
Fig. 3.8. Comparison of the various theoretical approaches
Formulas to determine the largest bending stresses due to the change in the curvature of the shell in the meridional direction were obtained for the corresponding ranges of deflections. With small I, we have
3.5 Regular and Singular Asymptotics
max 10"1 With large
=
48D ll f kRh 2
225
(3.5.94)
(r = r.).
f, we olbtain the asymptotic formula
max 10"1
=
DJ7 (1 + .j2H/h _ .j]
5.7 u
0.33b
1 4 /
)
(3.5.95)
Rh2b1/4
After combining the given expansions, we arrive at the relation max 10"1
=
48D ll kRh2
48b 1 / 4
1+
1l (
5.7k
f 033b 1/ 4 1 + .j2H/h-
(3.5.96) )
l'
Vi
2H
f
226
3. Continuous Systems
Table 3.5. Comparison of the theoretical and experimental data
No. 1 2 3 4
h [mm] 0.27 0.27 0.27 0.27
H [mm] 27.0 25.3 20.7 16.2
Rlh
qnO •
10 5
[N/m 720 768 941 1200
2
]
Qp.10 5
[N/m
2.32 2.36 2.06 1.56
2.8 2.6 1.9 1.6
2
max!
]
1.9 2.1 2.7 4.2
doseness to the isometric transformation of the sphere obtained by mirror reflection of the part of it relative to a plane make it possible to suggest that in the absence of a substantial effect from the fastening of the edge of the shell and constrained growth of the dents due to their interaction, each of them can be examined within the framework of the axisymmetric theory. Here, the symmetry axis for each dent will be its "own" axis. Its position will be determined by the point of the middle surface of the shell with the maximum deflection. Thus, we can expect that the relations obtained above might also be effectively used in other cases of nonaxisymmetric buckling of shells with large deflections. Figure 3.9 compares the results of calculations (curve 1) and tests [97] performed for an isotropic shell with the parameters 2H/ h = 6.87 and R/h = 365. The shell deformed in the elastic stageduring buckling. The conditions under which the shell was loaded by external pressure can be represented in the form
7J(T) =
lOTo7J. q.(5.9 -
o ~ TO < 0.1 \ TO)
5.8
7J.(178 -
350
TO)
0.1 3
~ TO
~ TO ~
<3 178,
where TO = T / A. The results from theory apd experiment agree well with one another. The da.shed line in Fig. 3.9 shows the results calculated by the method in [49d] , which corresponds to the solution of (3.5.92) with k = r~/(Rh) = 6.87. The agreement with the test data is somewhat poorer in this case. Calculations performed with both mathematical models showed that the difference between the results increases with an increase in the parameter 2H/ h. Also the difference in the description of the stress state by the different theoretical methods is substantial, a.s illustrated by curves 1 and 2 (relation 2 was obtained by the method proposed in the present study). Curves 2-4 in Fig. 3.9 show the way in which the dynamic stability of the shell is affected by the additional static component of the external pressure with qo = 0.1,0.2, and 0.3, respectively. In the ranges of qo of practical importance, there is a slight decrease in the critical dynamic pressure. As an example, we also examined the well-studied theoretical case of dynamic loading by an instantaneous applied external pressure which then
3.5 Regular and Singular Asymptotics
227
8r-------..-----..,....------.------;;~
max! 6
o ---~
-~--~-~__+______#_~~"--_+__-----~-.____i
I I I I I
4
._-----~--_.
,
o
•
I
2
-----
1~~--1
j
o
0.2
0.6
0.4
q.
0.8
Fig. 3.9. Comparison of the theoretical and numerical results remains constant over time. We took the load at which the corresponding unstable intermediate equilibrium position is attained as the critical value. For this load, there was also a sharp increase in the amplitude of the deflection. Figure 3.10 shows the results of the calculation performed by the proposed method (curve 1) along with the results obtained in [149, 44d] by powerful numerical methods (curves 2-3). The experimental data from [112] are represented by clear circles. The dark circles represent the critical values of pressure calculated in accordance with the asymmetric theory [149]. 0.8
r------..,....-------r-------,
0.4
6
10
14
Fig. 3.10. Comparison of numerical results with the two-point Pade approximants formula
Here, A = 2[3(1 - v 2 )jl/4(H/h)1/3. It is evident from Fig. 3.10 that the numerical solutions give relations which oscillate and decay with a decrease in the thickness of the shell and approach the above-found asymptotic value q ~
228
3. Continuous Systems
0.43. This phenomenon can be exploited to estimate the critical load, with consideration of the experimental data and the potential for the asymmetric buckling of spherical shells.
3.5.5 Asymptotic Investigation of the Nonlinear Dynamic Boundary Value Problem for a Rod As is well known, asymptotic approaches for nonlinear dynamics of continuous systems are well developed for infinite spatial variables. For systems of finite size we have an infinite number of resonances, and the PoincareLighthill method does not work. The use of an averaging procedure [59d] or the method of multiple scales [107] leads to infinite sytems of nonlinear algebraic or ordinary differential equations, and a subsequent truncation method does not provide the possibility of obtaining all the important properties of the solutions. In this chapter we use an asymptotic procedure which is based on the introduction of an artificial small parameter. Let us assume a governing boundary value problem in the following form 8 2U 8 2U _ U3 8x 2 - 8r 2 - -c , (3.5.97) where all variables are nondimensional, and c is a nondimensional small parameter (c « 1). From the physical point of view we have longitudinal vibrations of a rod with nonlinear drag. Let us introduce a change of the variable
t = wr.
(3.5.98) \
We will now search for solutions using the ansatzes U = Uo + cUI + c 2U2 + ... , w = 1 + cWI + c2W2 + ....
(3.5.99)
After substituting expresions (3.5.98) and (3.5.99) into the governinig boundary value problem (3.5.97) and splitting it with respect to c, one obtains 8 2Uo 82Uo 8x 2 - 8t 2 = 0, (3.5.100) 8 2U1 8 2U1 8 2UO 3 (3.5.101) 8x2 - 8t 2 = 2Wi 8t 2 - Uo . The solution (3.5.100) may be written in the form 00
Uo = C 1 sin x sin t
+ C 2 sin 2xsin 2t + ... =
2: Cisin ix sin it. i=l
Here C 1 is the amplitude of the fundamental oscillation, while the constants C i for i > 1 provide the next approximations. After some routine but cumbersome transformations, we arrive at the following infinite nonlinear algebraic equations
3.6 One-Point Parle Approximants
=
-32w 1 C 1 sinxsint = I : CaCb(3Ca+b-1
229
=
+ 3Ca+b+d + 6C1 I : C~,
a=l
a=l
=
-32w 1i 2 C i sin ix sin it = I : C aCb(3Ca+b-i
+ 3Ca+b+i + Ci-a-b)
a=l
=
+6Ci I:C~.
(3.5.102)
a=l
Systems like (3.5.102) may be obtained in various ways and the main problem in this approach consists in its solution. The truncation of the infinite system (3.5.102) ,cannot give any information about resonances of a higher order. We propose to introduce an artificial small parameter J.L, writing it near all nondiagonal members of system (3.5.102), and representing the unknown coeficients as an expansion: Cn WI
= C~O) + C~I)J.L + C~2)J.L2 + , = w~O) + w~l) J.L + w~2) J.L2 + .
n
= 2,3, ...
After splitting with respect to J.L, solutions may be obtained routinely. It may be easily shown that for even n C(k) n
=0
and
C2
(0)
= _-0.281 250 C I'
(0) _ WI -
(0)
= 0.0144927C1 ,
C s(0) = 0.0002071 CI,
wI
C3
2
-0.001438
I' (0)
C7
= 0.0000030C1 ·
Numerical results (the dependencies of the fundamental nondimensional frequency W upon the nondimensional amplitude C 1 ) are displayed in Fig. 3.11 for various values of the small parameter c.
3.6 One-Point Pade Approximants Using the Method of Boundary Condition Perturbation An analysis of plates under mixed boundary conditions represents a significant practical value: a lot of problems, arising in machine design, civil engineering, etc., are reduced to similar ones. These problems are usually solved using numerical methods [42d]. Nevertheless, the numerical approach does not adequately fit the requirements of optimal structural design or any other kind of optimal structural design ideology. The approximate analytical expansion, accurate enough, will be of great practical advantage for these needs. The basic idea of the present method may be described as follows. The parameter c is introduced in the boundary conditions in such a way that c = 0 corresponds to the common problem under consideration [1]. Then
230
3. Continuous Systems
0.8 0.6
_. ---- ---- ~~--+~--~+---~ I
-1- - ~--~. -~.- ~~~--+----
0.4
-(~-~--+----~---+-
0.2
I
i I
0.5
1
1.5
2
2.5
3
Fundamental amplitude 0 1 Fig. 3.11. Amplitude-frequency dependencies for fundamental oscillations for various values of the small parameter e
the c-expansion of the solution is obtained. As a rule, the expansion fails to converge at the point c = 1. The PA may be used to eliminate that drawback. Let us produce the PA-definition. For an expansion given by (3.6.1) 00
F(c) =
2: Cici i=O
.
,
(3.6.1)
-1
(3.6.2)
the fraction-rational function is
F(e)[m/nJ ~ (~",ei) (~biei)
F(c}[m/n] represents the pA of expansion (3.6.1) if the Maclaurin series of F(c} shows the coincidence of its coefficients with the corresponding coefficients of (3.6.1) up to terms of (m + n + 1}-th order. The features of the PA are as follows: it possesses uniqueness while m and n are chosen; it performes a meromorphic continuation of the function; for its definition from the source expans~on (3.6.1) the linear algebraic problem arises [42, 43]. It is quite often convenient to perform an asymptotic expansion in terms of a small parameter. Namely, the parameter c is introduced into the initial equations or the boundary conditions in such a way that for c = 1 we have the initial boundary problem, and for c = 0, a simplified problem admitting a simple solution. Then, an expansion in terms of the parameter c is constructed. Unfortunately, the obtained series is usually not convergent for c = 1. To eliminate this disadvantage, we can apply the PA. Here is an example. Dorodnitsyn A.A. has proposed a method of perturbation of the form of
3.6 One-Point Pade Approximants
231
the boundary conditions based on the introduction of a formal parameter e.. In order to remove the divergence for e. = 1, analytic continuation has been used but the efficiency of that technique is not high. The PA, on the contrary, gives good results. Let us consider the flexural vibration of a rectangular plate (-0.5k < x < 0.5k, -0.5 < Y < 0.5), simply supported at x = ±0.5k, and having mixed boundary conditions of the "clamped-simple supported" type, symmetrical to the y axis on the sides y = ±0.5. The initial equation is \74 w - AW
= O.
The boundary conditions have the form
= 0, W = 0, W
= 0 for x = ±0.5k; W,yy = H(x)e.(w,yy =f W,y) for y = ±0.5, w,:xx
where H(x) = H(x - J.L) + H( -x - J.L); H(x) is Heaviside's function. Substituting W and A into the form of the e.-series W
= Wo
A = Ao
+ e.Wl +
,
+ e.Al +
(3.6.3) (3.6.4)
,
after applying the usual procedure of the perturbation method, we have
Ao = 1r 4 1/J2 ,
Al = 41r2n2'"'tmn,
A2 = 41r 2 n 2 'V
{
/mn
_ 2n 2
1/J
'00, "
.
{
'"'tmn 1- - [1rO --cth (_I)Tn -1rO 1r21/J 2 2
~
o.cth(-I)i 0i + '"'t~m ~ 2 .
L-
~==1,3,5, ... }
2
+ -n1/J - -23] }
[
{(_I)i}] ¢i cth .¢i/2 .8icth (-I)' f3i/ 2
'
== 2,4,6, " .
> m 2 + n 2k } i 2 < m 2 + n 2k i2
{
2(0.5 - J.L) '"'tim =
4
1
'7r(m 2 -i 2 )
(-1 \Tn
+ ~ sin 21rJ.Lm,
,[{
i'1.
.
for i = m
m JSlll1rJ.L'tcos1rJ.Lm+
- { ';' } sin
"I'm "I'i] COS
for i
#
m,
and E' denotes the summation without the component i = m. Let us compare the frequencies given by this method with the exact values for the limit case (J.L = 0) when both sides y = ±0.5 are completely clamped. For the square plate A = (1.47831r)4. The PA of a segment of series (3.6.4) is
232
3. Continuous Systems
_ ao + ale A(e ) - bo + bl e ' where ao = Ao, bo = 1, al = Al + bIAO, bl = -A2/AI· For e = lone obtains A = (1. 70811r)4; the numerical solution is A (1. 70501r)4. Figure 3.12 presents the diagram of the relation of A to J.L for the cases of the symmetrical and the nonsymmetrical restraint layout. 1.75 . . . . - - - - , - - - - , - - - - , - - - - , - - - - r - - - ,
A/1r 1.70
1.65 _ _ _ _ _ -J:
1.60
y
--. - ~ -1-- .--.-
I . I I I
II I
__ (l.~~L .
O~--tri~
__
L.!=_~~_~
1.55
1.50 -- .---
e(1.8%)
. -.-. - --.Y
----~-t~ _~.--I_-...
1.40 '----'----'----'-----'-----'-----' -0.6 -0.4 -0.2 0 0.2 0.4 J1. 0.6
Fig. 3.12. The relationship be'tween the vibration frequency of the plate partially clamped on two opposite sides and the clamped segment length
Let us consider the application of this approach to the static analysis of the rectangular plate (-O.5a < x < O.5a; -O.5b < fj < O.5b), subjected to a uniform lateral load ij. The plate is simply supported along x = ±a/2 and subjected to mixed boundary conditions ("clamped-hinged"), symmetrical with respect to fj. The governing differential equation may be written as DV 4 W = ij.
(3.6.5)
Let us denote
W W=b'
x
x =-, b
3.6 One-Point Parle Approximants
k
ijb 4
a
= b'
q
= D'
233
(3.6.6)
Taking (3.6.6) into account, the governing equation (3.6.5) may be rewritten as \74W = q.
(3.6.7)
The boundary conditions may be formed as k W=O, Wxx=O, whenx=±'2;
(1 - H(x))Wyy ± H(x)Wy = 0,
W=O,
(3.6.8) 1
when Y = ±-, 2
(3.6.9)
where H(x) = H(x - J.Lk) + H( -x - J.Lk). Introducing the parameter £ into the boundary condition according to the procedure, one obtains 1
W = 0,
W yy = H(x)£(Wyy ± W y) when Y = ±'2'
(3.6.10)
The case £ = 0 gives us a plate which is simply supported along the boundary; the case £ = 1 corresponds to the problem under consideration (3.6.8)-(3.6.9). The intermediate values of £ are related to mixed conditions of a "simply supported-elastic clamping" kind with the elastic support coefficient u =
£/(1 - c). In order to solve the problem, let, us represent the deflection of the plate as
W =W1 +W2, q WI = 8k
(3.6.11) (_1)(m-I)/2 (
L
00
m=I,3,S,...
am
amth am + 2 12h ch 2a m y c am
ham ysh 2a m y ) cos 2a m xj c am 1 00 (_1)(m-I)/2 Am 2 W = -8 2 h (amthamchamy am c am m=I,3,S, ... -2amysh 2a m y) COS 2a m xj
+
(3.6.12)
L
1T'm
am =-·
(3.6.13) (3.6.14)
2k
Expression (3.6.12) describes the deflection of the simply supported plate subjected to the uniform lateral load q. Expression (3.6.13) describes the deflection of the simply supported plate, caused by the edge bending moments, distributed along y = ±0.5: 00
Myly=±o.s=
L m=1.3.S....
Am(-1)(m-I)/2cos2amx.
(3.6.15)
234
3. Continuous Systems
Satisfying boundary condition (3.6.10), one obtains the infinite linear algebraic system for the coefficients Ai as the unknowns:
f
Ai( _1)(i-I)/2 = c
'Yim( _1)(m-I)/2 Am [1 _ _1_ ( 40 m
m=I,3,5, ...
q
+th am )] + +c 8k
~. (m2~i2)
'Yim = {
'Yim
m=I,3,5, ...
.(ch~:m -thom), 2 (0.5 - J.L -
(_I)(m-I)/2
00
L
~m
ch am
o~
i = 1,3,5, ... ,
2;#£ sin 21rmJ.L) ,
(isin1rJ.Licos1rJ.Lm - msin1rJ.Lmcos1rJ.Lm) ,
(3.6.16)
i
= m
i
=1=
m.
Let us apply the perturbation technique to system (3.6.16), representing Ai as the c-expansion 00
Ai
=
L Ai(j)c
j
(3.6.17)
.
j=O
Substituting (3.6.17) into system (3.6.16) and splitting it into the powers of c, one obtains the reccurent formulas for Ai:
(3.6.18)
Ai(o) = 0;
L 00
Ai(l)
=
(_I)(i-I)/2
m=I,3,5, ...
q (_I)(m-I)/2 'Ymi 8k , 4 (am . am
(3.6.19) 00
L
Ai(n) = (_I)(i-l)/2
'Ymi( _1)(m-I)/2 Am(n-l)
m=I,3,5, ...
(3.6.20) The truncated perturbation expansion (with holding three initial non-zero terms) may be PA-transformed:
+ alc)(bo + blc)-l, al = A i (2) + bI A i (I),
A i [I/I](c) = c(ao
= A~(I), bo = 1,
(3.6.21)
bl = -Ai (3)/A i (2). Let us consider the limit cases for (3.6.21). Firstly, J.L = 0.5 corresponds to the simply supported edge y = ±0.5. The W'Yrni = 0, therefore Ai(j) = 0, W 2 = 0, W = WI, i.e., the exact solution for the plate, simply supported along the boundary. Secondly, J.L = 0.0 corresponds to the fully clamped edge y = 0.5. Here ao
I,
'Ymi = { 0 ,
when i when i
=m =1=
m
3.6 One-Point Pade Approximants
235
and the recurrent relation yields (3.6.22)
Ai(o) = 0; q 1 ( ai Ai(l) = 8k4 h2 ai
c
ai
-
thai
)
(3.6.23)
;
q 1 ai 1 A.( ) = - - ( - thai) 1 - - 4 [ , n 8k ch 2 ai 4a i
at
(
ai h2 c ai
n-l
+ thai ) ]
(3.6.24)
For c = 1, the PA for the truncated expansion (3.6.22)-(3.6.24) is
A'[l/l](c ,
ai - thai(aithai
q
= 1) = -2ar . ai -
+ 1)
.
(3.6.25)
0.5
Fig. 3.13. The relationship between the normal displacement of the plate partially clamped on two opposite sides and the damped segment length
th ai(aith ai - 1)
4.5 ,....----,...---.,.----..,.----..,-------, I
---J--~
W 10-3 q 4.0
I
3.5
3.0
2.5
- -
-
0.5
0
x
0.5
2.0
1.5
0
0.1
0.2
0.4 J1
The formula (3.6.11), taking (3.6.25) into account, describes the plate deformation when the x = ±0.5 edge is simply supported and the y = ±0.5 egde is clamped. The analysis, listed below, was carried out for the square plate. Expansion (3.6.21) for Ai was truncated to ten (initial) terms for c = 1. The deflection and bending moments in the centre of the plate are calculated for several given values of the parameter J1 (see Fig. 3.13-3.14).
236
3. Continuous Systems
5.0 .------,.-----.------,-----,-------,
M 10-2 q
4.5 .. -
4.0 .
-i--~ I
"----+----,.".H-
Arl-~~---t-1I1 x =0
y=o
3.5
3.0
I
2.0
---J -"""----"I
1.5 ' - - _ - - - J ._ _----'-_ _---'-_ _--'--_------' o 0.1 0.2 0.3 0.4 J1 0.5
Fig. 3.14. The relationship between the moment M 11 in the plate partially damped on two opposite sides and the c1aInped segment length
The results, obtained by the coupled series method [70d], are shown as dots. The dotted curves display the data computed by the finite element method. The discrepancy of the deflection, as well as the bending moments, does not exceed 5%, which confirms the acceptable accuracy of this method. Figure 3.15 shows the values of the edge moments My along y = ±0.5. Various problems of statics and dynamics of plates and shells, subjected to mixed boundary conditions, may be solved effectively on the basis of the approach presented here. Nonlinear problems can very often be solved by means of the perturbative approach. (
3.7 Two-Point Parle Approximants: A Plate on Nonlinear Support The present section deals with the problem of oscillations of a plate (a beam in the limit case) on a nonlinear elastic support. This problem may be solved by numerical or asymptotic methods. In the latter case quasilinear asymptotics are usually used [66, 67]. Then, one cannot obtain solutions for large amplitudes. Here a new nonquasilinear asymptotic is proposed. Heuristically it may be described as follows. In the long-wave approximation plate bending rigidity may be neglected, and we may investigate oscillations of a rigid body
3.7 Two-Point Pade Approximants
237
16 ...----......---------,--------;---,.-----,
Mq 10- 2
-r ,
14
~=0.3
!
.--1 ~-----~-~ /-£=0.2
12 10 . 8 --
/-£=0 6 I
-l
4
- -_.
2
--~
-L
0 -2
I 0.1
0
0.3
0.2
0.4
X
0.5
Fig. 3.15. The values of the moment My alone the line y = ±0.5 for various
p.
on an elastic nonlinear spring. In the short-wave approximation nonlinearity of the foundation may be neglected. It is typical for asymptotic methods that approximate solutions may be formulated for some limiting values of the parameters. For the intermediate case the analysis is very difficult. It is possible to overcome these difficulties using tw~point Pade approximants. Let us consider free vibrations of a beam on a nonlinear elastic support. The governing equation may be written in the form
+ k(w + '(31h2w3) + J.LttW = O. Jl/4; W = wi hI; x = (Trx)1 Lj
EJw xxxx
(3.7.1)
Here hI = {3 = k 2/k 1; k k 1L 41Tr4j J.L = 4 4 J.LL ITr : k 11 k2 are foundation coefficients. Let (without loss of the general character of the solution) the beam be simply supported:
w
(~Tr
1
t) = w (±2Tr t) = xx
1
O.
(3.7.2)
238
3. Continuous Systems
The initial conditions we assumed are as follows:
W(x,O) = Acosnx;
Wt(x,O) = O.
(3.7.3)
e,
Taking into consideration the new independent variables rewritten as weeee + e(w + aw 3 ) + w 2w rr = O. Here = nx; T = n 2(EJ/J.L)1/2tw; e = k/(EJn 4); a = {3h?
T,
(3.7.1) may be (3.7.4)
e
We assume a "'" 1 (this is the case of the essentially nonlinear foundation) and investigate two limit cases. For n "'" 1 (the long-wave case) e ~ 1, for n ~ 1 (the short-wave case) e «: 1. Let e « 1 (the short-wave case). The displacement W and the "frequency" w may be expressed as the e-expansions
W = Wo + eWI + e2 W2 + ... , W = Wo + eWI + e2W2 +...
(3.7.5)
After substituting (3.7.5) into (3.7.4) and splitting it bye, the following recurrent sequence of equations may be obtained:
w~4)
+ w5worr = (4) 2 _ WI + WoWlrr -
0,
2 3 - WOWIWO rr - Wo - aow ,
Satisfying the boundary and initial conditions (3.7.2), (3.7.3), we obtain , W = A cos ( cos T. (3.7.6)
.
The conditions of the absence of the secular term lead to the expressions Wo = 1;
9
WI = 0.5 + 32aA2;
(3.7.7)
W2 = -0.125 - ~aA2 _ 459 a 2 A 4. 32 ,2048 Now we are going to investigate the long-wave case (e » 1). The beam displacement and frequency square "ansatzes" are W -_ Wo + e-1 WI +-2 e W2 + ... , W _eO. 5 [W(O) + e- 1w(l) + e- 2w(2)
+ ...J.
(3.7.8)
Substituting expressions (3.7.8) into (3.7.4) and performing the e-splitting, one obtains the system of equations which permits us to determine the unknown expansion coefficient 2 w(O) = 1 + aA cos(. (3.7.9) Using tw~point Pade approximants, one may obtain an analytical solution for any value of the paremeter e.
3.7 Two-Point Pade Approximants
239
In our case we have =
W
Wo + (WI + w)cO. 5 + w(O) wc 1. 5 1 +wcO. 5
(3.7.10)
where w = W2/(W(O) - wd. We can obtain the solution for the linear case from formulae (3.7.7), (3.7.9), (3.7.10), assuming f3 = O. There is an exact solution in the linear case, and we can compare it with the approach presented above. The numerical results are plotted in Fig. 3.16, where the curves correspond to: 1 - the exact solution; 2 - the matched spectrum expression (3.7.10) ({3 = 0). The results are consistent with the physics of the problem and confirm the reliability of the approximate solution. Twcrpoint Pade approximants overcome the locality of the asymptotic expansions. Curve 2 coincides satisfactorily with the exact solution everywhere in the interval considered.
3,....-----r------r-------,.---.,....---.,...------r---...."..
I I --1- "-----_.
w 2.5
---------t--~~'"
2---- -
1.5
lL.---~--....l...--
o
_
L.-_
__l__ _~_ _____l.
234
1
5
6
_____J
c
7
Fig. 3.16. Comparison of the exact solution and the two-point Pade approximants formula
Now we investigate the case of a plate. Only final results of the asymptotic analyses are displayed here. The initial partial differential equation may be written as weeee
2 2l weeTJTJ
W
= w/h;
+
l4WTJTJTJTJ
+ c(w +
3 QW )
+ W2WTITI =
o.
= my; y = (tryd/L 1; l = L/L 1; Q = f31h2; = p2(mn)4w ; p2 = L 4/J1(Dtr 4); ko = kd(Dtr 4).
Here 7'1
+
TJ
Let the plate be simply supported:
trn
W
= wee = 0 for <' = ±"'""2;
W
= wTJTJ =
0
for
<' = ± tr~ ;
n
= 1,3,5, .
m = 2, 4, 6,
.
Cl
= ko(mn)4;
240
3. Continuous Systems
and the initial conditions be tV = A cos ( cos 1];
tV T = 0
for
T
= O.
As a result of this description of the asymptotic procedure, one obtains c »1, w 2 = w(O) + cW(I) + ... = 1 + A 2 cos 2 (cos 21], W 2 = Wo
c «1,
+ C -1-WI + C -2-W2 + ... ,
where:
Wo = (n -2 + l2 m -2)2; 27aA2 Wi = 1 + 64 9aA 2 9aA2 W2 = - 512 wo + 2048
+
1 9wo - WOl
+
(6
_
Wo _ WOl + 6(wo - W02)
1 0.284) 9wo - W02 - wo ;
WOl = 9n -2 + m -2l2; W02 = n- 2 + 9m- 2l 2. Using the twcrpoint Pade approximant we have W
2
=
Wo
+ (WO + WOW)c + w(O)w
--....;.....-----=~---
(1 +wc
where W = W2/(W(O) - WI)' : This formula approximately describeS the plate frequency spectrum for any c.
3.8 Solitons and Soliton-Like Approaches in the Case of Strong Nonlinearity The investigation of the dynamics of a number of structures, particularly the heat exchangers of electric stations, results in the barely studied computational schemes with many impact pairs [133, 134] that do not allow direct application of traditional methods of the theory of vibration-impact systems [165]. Utilization of the analytical approximation of shock interaction permits reduction of the initial problem, under specific conditions, to the computation of a mechanical system (beam, string) with nonlinear elastic supports. Consequently, certain general regularities of the dynamical behaviour of such systems, associated with the existence of regimes of soliton type therein, are successfully clarified. This section is devoted to a detailed numerical investigation of such regimes in the simplest system of the class under consideration, a one-dimensional chain of masses connected by means of a weightless string
3.8 Solitons and Soliton-Like Approaches
241
and interacting with strongly nonlinear elastic supports (unlike the cubic nonlinearity in Sect. 3.1). Let us note that chains with longitudinal exponential interaction have been well studied at this time [153]. The equations of motion of the system under consideration have the form
mWj
+ c(2wj
- Wj-l - Wj+l)
+ 2nwj + F(wj) = 0,
(3.8.1)
where Wj is a vector with the components W]l), w?); m is the magnitude of each of the concentrated masses; c is the stiffness of a linear spring connecting two successive masses. If S is the string tension and l the spacing between the masses, then the stiffness is c = Sjl. For infinite strings j = ... ,-2, -1,0,1,2, ...; in the case of a finite length j = 1,2, ... , N, where Wo = WN+l is in conformity with the conditions for fastening the string in the transverse direction. The nonlinear function F( Wj) that describes the interaction of impact type (an abrupt rise in the reaction for definite magnitudes of the displacements) is given as (3.8.2) Certain typical characteristics of impact pairs in vibrational impact systems cited in [165], say, can be approximated by (3.8.2) (for an appropriate selection of a and b). The equations of motion of an infinite chain allow exact periodic solutions in the form of standing waves:
wy) = aw]l),
W]l) = (-I)i rp(t),
where the function rp( t) satisfies the ordinary differential equation (a is any real number)
q + 4":'cp + 2..!::ep + m
m
a
my"1
+0 2
shbJI
+ a 2 cp =
0.
A soliton can be represented in the form of a modulated wave W]l)
= (-I)jvy) (t), w?) = aw]l),
where the functions vy\t) are not identical for different subscripts. If the set of functions v?\t) describing the modulation depends smoothly on the subscript j, it can be replaced approximately by a function of two variables v(x, t) for which, after the change of variables u =
,/1 + ",2W ,
T
=
/aht, e= V;;;,
Jab2 x cl
'
the following partial differential equation can be written 8 2u
8r 2
8 2u
+ 8e 2 + r 2 u + 26iL + sh u =
0,
(3.8.3)
242
3. Continuous Systems
where r 2 = 4c/ ab and 6 = n/ v!abm. For 6 = 0 and r 2 « I, (3.8.3) has a solution in the form of a localized standing wave, an envelope soliton
J w2 - 1 cos wr ] u(e, r) = 4arth [ wch (Jw2 _ Ie) ,
(3.8.4)
where
w~l\r) =
1 u(ej,r), w?)(r) = ow?)(r). (3.8.5) bv!l +0 2 The parameter w characterizing the frequency of vibrations in the variable r, determine wave amplitude and its degree of spatial localization at the same time. As w increases the soliton becomes so narrow that the conditions for applicability of the long-wave approximation are disturbed. In order to reduce the constraints on the soliton profile, the localized standing waves were investigated numerically on the basis of the initial system of equations (3.8.1), which was converted by the change of variables
(-l)j
J
Uj
=
bwj
,
T =
/¥it
to the form d 2uj - - 2 + {3(2Uj - Uj+l - Uj-d dr
.
Uj
+ 26Uj + -I-I sh IUjl = 0, Uj
(3.8.6)
where {3 = r 2 /4. The number of masses for numerical integration was taken as 40 or 41. The length of the chain was assumed to b~ equal to one, the numbering of the masses was from the left edge of the chain, and the edges were assumed fixed. The system of the first-order equations corresponding to system (3.8.6) was integrated numerically by the Runge-Kutta method with automatic sampling of the integration spacing. Assignment of the initial conditions was by a separate subprogram. During the computation the values of Uj were displayed (in graphical form) as were also those for dUj/dr and F(uj). At the end of the computation, the value of the total system energy was printed out (this quantity was conserved to 1% accuracy during the computation process). Integration of system (3.8.6) under conditions of applicability of the longwave approximation was performed first. As should have been expected, the comPl:lted data here are practically in agreement with the analytic solution, demonstrating the characteristic features of an envelope soliton under appropriate initial conditions. However, from the viewpoint of realizing intensive impact regimes, the cases of greatest interest are when the conditions mentioned are not satisfactory. As before, the initial conditions were given in the form corresponding to the solution of (3.8.5), (3.8.4), for r = 0 for the numerical investigation of system (3.8.6) in these cases. Evolution of the initial perturbation in the case of its comparatively small amplitude is shown in Figs. 3.17a and 3.17b.
3.8 Solitons and Soliton-Like Approaches
243
0.56
\
w
\
0.28
\
- -
,-----\ \
AA' o 'fvy
" ""
.......
/
/ /
I i--------=----I I I
-0.28 --
-056 . 0
0.5
x
o
1
a)
2.9
5.8
r
b)
Fig. 3.17. Space distribution of displacements (a); time evolution of displacements of selected masses (b)
The significant magnitude of its coupling parameter ({3 = 1) results in the fact that the time development of the process does not follow (3.8.4). Nevertheless, the typical behaviour of the soliton of an envelope is present; and the synchronized motion of all the masses with the spatial configuration attenuating from the centre of the soliton is conserved. Therefore, the initial perturbation of the form (1)
u· (0) =U(ej,O) = J
4. a 2 (-I)J rth
VI + 0
dUj(O) = dU(c. 0) = 0 dr dr ""J' ,
vw [~
2 -
wch
UJ~2)(0) = OUJ~I)(O),
1
V ~ M (ej - ~)
],(3.8.7)
2 dUJ )(0) - 0 dr -,
where M is the number of masses, practically corresponds, in this case, to the exact solution for the soliton of an envelope, although its time evolution cannot possibly be predicted quantitatively from (3.8.5) and (3.8.4). A systematic investigation of the influence of the coupling parameter {3 on the nature of the localized waves under intensive excitations of impact type is reflected in Figs. 3.18 and 3.19. The initial conditions were given in the form of (3.8.7). In the weak coupling case, a quite definitive spatial localization of the process is observed (Fig. 3.18a), which is completely conserved in time. Attention is turned to the synchronization of the behaviour of the
244
a)
3. Continuous Systems 4 ,-----+--------,
w 2
f---.
-
- ~
b)
4
~j=21 ~
w
-- - 1-
2
- - - - - - ~-
j=20
o
o "-
..
-2
-2
-4 L -_ _---l..-_ _ o 0.5 x 1
-4
----J
c)
4
d)
w 2
-2
-4
f---------
I
o
0.5
1.8
T
3.6
. (
- - - --
o
o
V
..
--
w
.A
--~.~
L
2
., ~ .
o
--,
4
---
f-----
~"-
L-~-
x
•
-~-
l--~------
r--------
0-1 1
-4
E~~ 0
0.5
x
1
Fig. 3.18. Space distribution Of displacements in the case of weak coupling (a); time evolution of displacements of the most excited masses (b) j initial space distribution of displacements, corresponding to strongly excited central masses (c)j final space distribution of displacements, corresponding to one strongly excited mass (d)
strongly and weakly excited masses. The time dependence of the displacement {Fig. 3.18b} is characteristic for a regular impact regime. Let us note that, in this case, the analytic solution {3.8.5} and {3.8.4} is inapplicable despite the low value of 13, since the degree of localization is too great and the passage to the long-wave approximation {3.8.3} is not yet justified. Figures 3.18c and 3.18d reflect the tendency to localization in the case when the initial excitation differs somewhat from {3.8.7}. The initial mode with just one mass oscillates in the impact regime. The deviation from the "exact" initial conditions for the envelope soliton results in the origination of
Solitons and Soliton-Like Approaches
3.8
245
4,....--.----,-------,---.....,...-----,--------,
w 2
o
-2 Fig. 3.19.
Time evaluation of displacements of the most excited masses
-4 '-------'------'""---------'----"""""--------' 8 10 12 14 r 16 6
an indefinite "background" that interacts with the soliton. This interaction is manifested in the quasi-periodicity of the process, the recurrence that is seen well on the time trajectories of the masses. An analogous tendency to localization is also observed as the coupling increases further. The appropriate time graphs (Fig. 3.19) also reflect the influence of the "background" and demonstrate a periodic return to almost the initial relationship between the amplitudes of the different masses. The existence of a spatially localized stationary solution makes the possibility of two, three soliton, etc., regimes evident. This possibility is illustrated in Fig. 3.20, where results are presented for a numerical integration of system (3.8.6) under the initial conditions U(l) (0) = 1
4
VI + a
2
wch 1
= au1(1) (0),
[J w~i
wch
vw
+arth
u (2) (0)
vw
(-I)i {arth
2
-1
[JW~il M (ei - ~)] dU(I) (0) 1
dr
=
du(2) (0) 1
dr
2
1
-
1M
(ei - ~) ]
} ,
=0
.
Conservation of the initial spatial distribution of the amplitudes in time indicates the weak interaction between the solitons. An analysis of the numerical results shows that in the case of strong coupling and weak spatial localization, when the impact interaction is not manifested, the first terms playa fundamental role in the power-law expansion
246
3. Continuous Systems
1.90
~
11\
I~
w
f~
I
0.95
o ~A~ ",
-0.95
-1.90
---
o
l ~y.,
-~--
L\- V t\ V t\ V t'\ 1/ V ~V ~:I I'::: / 0:
~A ...
-v, . .. A
'---~--
~ 0.5
a)
x
1
0
\11
\}
\1
4.2
T
\ 8.4
b)
Fig. 3.20. Space distribution of the displacements in two--soliton regime (a); time evolution of displacements of selected masses (b)
of the nonlinear characteristic. The long;wave approximation (3.8.3) can be used here, but without neglecting the tehn r 2 as compared with one. If the coupling along the chain is weak, while the localization is quite definite (the impact interaction case), then the total time period is determined with good accuracy by the equation of motion of the fundamental mass performing a sawtooth oscillation. We now discuss briefly the influence of damping. In the case of a "pure" soliton, comparatively little viscous friction will result in smooth damping of the process with the fundamental features conserved in a specific time. For a certain deviation from the initial conditions corresponding to a "pure" soliton, energy pumping is observed from the main mass to the adjacent mass such that the amplitudes of the latter can even grow substantially in the in!tial stage of motion. The features described above for an envelope soliton are conserved completely even in the case when impact interaction is concentrated in a number of masses along the chain. The initial conditions were given here also in conformity with the analytic solution for an envel~pe soliton. The spatial and time dependences of the displacements, as well as the impact interaction forces, reflect the synchronized motion of the masses and the quite definite localization of the process.
3.9 Nonlinear Analysis of Spatial Structures 4 r-----...,....----,
L!Q.l// w
, 2
247
6
[r=14.51
I
"\ \
- .
. . - --+-\-1 \
w
\
I
,
4
\ \
o
2
-2-~
.. "--_.~~ \
o
-------1--
I
\ -4 L...-
I I I
\
:
I / ,+/
~
v
"'-I
.A ~v
/
I
...J..-
0.5
. o vYVI
11\ ---'
x
a)
1
-2
o
0.5
x
1
b)
Fig. 3.21. Initial distribution of displacements (a); signal distribution of displacements (b)
Until now we have spoken about the evolution of the initial perturbation which is similar in shape to the soliton or the multi-soliton solution. Another extreme case is the excitation over nonlocalized modes, for instance over the spatial harmonic with a comparatively small number of half-waves. Investigation of the development of the initial perturbations by means of the first, third, and tenth harmonics indicates the tendency of the system to almost "sawtooth" configurations. This latter transformation of the vibrational process (the case of the first harmonic) can be assessed from Fig. 3.21 in which the tendency to destruction of the "teeth" with ~ubsequent localization is ~~.
'
3.9 Nonlinear Analysis of Spatial Structures 3.9.1 Introduction
In a large number of mechanical and hydromechanical systems there are stationary patterns which possess, in many situations, a coherent structure. For example, in Rayleigh-Benard convection of simple liquids both isotropic and anisotropic quasi-two-dimensional structures and normal and oblique rolls
248
3. Continuous Systems
can be detected. As shown by experimental results, convective instabilities in nematic liquid crystals are followed by transitions to normal rolls, wavy roll structures, as well as rectangular and chaotic roll structures. Although there exist some theoretical attempts to explain the observed phenomena, many problems are still open (in fact, only the normal-oblique transition has been even roughly estimated by either linear or nonlinear analysis). Additionally, this problem is a significant one in the theory of generation of vortices in superconductors, and plasmas or in liquid crystals [39, 53, 92, 129, 143, 47d, 61d]. From the mathematical point of view, the problem reduces to the nonlinear analysis of the so-called envelope (or amplitude) equation. The idea of introducing dynamics into the theory of coherent structures (governed by the envelope equation) was presented by Infeld et ai. [47d]. The crucial point of their approach included a time-dependent, supercritical structure and multiple time and space expansions. Two variants of the Ginzburg-Pitaevski universal model equations have supported their general ideas. Motivated by their observation that temporal development is more crucial for the theory of spatial structures in media such as superconductors, liquid crystals, simple fluids or plasmas than for static situations, we have outlined time-dependent terms in our mechanical analogy model of the above mentioned pattern-forming structures. In [129], the mechanical model of the buckling instability of an anisotropic elastic plate embedded in an elastic medium has been used for the interpretation of the periodic structures of anisotropic systems. Our idea is the extension of such a mechanical analogy by including real time-dependent forces (such as inertia and damping) and thu~ to examine their infuence on the stability threshold location of the possibl'e stationary patterns. 3.9.2 Modified Envelope Equation
We consider the small lateral displacement of a thin elastic plate extended in the X 1- and YI-directions, which is embedded in an elastic medium with a Duffing-type stiffness. The governing equation has the form 2 a u m at2
au
+ h at
, ( u3 ( a 4u + K,U + 1 = - Al a 4
XI
au 2
a 4u + A2 a 4
YI
a u) '
a 2u a 2u + 2A3 a 2 a 2
XI YI
2
+JL I aXI2 + JL2 aYI2
(3.9.1 )
where U(XI,YI,t) is a neutral surface of the plate, Ai (i = 1,2,3) are the bending coefficients, the loading is defined by JLI 'and JL2, m and h correspond to the mass and damping, and K, and "y are stiffness coefficients. In what follows, after rescaling of time, length and surface, we obtain 2 a v aT 2
av + c aT
+ V5
= R
{
4 2 2 a v a va v } - (1 + V' ) v - d ay4 - 21] ax2 ay2 + TV ,(3.9.2) 2 2
3.9 Nonlinear Analysis of Spatial Structures
249
where the related connections between the parameters are XI
1]
T
1/2 = ( -2AI) X, f.L I
=
A3f.L~
Alf.L~
YI
r =1_
-1,
~)1/2 = m t,
(
U
d_
1/2 = (2Alf.L2) Y, f.L I
4A4~, f.L~
(~)1/2 = 'Y .V,
=
C
-
A2f.L~ \
_ 1,
/\ I f.L2
~ (~) 1/2 , m
2
(3.9.3)
'Y
f.L~
R = 4AI~'
When the inertial term in (3.9.2) disappears, the problem reduces to the socalled "Swift-Hohenberg" model. In order to analyse the unstructured state locally (u = 0), we put v = exp{l1t
+ i{qx + pyH,
(3.9.4)
into (3.9.2) and we find that the eigenvalue characteristic equation
11
can be obtained from the (3.9.5)
in which P and q are the wave numbers of the mode considered. Furthermore, we consider the case in which the unstructured state u = 0 has one zero eigenvalue (the other cases will be consider elsewhere). Both eigenvalues are real and the second is equal to -c. The necessary condition which satisfies our requirements leads to the instability threshold defined by q~ = 1,
Pc2-0 - ,
r c = 0,
(3.9.6)
which is, for c = 1] = 0, in agreement with reference [129]. Furthermore, we take a small (positive) parameter, defined as r = e, and apply the second appropriate rescaling together with the introduction a blown-up version of the damping parameter c i.e., X = !e l / 2x 2 '
c = el / 2
Y = !e l / 2y 2'
S, (8
=
2
T = e l / 2t
t~h2"4AI~
m 'Y f.LI -
)) .
, (3.9.7)
We search for a solution of the form
v -_ e 1/2 ( VQ +1 e Vt/+ 2 eV2 + ) . .. ,
(3.9.8 )
and the dependence on X, Y and T reflects the disturbed structure. Applying well-known perturbation techniques, we obtain a recurrent set of linear partial differential equations.
250
3. Continuous Systems
For e 1/ 2 we have
a2vo R { aT2
+
SBvo} aT
= (-{I
+
V'2)2
V2
_
2 2 d a4v2 _ 2 a U2 a v 2 ) 8y4 1] ax2 a y 2
_2{ aVlax (1 + axaV12) aVl (1 d) B3Vl aVl aVl aVl ax + ay ay + + a y3 ay I
+ ( + 1]
-
)
(Bvay aVl aVl aVl) Bv Bv ax + ax 8Y ax ay 1
3 a2vo a 2vo { 2 ax2 aX2 1(
+ 2 1 + 1]
1
1 a2vo + 2aX2
) (a2vo a2vo a y 2 aX2
+
1}
a2vo a2vo 2(I + d) ay2 ay2 3
+
a2vo a2vo 8x 2 ay2
+
1 a2vo + 2 ay2
4 avo avo avo avo) ax ay ax 8Y
- 1 + Rv~ }.
(3.9.9)
We look for a solution of the form Vo = (A{X, Y, T)ei (x+lI)
+ A{X, Y, T)e- i(x+lI))/V3,
where A and A are complex conjugates, and from (3.9.9) we find 2 2 2 a A ) a A as 1 (a A 2 aT 2 + S aT = R aX2 + aY2 + A - IAI A.
(3.9.10)
(3.9.11)
The stationary solutions of (3.9.11) are sought in the form
A = Fexpi{QX + PY),
'.
(3.9.12)
and we obtain F = 0, and F2 = R-l{l- p2 _Q2). We examine both of them locally by introducing a small perturbation v (complex), A
= (F + v) exp{iZ),
which results in the equatiqn 2 a 2v av 1{ a v aT 2 + aT = RaX2
. av + 21Pay where v and
v
fj
Z=QX+PY,
+
a 2v aY2
(3.9.13)
av
+ 2iQ ax
2 2} ,
F v- F
fj
(3.9.14)
are complex conjugates. We put
= (v1ei(KX+LY) + v 2e- i (KX+LY))
I
(3.9.15)
into (3.9.15) and obtain the characteristic equat"ion
ao0'4
+ al0'3 + a20'2 + a30' + a4 =
0,
(3.9.16)
3.9 Nonlinear Analysis of Spatial Structures
251
where
ao = 1, a4 =
at
= 25,
a2 =
~2 {(K 2 + L 2)2 -
~ (K 2 + L 2) + 2F 2 + 8 2 ,
4{KQ + LP)2}
+ 2~2 (K 2 + L 2).
(3.9.17)
Applying the Routh-Hurwitz criterion to equation (3.9.11), we first analyse the threshold for instability of the unstructured state u = 0, which cor~ responds to F = O. The problem can be reduced to the consideration of the equation E =
~2
{K 2 (K 2 - 4Q2)
+ L 2 (L 2 - 4p2) + 2KL{KL - 4QP)}.{3.9.l8)
It is clearly seen that the relation K/Q = L/P = N governs the occurrence of the oblique rolls. For N < 2 all roots of (3.9.16) have negative real parts, which proves the stability of the unstructured state. The threshold of stability is found for the critical value N = 2. Investigating the second stationary solution, we find that the stability limit is defined by (3.9.19) which corresponds to the usual Eckhaus criterion: i.e., in the case of symmetry if P = Q > 1/V3, the unstructured state becomes unstable. Because the characteristic equation is of the fourth order only, one pair of conjugate roots Can occur. The examination of this possibility leads to the equation F4
+82 {~ (K 2 + L 2)2 + F 2 } + ~2F2 (K 2 + L 2) =
0,
+4(LQ + LP)2 (3.9.20)
which is never satisfied. Thus, in the case of the modified envelope equation only Eckhaus-type instability Can occur.
4. Discrete-Continuous Systems
4.1 Periodic Oscillations of Discrete-Continuous Systems with a Time Delay 4.1.1 The KBM Method One of the important problems of mechanical and automatic control engineering is active control of the oscillations of mechanical objects by means of control units, which can be treated as inertial systems with concentrated parameters and a time delay [137]. The objects to be controlled can be nonlinear mechanical systems with concentrated (futher referred to as discrete mechanical systems) or distributed parameters. The mixed situation, referred to as discrete-continuous systems, are dealt with in this chapter. In real control systems of this type, the control unit influences the object subjected to control and the state of the controlled object is monitored only in certain isolated points. It is usually possible to find controlled objects which are governed by partial differential nonlinear equations, as well as control units, which can be modelled by ordinary nonlinear differential equations. Let us consider a discrete-continuous system governed by the following equations:
8 2 u(t, x) _ (2m) 8t 2 - Lx {U(t, dy(t) ~
P
=
L
Xn + elI {X, U(t, x), y(t - JLn,
Apy(t - T p) + eF. {y(t - JL), U(t - JL), e}
(4.1.1)
p=O
subjected to the following non-homogeneous boundary conditions
L~h,j){u(t,xnlxES = eghj{y(t - JLn,
h
= 1, ... ,m.
(4.1.2)
The coordinate t denotes time and t E R; x is the vector of the coordinates and x E (G US), while S is the limiting set of G; u(t, x) is a certain scalar function determined in the set R x G, and L~h,j) is a linear operator of order 2m on Xj L~2m) is the linear differential operator of j < 2m - 1; Y and F. are vectores of an m-dimensional space; A p are constant matrices of order
254
4. Discrete-Continuous Systems
(m x m); Fl, II and
e
are functions of y(t - J.L), u(t - J.L, e), E (G US), while T p and J.L are time delays. Finally, we assume that e and J.L are small positive parameters. Thanks to this mathematical formulation of the problem, the presented analytical approach can be further used for many different discrete-continuous mechanical systems governed by (4.1.1). Thus we will continue our consideration first in a general form, and then, in order to demonstrate the physical insight of the problem, we will illustrate the method with an example from the area of mechanics [31]. The problem, including the non-homogeneous boundary conditions (4.1.2), can be reduced [31, 137] to one of homogeneous boundary conditions. Thus, we analyse the following system
a2v(t, 8t 2 x) dy(t)
dt
=
9h,j
= Lx(2m) {V(t,)x } + elI { x, { v t,) x ,y ( t -
2: Apy(t P
T p)
+ eFI {y(t -
J.L )} ,
J.L), v(t - J.L), e},
(4.1.3)
p=O
where v(t, x) fulfils the homogeneous boundary conditions L~h,,j) {v(t, X)}lxES
= 0,
h
= 1, ... , m.
(4.1.4)
From the first equation system (4.1.3), and for e = 0, we obtain L~2m){x(x)} + O"X(x) = 0,
L~h,,j) IXES
= 0,
h
= 1, ... ,m,
,
(4.1.5)
while from the other one, we obtain the characteristic equation
D(p) = det
{t
Ape- rpp
-
EP}'
(4.1.6)
p=O
In this dynamical system, oscillations will appear if 0"8 = w~. and/or if the characteristic equation (4.1.6) has the imaginary eigenvalues Pk = ±iwyk' Here we shall consider the case where 0"1 = w~ and the other eigenvalues of the first equation of system (4.1.5) amount to 0"8 1= {(p/q)wd 2 , where p and q are integers. Moreover, it is assumed that the characteristic equation (4.1.6) does not possess imaginary eigenvalues. We seek a one-frequency solution of the dynamic system (4.1.1) with the frequency approaching WI for e -+ and J.L -+ 0.' The approach suggested by Krylov, Bogolubov and Mitropolski will be used. We look for a sol ution in the form
°
K
v(t, x) = a(t)X 1 (x) cos 1/It +
L
2: 2: ek J.LlVkl {~, a(t), 1jI(t)}, k=ll=O
K
y(t)
=
L
2: 2: Ykl{ a(t), 1/1 (t)} , k=1 l=O
(4.1. 7)
4.1 Periodic Oscillations of Discrete-Continuous Systems
255
where (4.1.8)
(4.1.9)
(4.1.10)
:~ - L~2~){v(t, xl} = { ~~ Xl (xl - { (~~
r
Xl (xl
+L~2m) {Xl (xl) }a} cos,p - { 2~; ~~ + a ~:n Xl (xl sin,p + ~ ~ gk
~~
82Vkl + 81jJ2
82V kl 2 8a
l{ jJ,
(d1jJ) 2
ill
+
2
2
(da) 2 2 8 V kl d1jJ da 8Vkl d a dt + 8a8'I/J dt dt + 8a dt 2
8Vkl 81jJ
2
d 1jJ (2m) } dt2 - Lx {Vkl}'
(4.1.11)
From the second equation of (4.1.7), we obtain dy dt
K L = " " " " k l { 8Ykl ~ ~ g jJ, 8a
da _ 8Ykl d1jJ } dt 8'I/J dt .
(4.1.12)
Moreover, taking (4.1.9) into account, we calculate 2 ~ dt 2
=
{K L" k " "" LL-
g jJ,
k=l l=O
kl l dA da
}{K L" k " "" LL-
g jJ,
k=l l=O
l A kl
}
256
4. Discrete-Continuous Systems
lO dA= € 2A 10 da
ll A } dA+ € 2 J1- {dAlO - - A 11 + 10 da da
(4.1.13)
dA 20 dA lO } +O{€ k J1- I ;k+l=4, ) +€ 3 { ~AlO+~A20
2
tP ( d ) Xl (x) + L~2m){XI{X)} = dt K
{K L }2 L L€kJ1-1Bkl{a) Xl k=l 1=0
L
+2WI L L €k J1-' Bk1{a)X I = 2€w 1B lO X I + €2{2w IB 20 + B~O}XI k=II=O +2€JLW I B ll X I + 2€2 J1-{B lO B ll + w'IB2dX I (4.1.14) +2€J1-2WIBI2XI + 2€2{WI B 30 + B2lI)B lO } + O{€k J1-' ; k + l = 4), because in accordance with the first equation of (4.1.5), we have Xl (x)w? + L~2m){X{x)} = 0 and d dtP d~ dt = =
L } t;K ~L €k J1-' Akl {WIK + t; ~ €k J1-' Bkl €wlA lO + €2{w I A 20 + AlOBlO} + €JLWIA ll + €2J1-{w I A 21 +AllB lO + AlOB ll } + €J1-2 A l2WI + €3 {A 30WI (4.1.15) +A 20 B lO + A lO B 20 } + O{€kJ1-'; k + l = 4),
(4.1.16)
(4.1.17)
4.1 Periodic Oscillations of Discrete-Continuous Systems
257
Since Y and v can be expressed as the power series
y(t - JL) =
N
L
1 dny(t) n n! dtn (-JL) ,
n=O V
~ dnv(t, e) (_ )n (t _ JL, ~c) = ~ L- n! dtn JL,
(4.1.19)
n=O
and after limiting the calculation to n = 1, the functions £f, £F, Y and v can be expanded in a power series of the small parameters JL and £. Comparing coefficients of the same powers of £k JL l , we determine a sequence of reccurent linear differential equations: £ :
2
WI
a 2VlO (x, a, .,p) (2m){ } 8.,p2 = Lx VIO + 2wIB lO X Ia cos.,p +2w 1 A lO X I sin 'I/; + fE:(x, a, .,p),
WI
aYlO(a, '1/;) 8'1/;
~
L- ApYlO(a,.,p - Tpwd + FE: (a, .,p);
=
p=O
2 a2 V20(x,a,.,p) (2m){ } WI 8.,p2 =L x V20 +2wIB20Xlacos.,p +2w I A20 X I sin.,p + fE:2(x, a, '1/;), WI
aY20(a,.,p) 8'1/;
~
L- A pY20(a,.,p - Tpwd + FE: 2(a, '1/;);
=
p=O
a 2vll (x, a,.,p) (2m) { } WI a.,p2 = Lx Vll + 2wI BllXla cos 'I/; 2
+2wI All XI sin.,p + fE:p. (x, a, '1/;), WI
aYll (a,.,p) ~ a'l/; = L- ApYll (a,.,p - Tpwd
+ FE:p. (a, '1/;);
p=O
2 2 a V30 (x, a,.,p) (2m){ } WI 8.,p2 =L x V30 +2wIB30Xlacos.,p +2w 1 A 3o X I sin 'I/; + fE:3(x, a, .,p), WI
aY30(a, '1/;) ~ a,tP . = L- A pY30(a,.,p - Tpwd
(4.1.20)
+ FE: 3(a, .,p);
p=o
2 a2 Vi2(x,a,.,p) WI 8.,p2
=
(2m){ } Lx V I2 + 2w l B I2 X l a cos.,p
+2w I A I2 X I sin 'I/; + fE:p.2(x,a,.,p), WI
aYI2(a,.,p) a'l/;
=
~
L- A pYI2(a,.,p - Tpwd + FE:p.2 (a, '1/;);
p=o
258
4. Discrete-Continuous Systems
€
2
28 WI
f..£:
2
V21 (x, a, 'IjI) (2m){ } 8.,p2 = Lx V21
+2w I A 2I X I sin.,p WI
+ 2w I B 21 X l a cos 'IjI
+ fe21J.(x, a, 'IjI),
8Y21(a,'IjI) ~ 8'1j1 = L..." A p Y21 (a, 1/J -
TpW.)
+ Fe21J.(a, 'IjI);
p=o where € < f..£ and the functions fo, FO can be successively defined, as will be shown by an example. After expanding the function f(.) into a Fourier series, one obtains 00
f(.) =
L {b(.)n(a) cos n'ljl + C(.)n(a) sin n'ljl},
(4.1.21 )
n=1
where l
b(.)n (a) =
211'
! ! 2~l ! ! 2~l
dx
o
'IjI)X I (x) cos n'ljl d.,p,
0 211'
l
C(.)n(a) =
f(.) (x, a,
dx
o
f(.) (x, a, 'IjI)X I (x) sin n'ljl d.,p.
. (4.1.22)
0
If we equate the coefficients of X I (x) sin.,p and X I (x) cos.,p to zero, we obtain A kl and Bkl, which are given below: C(e)l(a) B ( ) _ b(e)l(a) A 10 () a - 2 , l O a - - :1') , ""WI a
WI
C(e 2)I(a) A 20 () a - 2w'
B
A
B
20
I
() = _ C(e 3 )I(a)
2 ' WI
30a
C(elJ.) I (a) A l l() a-2w'
B
__ C(eIJ.2)I(a) A 12 (a ) '
A 21 (a) = - C(e 21J.) I (a) ,
.
la
() __ b(e 3 )I(a)
30 a
-
2w la '
() _
b(elJ.) I (a) 2 ' wla B ()_ b(eIJ.2) I (a) 12 a - -......;....;.......;-.2wla ' B (a) = _ b(e 21J.) I (a) , 21 2wla lla--
I
2w 1
( ) b(e 2)I(a) a = - 2w '
2w 1
(4.1.23)
According to (4.1.9), we get
tP() da a = dt = €A lO
+ € 2 A 20 + € 3 A 30 + €f..£A ll + € 2 f..£A 21 + €f..£ 2 A I2 +O(€kf..£' ; k + l = 4),
w(a) =
~~
=
+ €BIO + €2 B20 + €3 B30 + €f..£B ll + €2f..£B 21 + €f..£2 B I2 +O(€kf..£' ; k + l = 4),
WI
at the initial conditions a(to) = ao, 'IjI(to) = .,po.
4.1 Periodic Oscillations of Discrete-Continuous Systems
259
From the first equation of (4.1.24) we obtain the dependence a(t), which upon introduction into the latter equation of (4.1.24) enables us to determine the dependence ..p{ a(t)}. Thanks to this, it is possible to analyse the slow transient processes leading to the steady state. The latter are analysed by assuming that da/dt = 0, which leads to the algebraic equation G(a, c:, f.L) = A lO + c:A 20 + c: 2A 30 + f.LA n + C:f.LA 21 + f.L2 A 12 = O. (4.1.24)
If the calculations are limited to order c:, we get from (4.1.24)
A lO
(4.1.25)
= 0,
which enables us to find: (a) one isolated solution; (b) a few isolated solutions; (c) no solutions. However, sometimes the phase flow of the starting equations can be very sensitive to changes in the amplitude a and/or the parameters c: and f.L. For these reasons, the full equation (4.1.24) should be taken into consideration. Now we briefly indicate the variety of problems which can be solved using this approach, and that cannot be solved by the use of a single perturbation method [30]. A. Suppose that the parameter c: undergoes slight changes, which are impossible to avoid. We want to control such changes by treating f.L as a control parameter. Inserting a = aO = const into (4.1.24) we can find G(c:, f.L, aO) = G(c:, f.L) = O. Thus, in accordance with the changes of c:, we can find the values of f.L in order to maintain a constant amplitude. Equation (4.1.24) is transformed into the form 2 A 30 c: + 2A~IC:f.L + A 12 f.L2 + 2A~0C: + 2A~If.L + A IO = 0, (4.1.26) where
A ,21
="21 A 21,
A'20
1 A 20, =2
A'11
="21 A 11·
( 4.1.27)
Equation (4.1.26) presents implicit second-order algebraic functions if A 30 , A~1 and A 12 are not equal to zero at the Same time. The form of the function is determined by the following expressions A 30 W
= det
A~1
A~o
S
=
A lO
A~1
A~o
A 12
A~1
A~1
A 10
+ A 12 , W2 2 = A 30 A lO
V -
= det (A~0)2, W ll = A 12 A lO
(4.1.28) -
(A~.)2.
.
By means of shifting the origin of the coordinate system and rotating the axes, it is possible to obtain the following functional forms (the expressions W, V, S are the invariants of such shifts and rotations): 1. V> 0, AW < O. Curve (4.1.26) is the ellipse c: 2/A 2 + f.L2/B 2 = 1. 2. V > 0, W = O. Equation (4.1.26) can be transformed to c: 2/A 2+f.L2 / B 2 = o and the solution is the point (0,0).
260
4. Discrete-Continuous Systems
3. V > 0, AW > O. Curve (4.1.26) is an imaginary ellipse (no real curve exists). 4. V < 0, W # o. Equation (4.1.26) is the equilateral hyperbola c 2 /A 2 J-L2 / B 2 = 1. 5. V < 0, W = O. The solution of (4.1.26) is the pair of intersecting lines c 2/A 2 - J-L2/B2 = O. 6. V = 0, W # O. The curve governed by (4.1.26) is the parabola J-L2 = 2pc. 7. V = 0, W = 0, W ll < 0 or W2 2 > O. Equation (4.1.26) represents a pair of parallel lines J-L2 - A2 = 0.· 8. V = 0, W = 0, W ll > 0 or W22 > O. The solutions of (4.1.26) are imaginary parallel lines J-L2 + A 2 = 0 (no real curve exists). 9. V = 0, W = 0, W ll = 0 or W 22 = O. The solution of (4.1.26) is a double line J-L2 = O. The coefficients of (4.1.26) are functions of the amplitude a and their values are determined by the functions /(.). B. Suppose that we would like to have a = a(c) and the shape of a(c) should be fixed a priori. The problem is then again reduced to implicit algebraic functions of second order.
C. Different branching phenomena can be expected. We can find the hysteresis algebraic points defined by the following equations
G(a, c, J-L) = 0, Ga(a, c, J-L) = 0, Gaa(a, c, J-L) = O.
(4.1.29)
If it is possible to eliminate the amplitude a from one of (4.1.29), then the other two enable us to find the hysteresis points. The bifurcation and isolated variety points are defined by the following three equations
G(a, c, J-L) = 0, Ga(a, c, J-L) = 0, Ge;(a, c, J-L) = O.
(4.1.30)
As mentioned above, (4.1.30) Can posses several different solutions for a. Thus, the M-multiple limit variety can be defined by the following equations
G(al' c, J-L) = 0,
G(a m , c, J-L) = 0,
4.1 Periodic Oscillations of Discrete-Continuous Systems
261
Using /..t as a parameter, we Can control the branching phenomena mentioned above. D. We can find the (e, /..t) set of parameters for which no real solutions of (4.1.24) exist. Thus, a domain of the assumed solution (4.1.7) can be defined in the two- parameter space. E. Suppose that we want to change the amplitude of oscillations, but the frequency of oscillations should not undergo any changes (or it should be controlled only by a linear pair of equations). In order to fulfil these requirements we have
= A IO + eA20 + e2A 30 + /..tAll + e/..tA 2I + /..t2 A I2 = 0, H(a, e, /..t) = BIO + eB20 + e 2B30 + /..tB ll + e/..tB2I + /..t2 B I2 = O. G(a, e, /..t)
(4.1.31)
After eliminating a from one of (4.1.31) there remains one equation which defines an implicit algebraic function of second order in e and /..t. One can freely choose one parameter and then calculate the value of the second one. Thus, by such an appropriate choice of the parameters e and /..t, the amplitude of the one-frequency oscillations will change; however, the frequency WI will always remain constant. We consider the following example from the field of mechanics [31]. An elastic beam of constant cross-section is connected by a spring k 2 with a discrete one-degree-of-freedom system (see Fig. 4.1). We assume that the
a,{3
EI~ m
M
Fig. 4.1. Self-excited vibrations of a beam connected with a one degree-of-freedom system
262
4. Discrete-Continuous Systems
linear coupling stiffness involves a time delay and that the nonlinearities, the time delay, and the amplitude of oscillations are small. Our system is an autonomous one, and the Van der Pol damping acting on the beam is responsible for oscillations. Within the framework of the usual assumptions of the elementary theory of bending we obtain the following set of governing equations:
(J4u lEI 8x4
82u
+ ml at2 = l(a -
My = -cfJ - (k()
2 8u _ _ ( A} (3u ) at - k2{y(t, x) - b(x - x)y t - /-l) ,
+ k2) + k2U(t - {.L,x),
(4.1.32)
where the damping coefficient a and (3 and the mass m are taken per unit length, and {.L is a time delay. The other standard parameters are given in Fig. 4.1. We have the following boundary conditions:
u(x, t)lx=o = u(x, t)lx=l = 0, 8 2 u(x, t) 8 2 u(x, t) = o. (4.1.33) 2 2 8x x=o 8x x=l In addition to the nonlinear mechanical system with a time delay, the governing delay nonlinear differential equations can be found in problems related to biology, blood circulation, and control systems. Therefore, we transform the dimensional equations (4.1.32) to nondimensional form. Thanks to this we reduce the number of valid parameters, and our further calculations are valid not only for the mechanical system shown in Fig. 4.1, but for other systems as well. The new nondimensional, governing equations are: 2 8 w(r, e) + 4 B4w(r, e) = (1 _ 2( '. C)) 8w(r, e) _ A ( i) 8r2 p 8e4 e w r, ~ 8r e w r, ~
+eBb(e - {)ry(r - /-l), d 2 ry( r) dry _ dr2 = -eC dr + eFw(r - /-l, e),
(4.1.34)
and the new boundary conditions are
w(e, r)le=o = w(e, r)le=l = 0, 8 2 w(e, r) B2w(e, r) 2 8 2 8r e=o r e=l
=
O.
The nondimensional parameters are defined as follows: r = nt, w = ((3a- 1)1/2 u , A = k2a- 1 l- 1 n- 1 , D = (ko + k 2)M- 1 n- 2, /-l = n{.L, e = a(mn)-l, B = k2a- 1/ 2(31/2 n- 1 , = Xl-I, p4 = E1m- 1 n- 2l- 4 , C = cma-1M- 1, F = k2ma-l/2fj-l/2n-ll-IM-l.
e
(4.1.35)
(4.1.36)
4.1 Periodic Oscillations of Discrete-Continuous Systems
263
In order to avoid tedious calculations we assume that d1J 1J(r-j.£) = 1J(r) -j.£dr'
-
w(r-j.£,e)=w(r)-j.£
8w(r,{) 8r
.
(4.1.37)
. Taking (4.1.37) into consideration, we obtain from (4.1.33) the following set of equations: 2 8 w(r, e) 4 B4w(r, e) = (1 _ 2( c)) 8w(r, e) _ A ( C) 8r 2 + p 8e4 e w r, ~ 8r e w r, ~ - d1J +eBb(e - e)1J(r) - ej.£Bb(e - e) dr'
(4.1.38)
2 d 1J(r) D __ C d1J F( C) _ F 8w (r,{) dr 2 + 1J - e dr + e r, ~ ej.£ 8r' From the first equation of (4.1.38), and for e = 0, we determine the frequency v = (n1rp)2, n E N. Limiting our calculations to n = 1, and with regard to the earlier section, the solutions of (4.1.38) are sought in the form w(e, r) = a(t) sin 1re cos 1/J(r) + eWlO(e, a, 1/J) + e2W 20 (e, a, 1/J) +e3 W 30 (e,a,t/J) + ej.£Wll(e,a, t/J) +e 2j.£W21 (e,a,1/J), (4.1.39) 1J(r) = e1JlO(a, t/J) + e21J20(a, 1/J) + e31J30(a, t/J) +ej.£1Jll (a, t/J) + e2j.£1J21 (a, t/J), where Wkl(e, a, t/J) and 1Jkl(a, t/J) are the limited and periodic (with regard to t/J) functions to be obtained. The unknown amplitude a( r) and phase t/J( r) are calculated from da
dt = eA10(a)
dt/J dt
= v
+ e2 A 20 (a) + e3 A 30 (a) + ej.£A ll (a) + e2 j.£A21 (a),
(4.1.40)
+ eBlO(a) + e2B20(a) + e3 B 30 (a)
+ej.£Bll(a) +e2j.£B21(a).
(4.1.41)
Proceeding in an analogous way, we find the sequence of recurrent linear equations e : v
282WlO 81/J2
+ P4
B4WlO . 8e4
-
. .. 2vB lOa cos t/J sm 1re - 2vA lO sm t/J sm 1re
= -av sin 1re sin 1/J + va 3 sin3 1re sin t/J cos 2 t/J -aA sin 1r{ sin 1/J cos 1/J, 2 2 8 1J10 . v 8t/J2 + D1JlO = aF sm 1re cos 1/J;
(4.1.42)
264
4. Discrete-Continuous Systems
2 28 W 20 v 8tP 2 =
48
4
+P
W 20 8e 4
-
. ~/.· C v lOa cos 'I/J sin 1l"~c - 2v A 10 sm If' sm 1l"~
2 B
82WlO B2wlO ( dA lO -2vB lO 8tP 2 - 2vA lO 8a8tP - AlO""""da
-
B~oa) cos'" sin 1r{ + (2A lO B lO + a d:~o AlO)
sin'" sin 1r{
. .. 8WlO +A lO costPsm1l"e - aB lO sLm'I/Jsm1l"e + v 8tP
(4.1.43)
- A lO a2 cos 2 tP sin 2 1l"e + 2a 2 WlOv sin tP cos tP sin 2 1l"e 3 . 2· 3 8WlO 2 2 . 2 +a BIO SIn tP cos tP sm 1l"e - 8'I/J va cos 'I/J sm 1l"e -AWlO + Bb(e - e)77lOl 2827720 B27710 8271I0 v 8tP 2 + D7720 = -2vAlO 8a8tP - 2vB lO 8'I/J2
-c v
28
871I0 8'I/J + FWlO ;
2
W 30 8'I/J2 = -
48
+P
4
W 30 8e4
-
2 B
~/.·
C
v 30 a cos If' sm 1l"~ -
. ~/.· C V 30 sm If' sm 1l"~
2 A
dA 20 dA lO ). ( """"daAlO + ~A20 - 2aBlOB20 costPsm1l"e
+{ 2A
20 B lO
+2A
IO B 20
-i\a( d:~o A20
2 dB 20 ) } 2 8WlO 8 W lO +""""daAlO sin tP sin 1l"e - A lO 8a 2 - 2 8a8tP (vA 20 8WlO dA lO 8 2W lO 2 +AlOBlO ) - AlO 8a """"da - 8tP2 (B IO + 2vB20) 8 2W 2O 8 2W 20 A 8WlO 'dB lO - 10 8'I/J """"da - 2vA lO 8a8tP - 2vB lO 8tP 2 +A 20 cos 'I/J sin 1l"e - aB20 sin tP sin 1l"e A 8WlO B 8WlO 8W20 + 10 8a + 10 8'I/J + v 8tP -2aA lO W lO cos 2 'l/Jsin 2 1l"e - a2A 20 cos 3 'I/J sin 3 1l"e (4.1.44) +Wro va sin 'I/J sin 1l"e + 2va 2W 20 sin tP cos 'I/J sin 2 1l"e +2a 2B IO WlO sin tP cos 'I/J sin 2 1l"e + a3 B20 sin 'I/J cos 2 tP sin 3 1l"e 2 8WlO 2 . 2 8WlO -a A lO da cos tP sm 1l"e-2VaWlO~cos'I/Jsin1l"e 2 8 W lO 2 . 2 2 8W20 2 2 -a BIO 8tP cos 'I/J sm 1l"~ - a v 8'I/J cos 'I/J sin 1l"e
4.1 Periodic Oscillations of Discrete-Continuous Systems
265
• 2 8 2W 21 4 B4w30 v 8tP 2W21 + P 8e4 - 2vB 21 acostPsm1l"e -2vA 21 sin 'l/J sin 1I"e = Au cos tP sin 1I"e
-aB II sin tP sin 1I"e + v 8~ I
-
a2All cos 3 tP sin 3 11"e
+2va 2W u sin 'l/J cos 'l/J sin 2 1I"e + a3 B u sin tP cos 2 tP sin 3 11"e 2 8WU 2 . 2 -va 8tP cos tPsm 1I"e - AWu + B11u O(e - e) (4.1.45) -Bvo(e - () 8;~0, 2 82 1121 v 8'l/J2
+ D1121 = -Cv 811u 8'l/J + FWII
-
F
A' 10 cos tP sm 1I"e
.. 8W1O +aF B IO sm tP sm 1I"e - Fv 8'l/J ;
eli-
v
282W11 8tP 2
2 82 1111 v 8tP 2
4 84Wu 8e 4
+P
= 0, •
(4.1.46) •
-
+ D1111 = avF sm tP sm 1I"e.
The solution of this set of equations gives: WIO = -
12~OV a3 sin 311"e sin 'l/J - 12~V a3 sin 1I"e sin 3'l/J 1 3 · 3 c . 3~/. 1152va sm 11"~ sm 'f',
Fa 1110 = (D _ v 2) sin 1I"e cos 'l/J,
e E (0,1),
W20 = Al sin 311"e sin tP + A 2 sin 1I"e sin 3tP + A 3 sin 311"e sin 3'l/J + A 4 sin 311"e cos 'l/J + As sin 11"e cos 3tP + A 6 sin 311" e cos 3'l/J,
266
4. Discrete-Continuous Systems
3 3 _ a _ a3 } Al = 8011 2 - 12801l A sin 7re - 3211 A sin 7re + 12801l A , 3 3 3 3a 3a } 1 {27a A 2 = -2 --Asin7re--Asin7re---A, 811 12811 3211 12811 3 3 3 1 {9a _ a _ a } Asin7re- 32 A sin 7re+ 1152 A , A3 = --2 7211 115211 1I 1I 1 {a
1 {a
3 }
A4
=
80112
At>
=
1 { 3 72112 1152 a
1120 =
640 '
As
=
1{9 3}
8112
64 a
,
3} ,
F FC}. c' ~/. { 211 A 10 (D _ 1I 2)2 + all (D _ 1I 2)2 sm 7r~ sm 0/ F _ a3 F IO +211aB (D _ 1I2)2 sin 7re cos 1/; - 12801l(D _ 1I 2) sin 37re sin tP
3a 3F _ a3F -12811(D - 911 2) sin 7resin 3tP - 115211(D _ 911 2) sin 37re sin 31/;, Fall
W II = 0,
A lO =
.
f'
1111 = (D _ 1I 2 ) sm 7r~ sm
~a( 1 -
136a2) ,
BlO
~/.
(4.1.47)
0/,
1
= 211 A~in 'Ire, A 20 = 0, ,
1{ 1 3 2 1 2. 2'BF . -} B20 = -~ "8 + 64 a + 811 2A sm 7re + 2(D _ 1I 2) sm7re , _ 3 a3 2. 2 3BFa 3 . 3a 3 A 30 - - 256114 A sm 7re - 6411 2(D _ 1I 2) sm 7re + 25611 2 BFa . - BFa (1 - fBa 2) . 2(D _ 1I 2)2 sm 7re + 2112(D _ 1I2) Sll17re -
CBFa - 2(D _ 1I2)2 sin 7re, B
_ 30 -
a 2113
{I'8 3 1
2 2. 2 BF . -} + 64 a + 811 2 A sm 7re - 2(D _ 1I 2) sm 7re
BF 2 • 3 - 411 2(D _ 1I2)2 A sm 7re,
All = 0, A 21
=-
B II = 0, , BFa .211(D _ 1I2) (ll + 1) sm 7re,
B21
=
O.
In the calculations we have not taken into account harmonics of order greater than three, and we have omitted powers of the amplitudes which
4.2 Simple Perturbation Technique
267
were greater than three. Let us consider the stationary state which leads to the following algebraic equation: A lO
+ e2A 3o + etLA2.
= 0.
(4.1.48)
Because we have limited the calculations to the first power of tL in our example, we can use the general discussion given earlier by substituting tL for a and considering further the implicit function (4.1.48) with regard to a and
e. From (4.1.48) one obtains 2 BF { 2 }.e v 2 (D _ v 2 ) D - v (C + 2) sm 7re
BF(v + l)tL. 1 2 -e v(D _ v 2 ) sm 7re - 16 a + 1 = 0.
(4.1.49)
We have also determined
3BF -{ 2 W = 16v 2(D _ v 2)2 sin 7re (v (C + 2) - D)
+ tL 2
BF(v + 1)2 -} 4 sin 7re ,
3BF 2 • V = 16v 2(D _ v 2)2 {v (C + 2) - D} sm 7re.
(4.1.50)
These results allow us to come to some important conclusions from the point of view of possible applications. If v 2 (C + 2) > D, then in the considered system, a one-frequency periodic solution does not exist, because W > 0. If v 2 (C + 2) = D, then W > 0, and the curve e(a) is a parabola. If v 2 (C + 2) < D and W =I 0, then the curve e(a) is an equilateral hyperbola. If v 2 (C + 2) < D and W = 0, we have two intersecting lines.
4.2 Simple Perturbation Technique To introduce the reader to a simple perturbation technique applied to discrete--continuous systems we consider equations of the form [29] 8 2 u(t., x)
Bt2
•
2
=
C
8 2 u(t., x) 8x 2
+f L.[y(t), rd
=
( e, x, u(t., x),
cp[e, y(t.), u(t. -
8u(t., x) 8u(t., x) ) Bt. ' 8x ,y(t. - r),
(4.2.1)
r, e)]
with the homogeneous boundary conditions u(t.,O) = u(t.,l) = 0,
(4.2.2)
268
4. Discrete-Continuous Systems
°
where:
is a certain nonlinear function assuming to be zero for x = and x = l; L I (y, 'Tr ) is a linear differential operator with delay of the form Ldy, 'Tr ] = 2::=0 2:~oapry(P){tl - 'Tr ), 'To = 0, 'Tr > 0;
f
y{t l - 'T) u (t I
_
=
dy{td y{tl) - 'T dtl
1
+ '2'T
2d
c) - (t C) _ 8u{tl,e) 'T, ~ - U I, ~ 'T at l
2y{t l ) dt~
... ,
l ,e) + ~2 'T2 82 y{t 8t~
...
(4.2.3)
Further calculations will be limited only to the first three terms of series (4.2.3) in (4.2.1) and we obtain
8 2u{tl, x) at~
2
=
~U{tl' x)
C
8x2
8u{t l , x) &(tl,x) dy d 2y ) +II ( c:,X,U(tI'X), at l ' 8x ,'T,y, dt l ' dt~ , 8u{t l , e) 8 2u{t l , e)) Ldy{td,'Tr ] =
.
where the functions II and
P
B{p) = LLaprppe- TrP ,
(4.2.5)
p==Or=O
and that its eigenvalues are different from zero and have purely imaginary values. This means that oscillations are not generated by the discrete system. The starting solution for the analytical approximate method, with c: = 0, 'T = 0, is of the form
U~.){tl' x)
00
= L
sin
1l"t [a?)n cos{naotd + b?)n sin{naotd] ,
n=1
y~.) {td = 0,
(4.2.6)
4.2 Simple Perturbation Technique
269
where the operator (*) denotes T or €, a?)n' and b?)n are the amplitudes, and To = 21r lao = 2llc is the period of oscillations of the linear part of the system described by the first equation in (4.2.4). For € =1= 0 and T =1= 0 in a satisfactorily close neighbourhood of zero, we seek the periodic solution of system (4.2.4) a little different from (4.2.6). Generally, the contribution of higher harmonics to the solution quickly decreases, and it is sufficient to consider only a few of the first harmonics in the calculations. The period sought is equal to
T
=
To [1 + 1](€, T)]
(4.2.7)
and evidently depends on both of the perturbation parameters. Let us introduce a new dimensionless time t according to the equation
t = 1 + TJ{€,T)t, 1
(4.2.8)
0.0
which allows us to seek a periodic solutions with period 21r. Substituting (4.2.8) in (4.2.4), we obtain the equation 2 8 2u{t, x) 2 8 u{t, x) at 2 = C 8x 2
8u{t, x) 8u{t, x) dy d 2y ) +F ( €,X,u{t,x), at ' 8x ,T,y, dt' dt2 ' &(t, e) 8 2u{t, e)) Ldy{t),Tr ] = ¢>l ( €,y{t),t,u{t,e), at ' 8t 2 '
(4.2.9)
where: F= L
{I + TJ2)l2 1r 22 c
!I,
(4.2.10)
['(1,,: 1/)1' y(p)(t) - T}(l,,: 1/) y(P+l)(t)
= ~~a...{
+~T;y(P+2) (t) }, 4> =
['(1,,: 1/) l'
'1'.
The nonlinear functions ¢> and F as well as the solution sought, y, u and TJ, are presented in the form of power series: . _
2
2
¢> - ¢>o + €¢>e + € ¢>u + + T¢>T + T ¢>TT + + T€¢>u + , F = Fo + €Fe + €2 Fu + + TFT + T2Fn + + T€Fu + , _ 2 2 Y - Yo + €y~ + € Yu + + TYT + T Yn + + T€YTe + , (4.2.11) u = Uo + cUe + €2Uu + + TUT + T2UTT + + T€U Te + , TJ -_ TJo + €TJe + € 2 TJee + ... + TTJT + T2 TJn + ... + T€TJTe + ....
270
4. Discrete-Continuous Systems
Having substituted (4.2.11) into (4.2.9), and having equated the expression representing the same powers of the small parameters T and £ as well as the same powers of their products T m £l (m,.l = 1, 2 ... ), the recurrent systems of linear equations are obtained. While solving the subsequent equations of the system, we use the harmonics balance method. Let us assume that we have determined the first system of recurrent equations, where the operator (*) means T or £. Having substituted the solutions (4.2.6) for the nonlinear functions F(*) and ¢>(*) (this time for the equation we assume tl = t and ao = 1) and having developed these functions into a Fourier series, we obtain 00
00
F(*)(t, x) = L L sin n;x [A~*2 cos(kt) p=Ok=O
+ B~~ sin(kt)],
00
¢>(*) (t) = L[Ci*) cos(kt)
+ Di*) sin(kt)],
(4.2.12)
k=O
where:
ff = ff C~*) ~ f ~f
21t'
l
A~*2 = :l
o
0
21t'
l
2 1rl
(*) B nk
F(*)(t, x) sin n;x cos(kt) dt dx,
o
.
n1rX F(*) (t, x) sm l - sin(kt) dt dx,
0
21t'
¢>(*) (t) cos(kt) dt,
=
.,
.
(4.2.13)
o
21t'
Di*) =
¢>(*)(t) sin(kt) dt.
o We seek the solutions of'the system of equations formed by comparison of the expression next to (*) in the form of N K U(*) (t, x) = L L sin n;x [a(*)nk cos(kt) n=l k=O
+ b(*)nk sin(kt)] ,
K
y(*;(t) = L[C(*)k cos(kt) k=O
+ d(*)k sin(kt)].
(4.2.14)
The solution of the first equation of system (4.2.9) is explicitly determined only when
4.2 Simple Perturbation Technique
Qe.)n{a?)8' b?)6' 11e) =
271"
ff l
271
Fe.) (t, x) sin n;x sin{nt) dt dx = O. (4.2.15)
o 0 Conditions (4.2.15) allow us to neglect the resonance terms which exponentially grow with time. Thus we obtain 2N of the equations, whereas the unknowns a?)s' b?)8 and 11e are 2N + 1. In this case, however, dealing with an autonomous system, we may assume that be.)N = o. As an example let us consider the discrete-continuous system described by the equations 2 2 8 u = (30)2 8 u{t., x) 2{t )]8u{t.,x) !:U2 8 2 + £[0003. u. ,X !:U UL.
X
~
UL.
+£8{x - x)y{t.) - £T8{x - x) dd
y
t.
8u{t., x) _ 8 2u{t., X)) _ (8U{t., X)) 3 +T ( at. 8x2 T at. '
(4.2.16)
d2 y dy 2 + 10£-d + 400y{t.) = 10£u{t., x), u{t., 0) = u{t., 1) = 0, d t. t. where for the sake of simplification of the calculations, the delay T and the small parameter £ are in the evident form (4.2.1) and x E [0,1] is the association point of the discrete system with the continuous one. In the discrete system described by the second equation of system (4.2.16) accompanied by the lack of interaction on the side of the continuous system and as a result of damping in the system, oscillations cannot occur. The oscillations are excited in the continuous system because of damping of the Van der Pol type described by the second term on the right-hand side of the equality sign. For T = £ = 0 the period of this solution is equal to To = ~ /15. We seek the periodic solution of system (4.2.16) with period T, insignificantly different from the period To. Accordingly, let us first make use of the independent variable
t. = 1 + ~~£, T) t.
(4.2.17)
Having substituted (4.2.17) into (4.2.16), we obtain 2 8 u{t, x) = 2- (I + )2 B2u{t, x) + (I + 11) [0 003 _ 2{t )] 8u{t, x) at2 ~2 11 8x2 £ 30 . u, x fJt +£
(I +'11)2 1+11 dy 900 8{x - x)y{t) - £T 8{x - x) dt
30
1 + 11 8u{t, x) (I + 11)2 B2u{t, x) +T 3O at - T 900 8x2
-T~ 1+11
2
(8U{t,X))3. fJt
'
d y 1( ) dy 4 2 £ 2dt 2 +£3 1 + 11 dt + 9{1 + 11) y{t) = 90{1 + 11) u{t,x).
(4.2.18)
272
4. Discrete-Continuous Systems
The parameters T and € are treated as independent. Assuming one of them to be equal to zero, the problem is reduced to the classical perturbation method. We assume the starting solution in the form of u(O) = u(O) + u(O) = a(O) sin1rxcost + a(O) sin1rxcost , T E: E: T y(O) =
O.
(4.2.19)
The amplitude sought, a~O), will be determined from the first recurrent equation formed by the compan~ion of expressions that are coefficients of the parameter €, whereas the amplitude a~O} will be determined from the first recurrent equation formed by the comparsion of expressions that are coefficients of the parameter T. From the first equation of system (4.2.18), having equated the expressions that are coefficients of the parameters T, we obtain
8 2u 1 8 2u 277T 82u~0) 1 8u~0) - 2= -2- -2 + + --8t 1r 8x 1r 2 8x 2 30 8t 1
82u~0)
- 900 8x2 - 30
) at" .
(8u~0)
3
(4.2.20)
Having equated the resonance terms to zero, we obtain 77T = 0.0055, a}O) = 0.044.
(4.2.21)
The solution of (4.2.20) is UT
=
•
1~8 (a~0))3 sin 31rxsint - 1~8 (a~0))3 sin1rx sin 3t +a sin 1rX cos t,
(4.2.22)
T
where the amplitude aT will be determined from the subsequent reccurent equation. This equation is of the form 8 2u TT 1 8 2u TT '2 8 2u T 2 82u~0) 8t 2 1r2 8x 2 + 1r 277T 8x2 + 1r 277TT 8x 2 1 8u~0) 1 Bu T 1 8 2u T 2 82u~0) 77T + 30 77T ~ + 30 8t - 900 8x 2 - 900 8x 2 8{u~0))3 8{u(0))2 uT -90 at + 3077T 8t
(4.2.23)
From (4.2.23), having equated the resonance terms to zero, we obtain
-2(0) _ 2_ ~ (0) 2 _ 30 77TaT 30 aT 16 (aT ) aT 2 (0) - 77Ta T - aT 77TT
0,
1 2 2 (0) 2 + 900 aT1r + 90077TaT 1r
135{ (0))5_ 135 {a(0))3=0. aT 8 77T T
+ 206
(4.2.24)
4.2 Simple Perturbation Technique
273
Solving the system of equations (4.2.24) we get: aT
= 0.00002,
17TT = -0.00006.
(4.2.25)
From the second equation of system (4.2.18), we obtain YT = O.
(4.2.26)
Let us now determine the perturbation equations formed owing to the comparison of the expressions that are coefficients of the parameter £. From the first equation of system (4.2.18), we obtain 8 (0) 82 82 (0) 82 e ~ = ~ U [0 003 _ { e(0) )2] ~ (2 2 ) at2 1r2 8x2 + 1r2 217e 8x2 + 30' u fJt' 4.. 7
2-
2-
2-
and having equated the resonance terms to zero, we obtain a system of algebraic equations. Solving it, we have a~O) = 0.12649,
17e = O.
(4.2.28)
The solution of (4.2.27) is Ue
=
ae sin{1rx) cost - 5.10- 7 sin31rxcost.
(4.2.29)
From the second equation of system (4.2.18), we obtain
d y ~ _ 2- (0) dt2+9Ye-90ue (t,x). 2
(4.2.30)
We seek the solution of.{4.2.30) in the form
Ye = be cost
+ Ce sint.
(4.2.31)
Having substituted (4.2.31) into (4.2.30), we find
be = -0.088 sin{1rx) ,
Ce
(4.2.32)
= O.
From the second equation of system (4.2.18), having equated the coefficients of £2, we obtain
d 2 yu dt 2
4
+ gYu
=
1. "3besmt
1
.
_
+ 90ae sm{1rx) cost.
(4.2.33)
We seek the solution of (4.2.33) in the form
Yu = bu cos t + Cu sint. Having substituted (4.2.34) into (4.2.33), bee
= -
8
5~ sin{ 1rx),·
(4.2.34) ~e
calculate:
u = -0.0048 sin 1rX.
(4.2.35)
C
From the first equation of system (4.2.18), having equated the coefficients of £2, we obtain 2 82 u _-:::e_e _ _1 8 u U at 2 1r 2 8x 2
+ _2217u 8
1r 1 (0) 8u~0) - 15 u e ue----at
2 e(0) U 8x 2
+ 0 0001 8 U e _
+ Yeo{x -
.
_
x).
at
1 8 {(0))2 U e e 30 u at (4.2.36)
274
4. Discrete-Continuous Systems
From (4.2.36) we finally calculate: 2 a = 0, TJu = -0.01 sin ?rI.
(4.2.37)
e
By means of analogous calculations it is possible to determine the recurrent equations occurring with the combinations €krl, where k ~ 1 and l > 1. In the case when the characteristic equation (4.1.1) does not have imaginary eigenvalues, the periodic solution is sought in the form [37] K
=
v(t,x)
L
LL€k/LlVkl{x,a(t),tb(t)}, k=ll=O K
y(t) = a(t){aei,p(t)
L
+ a:e-i,p(t)} + L
L
€k /LlYkl{ a(t), tb(t)}.
(4.2.38)
k=ll=O
where K
L
~; = LL€k/LlAkl{a(t)}, k=l l=O
dtjJ
ill =
K
w
L
+ L L €k /Ll Bkl{a(t)}.
(4.2.39)
k=l l=O
a and
a: are determined from the equations
p
p
(Ape-Tpwi - Ewi) a = 0,
L
L
p=O
(ApeTpWi
p=O
+ Ewi) a: = o.
(4.2.40)
: '.
The eigenvectors (3 and ~ of the adjoint set of equations are obtained from p
L
p
(A;e
Tpwi
+ Ewi) (3 = 0,
p=O
L
(A;e- Tpwi - Ewi) ~ = 0,
(4.2.41)
p=O
where A; are the matrices, conjugate to the A p matrices. From the first of equations (4.2.38) we have K
':: =
2 8 v _ fJt2 -
L
{;~E.I"
{a;:, (~;) + a~, (~~)},
2 ~ ~ k l{ 8 Vkl L-L-€ /L 8a
k=l l=O
+ 8Vkl (d2a) 8a
dt 2
+
2
(da)2 dt
dtjJ da + 282Vkl -
B2vkl (dtb )2 8a 2
dt
+
8Vkl (d 2tb ) 8:,p dt 2
From the first of equations (4.1.1) we obtain 2
8 v _ fJt2
L(2m){ (t z v,
x
)}
=
~ ~ k l { B2V kl L.; L.; € /L 8a 2 k=l l=O
(4.2.42)
8a8tb dt dt
(da)2 dt
}
.
4.2 Simple Perturbation Technique
82Vkl dt/J da 8Vkl (d 2 a) 8 2Vkl (d1jJ) 2 +2 8a(Jt/J dt dt + 8a dt 2 + 8a 2 dt Vkl (d 21jJ) 8 + 81jJ dt2 - Lx(2m) {Vkl} } ,
275
(4.2.43)
while from the second one we obtain dy = da (aeh/J(t) dt dt K
+ ae- I1/J(t») + ia(t) dt/J
(ael1/J(t) _ ae-I1/J(t»)
dt
L
' " ' " k l { 8Ykl da + L- L- e J.L 8a dt
+
k=l l=O
8Ykl dt/J } 8t/J dt
(4.2.44)
.
From (4.2.39) it follows that 2
d a _ 2 A dA IO dt 2 - e 10 da
2
+ e J.L
20 +e3 (dA ~ A IO da dt/J dt dt = ewA IO
(dA IO A da 11
+
dA 11 A ) da 10
dA IO A20 ) + O(e k J.L l ; k + l = +~
2
4),
2
+ e (wA20 + A IO B IO ) + eJ.LWA11 + e J.L(wA21 +A ll B IO + A IO B ll ) + eJ.L2A I2 W + e 3 (A30w +A20BIO + A 10 B20) + O(e kJ.Ll ; k + l = 4),
2 d 1jJ _ 2 dB lO A 2 dt 2 - e da 10 + e J.L +e 3
{
dB 10
(4.2.45)
{
dB IO A da 11
dB 20
~A20+ ~AIO
+
(4.2.46)
dB ll A } da 10
} +O(ek J.L l jk+l =4),
(4.2.47)
(4.2.48)
(4.2.49)
t _ ) = ~ .!.- dny(t) (_ )n Y( J.L L- n! dt n J.L, n=O
lI(t - J.L {) = ,
~ .!.- dnll(t, {) (_II.)n L- n! dtn fA"
n=O
(4.2.50)
276
4. Discrete-Continuous Systems
then the function ef and eF can be expanded in a power series of the small parameters /.L and c. Further calculations were carried out for n = 1 (YI -/.L(dy/dt) and VI = -/.L(dv/dt)) and under the assumption that N
•
_ _
VI -
L
~~ k l /.L L- L- e /.L
da 8a dt
{ 8Vkl
k=1 l=O
in
= -I' { _
~~ (",,''''(t) + tie-I'" (t»
~~
k l (8 Y kl
da
8Vkl
+
d'I/J }
81/; dt
I'a(t) { ~; i (o<e'''' (t)
-
8Ykl
(4.2.51)
' - tie-I'" (t»
d1/;)}} .
8a dt + 8'I/J dt
/.L L- L- e /.L k=ll=O
The necessary derivatives of the functions f and F were calculated at the point /.L = e = 0 and Vo = 0, Yo(t) = a(t){aeiw(t) + ae-iw(t)}. The sequence of recurrent linear differential equations obtained are of the form
e:
2
W
B2 VlO81/;2 (x, a, 'I/J)
+ f e,
= L(2m) {lI; } x
10
p
- -iW) , W8YlO(a, &¢ 1/;) = ~ L- A pYIO (a, 'I/J - TpW ) - A 10 (aeiw + ae p=o p
e2
-iaBIQ(aeiW - ae- iW ) - LTpA p{ AlO(aei(w-Tpw) (4.2.52) p=o +ae-i(W-TpW)) - iaBIQ(aei(~-TPW) - ae-i(W-TpW))} + Fe; , 2 w 2 8 V20 (x, a, 'I/J) = L(2m) {l/. } +' f 81/;2 x 20 u, p
'I/J) = """ W8Y20(a, &¢ L- ApY2O(a, 'I/J - TpW) - A 2O (ae'l11' p=o
+ ae- 1'11' )
p
-iaB2O(aei~ - ae- iW ) -
L A p{ TpA20(aei(W-TpW)
(4.2.53)
p=o +ae-i(W-TpW)) + iaB2O(aei(~-Tpw) - ae-i(W-TpW))} + F u 2
W28 Vj, I (x, a,1/;) 81/;2
=
;
L(2m) {V; } + f x 11 ep.,
p
(a, 'I/J) = """ W8Yll&¢ L- ApYll (a, 'I/J - TpW) --:- All (ae1'11' + ae- 1'11' )
p=o p
-iaB ll (aeiW - ae- iW ) -
L
TpA p{ All (aei(W-TpW)
(4.2.54)
p=o
+ae-i(W-TpW)) - iaB u (aei(W-TpW) - ae-i(W-TpW))} + Fep.;
4.2 Simple Perturbation Technique
W
2 82V3C1{X, a, tP)
= L(2m){TT } +
8tP 2
x
"30
t
u
3
277
,
p
W 8Y30{a, 8t/! tP) = '" L- ApY30 (a, tP
- TpW ) -
A 30 (aeiw + o:e - -iW)
p=O p
-iaB30 {aeiW - ae- iW )
- LTpAp{ A 30 {ae i (W-TpW)
(4.2.55)
p=O
+ iaB 12 (aei(W-TpW)
+ae-i(W-TpW))
2
W
- ae-i(W-TpW))}
+ F e3;
82V12{x,a,'I/J) _ L(2m){l!, } +1
8tP 2
-
12
x
e1J.2,
p
" ApYI2{a, 'I/J - TpW) - A 12 {ae1'w + ae- 1'w ) W8YI2(a, 8,p 'I/J) = 'Lp=O p
-iaB I2 {ae iW - ae- iW ) -
LAp {TpA I2 (aei(W-TpW)
(4.2.56)
p=O +ae-i(W-TpW)) + iaB I2 {ae i (w-Tpw) - ae-i(W-TpW))} +
F e1J.2;
p
-iaB21 (aeiW -
ae- iW ) -
LA
p { Tp
A 21 (aei(W-TpW)
(4.2.57)
p=O +ae-i(W-TpW))
+ iaB 21 (aei(W-TpW)
- ae-l(w-Tpw))}
Here
le
3
_ 8 21 2 81 8v2 Vio + 8v V20 +
m
L l=1
8 21 2 8 2 Y(10)l + y,
81 Y(20)l L a l=1 Yl m
+ Fe 2 w
278
4. Discrete-Continuous Systems
4.2 Simple Perturbation Technique l aF { - AIQ(ae'w +2 ~ L- ~
n=1
+ae- 'w
279
I
)
YIn
- -iW) - --w 8yll } -aB 10 (Q,eiw + a:e at/; . To achieve a complete ordering of all the recurrent equations we take the additional condition that €i-I < Ii, where i is a positive integer. After expanding the function f(.) into a Fourier series, one obtains 00
f(.) =
00
~ ~ {b(.)sn(a) cosnt/; + C(.)sn(a) sinnt/; }Xs(x),
(4.2.59)
s=1 n=l
where l
b(.)sn(a) =
211'
JJ 2~l J J
2~l
dx
o
(x, a, t/;)Xs(x) cosntj.J dtj.J,
f(.)
(x, a, t/;)Xs(x) sin ntj.J dt/;.
0 211'
l
C(.),m(a) =
f(.)
dx
o
(4.2.60)
0
The functions F( *) are expanded into a complex Fourier series (4.2.61 ) n=-oo
where
C(.)n(a) =
2~
21t'
J
F(.)e inW dt/;,
n = ±1, ±2, ...
(4.2.62)
o We describe a procedure of solving the recurrent set of ODE's based on (4.2.52). In order to avoid terms ascending unrestrictedly in time in these equations, the following conditions must be satisfied: -{A lO (a)
+ iaB lO (a)} {("', m+ ~ 7". (Ap"" {3)e- T • Wi }
+(C(lo)l(a),f3) :...- O.
(4.2.63)
Here (a, b) denotes the scalar product. By equating to zero the real and imaginary parts of (4.2.61), we obtain two equations to determine the quantities AIQ(a) and BIQ(a). Then we can find VIQ and YIQ, and further successfully solve the recurrent set of equations. Analytical conditions for the existence of a two-parameter family of periodic and quasiperiodic orbits in autonomous and non-autonomous systems Can be found in [32, 35, 36, 38].
280
4. Discrete-Continuous Systems
4.3 Nonlinear Behaviour of Electromechanical Systems 4.3.1 Introduction UsuaHy the dynamics of nonlinear discrete-continuous systems governed by ordinary and partial differential equations (the case considered here) causes some difficulties in nonlinear analysis. It is often brought about by the timeconsuming numerical techniques used to find the solution of the partial differential equations. Additionally, real physical systems possess many parameters which can be changed over wide ranges and in practice direct simulation of the governing equations is costly and tedious. For simple dynamical systems, averaging formulas can be derived without computers. However, in the case of complex systems (such as the example considered in this book) this classical approach leads to serious difficulties. Therefore, the idea of applying the averaging technique has been supported by symbolic computation with the use of the Mathematica package. A program has been written in the it Mathematica language which has yielded the averaged equations. This set of equations has been transformed (using one of the Mathematica options) to Fortran expressions, and further, a numerical analysis has been carried out. Admittedly, in this section, on the one hand, the electromechanical system serves as an example for a systematic strategy of solving many other relative problems, which can be found in nonlinear mechanical, biological or chemical dynamical systems. It consist~ of a few steps: 1. Dynamical equations are derived. 2. The averaging method is proposed and the program for symbolic computation yields the averaged equations (AVE). 3. Further systematic study of the obtained ODE's is developed. The model was first discussed by Rubanik [137], where attention was focused on the averaging procedure only starting with the governing equations. Here, the system under consideration is discussed in some detail, and the symbolic computation is used to obtain the differential amplitude equations based on the application of the Mathematica package. Contrary to the approach in [137], also the numerical analysis of the averaged differential equations found is carried out to show interesting nonlinear phenomena. The averaging method, also based on the symbolic computation, has been proposed by one of the authors earlier [21d, 22d]. Here two parts of the project are clarified. First, more attention is paid to the discussion of the electromechanical nonlinear system including its electrical model. Second, instead of an approximation of the time delay function by a Taylor series, a Galerkin method is applied. As has been pointed out by Elsgolc [33d], the truncation of the Taylor series to one term, as well as the use of more Taylor terms in the function with the time delay approximation, does not lead to an improvement in accuracy of the numerical calculations. For this reason a new approach, based on a Galerkin approximation, which does not possess any
4.3 Nonlinear Behaviour of Electromechanical Systems
281
limitations because of the delay magnitude, is used. Third, the qualitatively new numerical results are discussed and illustrated now in comparison to the author's previous works [34, 24d, 25d]. 4.3.2 Dynamics Equations
A string (a continuous system) is embedded in a magnetic field. For a certain set of parameters the stationary position of the string becomes unstable and the string starts to vibrate. Because the string possesses an inductance L, a resistance R and a capacitance C the movement in the magnetic field causes the occurrence of voltage and current. The amplifier controls the change of the current amplitudes with the time delay as the experiment shows. Figure 4.2 presents the electromechanical model and its electric scheme.
c s L
~u 'X
F, P R
U;up",! N
Uoutput
a)
b)
Fig. 4.2. Scheme of a string embedded in a magnetic field (a) and its electrical model including an amplifier (b)
The magnetic induction B(x) acting along the string generates voltage at the ends of the string according to the following equation
J () l
Uinput () t =
B
X
8u(t,x)
at
dx,
(4.3.1)
o where x is the spatial coordinate, t denotes time, u(t, x) is the amplitude of oscillations of the string at the point (t, x) and 1 is the length of the string. The amplifier gives the output voltage Uoutput(t)
where
hi (i =
=
hIUinput(t) -
h2Uinput(t),
1,2) are constant coefficients.
(4.3.2)
282
4. Discrete-Continuous Systems
The current oscillations including the time delay in the amplifier are governed by the equation
i(t) + 2Ai(t) + kI(t) = Uoutput(t - v)
(4.3.3)
where:
2A = RL- I ,
k = (LC)-I,
Uoutput(t) = hI Uinput(t) - h2Ui~put (t)
hI
= L -1-hI,
h2
(4.3.4)
= L -1-h2 ·
In the above expresion, the "dot" denotes differentiation with respect to t, I(t) denotes the changes of the current and v is the time delay. The changes in time of I(t) and the changes in x of B(x) play the role of the force acting on the string, whose oscillations are governed by the equation
8"':i:; x) _ t?8"~~ x) ~ ~ (2ho8u~ x) _ B(X)I(X)) ,
(4.3.5)
where h o is the external damping coefficient, p is the mass density per unit length, c2 = F / p, F denotes the string's tension and € is a small positive parameter. The frequencies of free oscillations of the string are given by W s = (s = 1,2, ...), and the homogeneous boundary conditions are as follows:
1rcs/l
u(t,O)
= u(t, l) = O.
(4.3.6)
4.3.3 Averaging
<
•
Our considerations are limited to first· order averaging. For € = 0 the solution to (4.3.5) is given by Uo
. (1rX) = al cos(wlt + (}I) sm -l-
+ a3 cos(3wlt +
(}3)
. (31rX) -l-
sm
(4.3.7)
where aI, a3 are the amplitudes and (}I, (}3 are the phases. For € =I- 0, yet small enough, the solution to (4.3.5) is expected to be of the form u = Uo +€UI(x,al,a3,(}I,(}3) +h.o.t.
(4.3.8)
where "h.o.t." denotes higher-order terms. Supposing that B(x) is symmetric with respect to the ends of the string, i.e., B(x) = B(l- x), we take
B
= B, sin C"t) + B 3 sin (3~X) .
(4.3.9)
From (1) we obtain Uinput =
-~BlaIWll sin(wlt + (}.) - ~B3a3wIl sin(3w l t + (}3)
(4.3.10)
4.3 Nonlinear Behaviour of Electromechanical Systems
283
and the right-hand side of (4.3.3) is calculated using the symbolic calculation: Uoutput(t -
v)
= -~B.a.h.lw~ cos(!liIO ) - ~B3a3h.lw~ COS(!li30)
+~B~a~h2l3wt
cos(!li.o) sin 2 (!li30)
(4.3.11)
27 2 B a.2 a3h2l 3w.4 cos ( !li30 )sm . 2(!li ) +gB. .0 3
+~B~B3a~a3h2l3wtsin (2!li.0) sin (!li30 ) + 81 B.B~a.a5h2l3wt sin (2!li30) sin (!liIO )
8 27 2 2 3 4 ( ). 2 ( +gB.B3a.a3h2l w. cos!li1o sm !li30 )
243 3 2 3 4 . 2 + 8 B3a3h2l w. COS(!li30) sm (!li30) , where:
= wt + O.
!li1O !li30
=
- vt, 3wt + 03 - 3vt.
(4.3.12)
From (4.3.11) we take only the harmonics sin(iwt), cos(iwt) (i = 1,3), and therefore Uoutput (t
- v)
= b. c cos wt + b. s sin wt + b3c cos 3wt + b3s sin 3wt
(4.3.13)
where: b. c = bls
=
AcosO~ - 392B~B3a~a3h2l3wtcOS(20~ -
-A sin 0; - :: B~B3a~a3h2l3wtcos(20; - 0;),
b3c = A cosO; b3s
A
=
0;),
-A sin 0; -
1
332B~B3a:h2l3wt cos(30;), 332B~a:h2l3wtsin(30n, 2
= -"2B.a.h.lw.
3 3 3 3 4 27 2 2 3 4 + 12B.a.h2l w. + 16B.B3a.a3h2l w.,
2 2 3 4 243 3 3 3 4 -"29 B 3a3 h .lw.2 ~ 27 16 B. a. a3 h 2l w. + 32 B 3a 3h 2l w., O~ = O. - vt, OJ = 03 - 3vt.
C
=
(4.3.14)
The solution to the linear equation (4.3.3) has the form Io(t) =
L
{(bicMi - bisNd cos iwt
i=.,3 + (bicNi - bisMi) sin iwt}
+ h.h.
(4.3.15)
where the abbreviation "h.h." denotes higher harmonics which are not taken into account, and Mi, N i are given below:
284
4. Discrete-Continuous Systems
Mi = Ni =
k - i2W~
(k - i2W~)
2
+ 4i2 >..2W~ '
2>" iw l
wn
(k - i 2
(4.3.16)
+ 4i 2 >..2w ?
2 '
The further analysis is straightforward for the perturbation technique. Because B(x) and I(t) are defined, therefore (4.3.5) can be solved using a modified classical perturbation approach (it is assumed that UI (x, alt a3, (h, (3) is a limited and periodic function). Substituting (4.3.8) for (4.3.4) and taking into account that ai = ai(t) and Oi = Oi(t) (i = 1,3) slowly change in time, the following resonance terms are calculated from the right-hand side of (4.3.5) (further referred to as Ri) l 211'
~e =
:l JJR
sin 1T";X cos !PiO d!PiO
o Ria
= :l
0 l 211'
JJ
o !PiO
(4.3.17)
R sin 1T";X sin!PiO d!PiO
0
= iw + 0i,
(4.3.18)
i = 1,3
where Ric, Ria correspond to the coefficients of cos i!PiO and sin i!PiO, respectively. The comparison of the coefficients of cos i!PiQ and sin i!PiO and generated by the left-hand side of (4.3.5) to those defined by (4.3.17) leads to the following averaged amplitude equations ; I
· ehoal al = p · ehoa3 a3 = -
P
eBl . -2- {(bleMI - blaNI) sm 0 1 {JW eB3 . -2- {(b 3e M 3 - b3a N 3) sm03 {JW
· eB I 01 = - 2 {(bleMI ..... blaNI) cos 01 al{JW · eB3 03 = - 2 {(b 3e M 3 - b3a N I) cos 03 a3{JW
+ (bleNI
-
blaMI) cos 0d ,
+ (b 3e N 3 -
b3a M 3) COS03} ,
+ (bleNI
•
-
blsMI) sm 0d ,( 4.3.19) .
+ (b3e 1V3 - b3a M 3) sm 03} .
The analysed set of equations has some properties which can cause difficulties during numerical analysis. First of all, this is a stiff set of equations (note tbe occurrence of a3 in the denominator ofthe last equation of (4.3.19)). As is assumed by the averaging procedure, the amplitudes ai and the phases Oi change with time very slowly, and a long integration to trace the behaviour of the system is required. For the further analysis of the time dependent solutions we transform (4.3.19) into the amplitude equations. For this aim we assume
4.3 Nonlinear Behaviour of Electromechanical Systems
Uo =
(Yl COSWlt + Y2 sinwlt) sin
285
(~X)
+(Y3 cos w3t + Y. sin w3t) sin
(~x)
.
(4.3.20)
Comparison with (4.3.7) yields the following relations: Y1 (t) = al (t) cos 81 (t) Y2(t) = -al (t) sin 8 1 (t)
(4.3.21)
Y3(t) = a3(t) cos 83(t) Y4(t) = -a3(t) sin83(t).
In what follows, the set of the amplitude differential equations has the form
Yi (t)
= al (t) cos 81 (t) - al (t)B I (t) sin 81 (t),
Y2(t) =
-al (t)
sin 81 (t) -
al
(t)B I (t) cos 81 (t),
(4.3.22)
Y3(t) = a3(t) cos 83(t) - a3(t)B3(t) sin 83(t), Y4(t) = -a3(t) sin 83(t) - a3(t)B3(t) cos 83(t)
where
ai
and 8 1 are given by (4.3.19) and
8,
= arctan ( - ~: ),
al
= (y12 + Yl)1/2,
83 a3
= arctan ( - ~:) ,
(4.3.23)
= (y32 + Yl)I/2.
4.3.4 Numerical Results We consider both time-dependent and time-independent solutions. In order to get the stationary solutions we solve the nonlinear algebraic equations obtained from (4.3.19). For this a Powell hybrid method and variation of Newton's method have been used. It takes a finite-difference approximation to the Jacobian with high precision. Figure 4.3 presents an example of numerical calculations for the following fixed parameters: l = 0.1, w = 30.0, A = 0.01, k = 35.0, h 2 = 5.0, p = 1.0, ho = 0.005, II = 5.0, B I = 5.0, B3 = 1.0. hi has been taken as the control parameter. The increase of hi damps the value of the first harmonic oscillations. However, during the change of hi the amplitude a3 as well as the phases 81 and 83 remain constant. Now (4.3.22) and their time-dependent solutions will be analysed. The system of equations is stiff and the Gear method routine from the IMSL Library is used to solve the problem. Let us consider the following set of parameters: l = 0.1, W = 190.0, A = 0.1, hI = 0.58, h2 = 0.02, II = 0.1, p = 1.0, Bl = 6.3, B 3 = 0.08, € = 0.05. Calculations have been performed with TOL = 10- 8 , which is proportional to the calculation step. As can be seen from Figure 4.3a, the variables YI,2(t) change in an oscillatory manner, whereas Y3,4 (t) decay exponentially. This allows the interpretation that the ampli-
286
a)
4. Discrete-Continuous Systems 0.020
"I ...............
al
a3 0.018--~----~~-~,-~"'~~----.j~f--.-a-' 1---
0.016 ~-0.014
'---1
----_____
-'T.---.-.-----.-.. . .-.-. --~
0.012·--·'-·~-
I------- - - -
-
--
----=9"'"-~;;::::_=1
. - - -~
------~
--
a3
0.010 I-----+------+-------.---~----+_-------____i 0.008 "-
--
---1-.---. ! _.
0.006 0.004
------1--------
-_._.
--.-
0.02
0
--I--.-------\-- - - - ' - " -
0.04
-'
0.06
-~-
"
..
0.08
~--
---'
hi
0.10
b) (2) -- -- - -
(2.5) - --
-~---.--
t I
-
l
(3)
(3.5) (4)
I-- -.- -
-~-'-
-
i
-~--.- -
----I--~
I --.-
L..-_ _---'-I
o
----
~.- --~--
-------- -
(Ji-- __
.----
l-
....1.-_ _- - . 1
-
--.
-
-------
~
.
.-
_
---'-_ _- - - - - '
0.02 0.04 0.06 0.08 Fig. 4.3. Amplitudes (a) and phases (b) versus parameter hI
hi
0.10
tudes of the first and the third modes of the string behave quite differently. Furthermore, it implies that quite different analytical types of solutions can be assumed from the mathematical point of view. The other figures illustrate the change of the mode amplitudes with the incre~e of the coefficient h2 . As can be seen from these figures, the increase in the nonlinear term h2 leads to the extension of transient periodic oscillations. Finally, let us consider the following fixed parameters: l = 0.1, W = 900.0, k = 7250, A = 0.1515, hI = 5.48, h2 = 65.0, P = 1.0, B I = 0.65, ho = 0.00001, B 3 = 0.089, € = 0.009, II = 0.00001. Strong nonlinear behaviour is observed. In the beginning all variables do not exhibit oscillatory behaviour. After the time of about 20000 units a sudden occurrence of strong nonlinear oscillations of YI,2(t) Can be seen, whereas the variables Y3,4(t) do not change in an oscillatory manner again (Figure 4.4a). Increasing the time delay a strange
4.3 Nonlinear Behaviour of Electromechanical Systems
287
~'Y;'YJ'~
a)
0.12 0.10 0.08
.~-+---~-~-~
-----
0.06 0.04 0.02
oL:..=---l-~.--L-=:~J=~:::~..~..~..===l""I!l!I-----1 (0.02) (0.04)0
100000
200000
300000
400000
500000
600000
time
b)
~'Y;'Y:I'Yt 0.15 - - - - + ~.- ------ -
0.10 0.05
"'--Y4-~---------+-·_-~+------- - - -
.........': •
}J:·
.. a.
··~.":':."::.-::oIftolft
3
o '\y
'. 2
". (0.05) --- -~--~-.. (0.10) 0
-.
---->........
........
100000
-~--
200000
300000
400000
500000
600000
time Fig. 4.4. Time evolution of the amplitudes with an increase of coefficient
h2:
(a)
h2 = 0.02; (b) h 2 = 0.04
transitional state is observed: strong nonlinear oscillations Yl,2(t) vanishing in time are shown. The amplitude Y3 decreases linearly, and its derivative remains constant. This means that the first and the third mode amplitudes behave qualitatively differently. To summarize, the analysis is focused on the numerical observation of the averaged differential equations derived from the dynamical examination of the string-type electromechanical generator. Considerable attention is paid to the derivation of the averaged equations through the application of a modified perturbation technique supported by symbolic computations. The obtained set of equations is nonlinear and stiff. The numerical calculations based on solving the initial value problem have been performed to reveal some interesting results, which are here briefly summarized.
288
c)
4. Discrete-Continuous Systems ~,Y;,Y;,~ 0.15 ...-------,-----,----r-----,------,------,
0.1 0.05 0 (0.05) (0.1) (0.15)
d)
0
100000
200000
300000
400000
500000
600000
time
~,Y;,Y;,~ 0.15 ,,
0.1
I
--
,I
0.05 0 (0.05) (0.1) (0.15) 0
e)
1ססoo0
200000
30ססoo
400000
500000
600000
time
~,Y;,Y;,Yt 0.15 .----,------,------r------.------..----,
0.1 0.05
o (0.05) t--\----l~H_+-_H~-\-I-_____+____I__~~\__-+____::_--I (0.1) 1----'\--,f+-t-l----+--\----I-4r.-V-~~~--~--.I (0.15)
L - - _ - - - - L_ _--L-_ _~_ _- ' - -_ _..i...._._
o
100000
200000
300000
400000
500000
_..J
600000
time
Fig. 4.4(continued). Time evolution of the amplitudes with an increase of the coefficient h 2 : (c) h 2 = 0.05; (d) h2 = 0.055; (e) h 2 = 0.0554
4.3 Nonlinear Behaviour of Electromechanical Systems
289
~,Y;,r;,~
a)
2
1.5 1
0.5
-
~
..-.
~ tA~., .....
... -_....-..-
~.-
(1)
- ...
....-•....•
o
10000
5000
15000
"i
~·:·::~..
y
25000
.
30000
35000
time
._
l~
,
!~
I
·.. ···h-···
"', I
1':'
~· . ·. •·.. ~ x..... r--...... .- ··
,'\Y,4
T
"'r;l
'JI'.
.
,.
--
-.. . .
_.
- ~ - -- ...----I----+--.--.-~c----~--+--
(1) (1.5)
20000
•
4
........,
(0.5)
"
f - - - - - +----~-~+--~+-~~+_--+_---_+_-__1
. . R o
0.5
.... /c. ", I} .. ....... r;1 ~
~
f
-
~,~,r;,~ 1.5 ... 1
.....-1"; ..... .........,
""
-- -----t
(0.5)
b)
----
'c-
~
o
(1.5)
~ l'1
-
1------
--t--------j
-~ - --------+---------+~_.-- - -+-----+----+---~---I
L.-_....!.-_-----J._ _....L..-_---l--_ _" - - _ . . . . l . - _ - - - - '
o
10000
20000
30000
40000
50000
Fig. 4,5. Strong nonlinear amplitudes oscillations: (a) v
60000
70000
time
= 0.00001; (b)
v = 0.0001
It has been shown that in the case of stationary solutions the increase in the control parameter hI damps the first harmonic oscillations. The amplitude a3 and the phases remain constant during the change of the control parameter. In the case of time-dependent solutions interesting nonlinear behaviour has been reported in general. Amplitudes of the first two modes behave in a qualitatively different manner. YI,2(t) change in time with oscillations, whereas Y3,4(t) decay exponentially. Furthermore, strange nonlinear phenomenon has been exhibited. After a long nonoscillatory transitional state strong nonlinear oscillations suddenly occur.
General References
[1] Abdel-Ghaffar A.M., Rubin L.L, Nonlinear free vibrations of suspension bridges: theory and applications. J. Engn. Mech. Amer. Sol. Civil Engn, vol. 109, No I, 1983, 313-345. [2] Ambartzumyan S.A., General theory of anysotropic shells. Nauka, Moscow, 1974 (in Russian). [3] An album of fluid motion. Van Dyke M. (ed.). Parabolic Press, Stanford, California, 1982. [4] Andersen C.M., Geer J.F., Power series expansion for frequency and period of the limit cycle of the Van der Pol equation. SIAM J. Appl. Math, vol. 42, No 3, 1982, 678-693. [5] Andersen C.M., Geer J.F., Dadfar M.B., Perturbation analysis of the limit cycle of the free Van der Pol equation. SIAM J. Appl. Math, vol. 44, No 5, 1984, 881-895. [6] Andrianov LV., On the theory of Berger plates. PMM, J. Appl. Maths Mech, vol. 47, No 1, 1983, 142-144. [7] Andrianov LV., The use of Parle approximation to eliminate nonuniformities of asymptotic expansion. Fluid Dynamics, vol. 19, No 3, 1984, 484-486. [8] Andrianov LV., Construction of simplified equation of nonlinear dynamics of plates and shallow shells by the averaging method. PMM, J. Appl. Math Mech, vol. 50, No I, 1986, 126-129. [9] Andrianov LV., Constituting equations of structurally orthotropic cylindrical shells. J. Appl. Meeks Technical Physics, vol. 27, No 3, 1986, 468-472. [10] Andrianov LV., Application of Parle.approximants in perturbation methods. Advances in Mechs, vol. 14, No 2, 1991, 3-25. [11] Andrianov LV., Laplace transform inverse problem: application of two-point Parle approximant. Appl. Math. Lett., vol. 5, No 4, 1992, 3-5. [12] Andrianov LV., Blawdziewicz J., Tokarzewski S., Effective conductivity for densely packed highly condensed cylinders. Appl. Physics A, vol. 59, 1994, 601-604. [13] Andrianov LV., Gristchak V.Z., Ivankov A.O., New asymptotic method for the natural, free and forced oscillations of rectangular plates with mixed boundary conditions. Technische Mechanik, vol. 14, No 3/4, 1994, 185-193. [14] Andrianov I.V., Ivankov A.O., New asymptotic method for solving of mixed boundary value problem. Intern. Series Numer. Maths, vol. 106, 1992,39-45. [15] Andrianov LV., Ivankov A.O., On the solution of plate bending mixed problems using modified technique of boundary conditions perturbation. ZAMM, vol. 73, No 2, 1993, 120-122. [16] Andrianov LV., Kholod E.G., Non-linear free vibration of shallow cylindrical shell by Bolotin's asymptotic method. J. Sound Vibr., vol. 160, No 1, 1993, 594-603.
292
General References
[17] Andrianov LV., Krizhevsky G.A., Investiga.tion of natural vibrations of circular and sector plates with consideration of geometric nonlinearity. Mechs of Solid, vol. 26, No 2, 1991, 143-148. Andrianov LV., Krizhevsky G.A., Free vibration analysis of rectangular plates [18] with structural inhomogenity. J. Sound Vibr., vol. 162, No 2, 1993, 231-241. [19] Andrianov LV., Loboda V.V., Manevitch L.I., Formulation of boundary-value problems for simplified equations of the theory of eccentrically reinforced cylindrical shells. Soviet Appl. Mechs, vol. 11, No 7, 1976, 726-730. [20] Andrianov LV., Manevitch L.L, Calculation for the strain-stress state in an orthotropic strip stiffened by ribs. Mechs of Solids, vol. 10, No 4, 1975, 125129. [21] Andrianov LV., Manevitch L.L, Approximate equations of axisymmetric vibrations of cylindrical shells. Soviet Appl. Mechs, vol. 17, No 8,1982,722-727. [22] Andrianov LV., Manevitch L.L, Asymptotology 1: Problems, ideas and results. J. Natural Geometry, vol. 2, No 2, 1992, 137-150. [23] Andrianov LV., Starushenko G.A., Solution of dynamic problems for perforated structures by the method of averaging. J. Soviet Maths, vol. 57, No 5, 1991, 3410-3412. [24] Andrianov LV., Starushenko G.A., Asymptotic methods in the theory of perforated membranes of nonhomogenous structures. Engn. Trans., vol. 43, No 1-2, 1995, 5-18. [25] Arimoto S., Learning control theory for robotic motion. International Journal of Adaptive Control and Signal Processing, vol. 4, 1990, 549-564. [26] Atadan A.S., Huseyin K., Symmetric and fiat bifurcations, an oscillatory phenomenon. Acta Mechanica, vol. 53, 1984, 213-232. [27] Atadan A.S., Huseyin K., On the oscillatory instability of multiple-parameter systems. Int. J. Engng. Sci., vol. 23, No 8, 1985, 857-873. [28] Awrejcewicz J., Determination of the limits of the unstable zones of the unstationary non-linear mechanical systems. Int. J. Non-Linear Mechanics, vol. 1 23, No 1, 1988, 87-94. '. [29] Awrejcewicz J., Determination of periodic oscillations in nonlinear autonomous discrete-continuous systems with delay. Int. J. Solids Structures vol. 27, No 7, 1991, 825-832. [30] Awrejcewicz J., Vibration control of nonlinear discrete-continuous systems with delay. Proc. 1st Intern. Conf. on Motion and Vibration Control, Yokohama, September 7-12, 1992. Eds. Seto K., Yoshida K., Nonami K., 958-963. [31] Awrejcewicz J., Stationary and nonstationary one-frequency periodic oscillations in nonlinear autonomous systems with time delay. J. Tech. Phys., vol. 34, No 3, 1993, 275-297. [32] Awrejcewicz J., Analytical condition for the existence of an implicit twoparameter family of periodic orbits in the resonance case. J. Sound Vibr., vol. 170, No 3, 1994, 422-425. [33] Awrejcewicz J., Modified Poincare method and implicit function theory. In: Awrejcewicz J. (ed), Nonlinear dynamics: new theoretical and applied results. Akademie Verlag, Berlin, 1995, 215-229. [34] Awrejcewicz J., Strange nonlinear behaviour governed by a set of four averaged amplitude equations. Meccanica, vol. 31, 1996, 347-361. [35] Awrejcewicz J., Someya T., Analytical condition for the existence of twoparameter family of periodic orbits in the autonomous system. J. Phys. Soc. Jap., vol. 60, No 3, 1991, 781-784. [36] Awrejcewicz J., Someya T., Analytical condition for the existence of twoparameter family of quasiperiodic orbits in the autonomous system (Nonresonance case). J. Phys. Soc. Jap., vol. 61, No 7, 1992, 2231-2234.
General References
293
[37] Awrejcewicz J., Someya T., Periodic oscillations and two-parameter unfoldings in nonlinear discrete--continuous systems with delay. J. Sound Vibr., vol. 158, No 1, 1992. [38] Awrejcewicz J., Someya T., Analytical conditions for the existence of a twoparameter family of periodic orbits in nonautonomous dynamical systems. Nonlinear Dynamics, vol. 4, 1993, 39-50. J., Someya T., On introducing inertial forces into non-linear Awrejcewicz [39] analysis of spatial structures. J. Sound Vibr., vol. 103, No 3, 1993, 545-548. [40] Aziz A., Lunardini V.J., Perturbation techniques in phase change heat transfer. Appl. Meehs Rev., vol. 46, No 2, 1993, 29-68. [41] Babcock C.D., Chen J.C., Nolinear vibration of cylindrical shell. AIAA J., vol. 13, No 7, 1975, 868-876. [42] Baker G.A., Essential of Pade approximants. Academic Press, New York, 1975. [43] Baker G.A., Graves-Morris P., Pade approximants. Addison-Wesley Pub!. Co., New York, 1981. [44] Bakhalov N., Panasenko G., Averaging processes in periodic media. Mathematical problems in mechanics of composite materials. Kluwer Academic Publishers, Dordrecht, 1989. [45] Barantsev R.G., Asymptotic versus classical mathematics. Topics in Mathematical Analysis, 1989, 49-64. [46] Barenblatt G.I., Similarity, self-similarity and intermediate asymptotics. Plenum, New York, London, 1979. [47] Barenblatt G.!., Dimensional analysis. Gordon and Breach, New York, London, 1987. [48] Barenblatt G.!., Intermediate asymptotic, scaling laws and renormalization group in continuum mechanics. Mecca nica, vol. 28, 1993, 177-183. [49] Bender C.M., Milton K.A., Pinsky S.S., Simmon L.M.Jr., A new perturbative approach to nonlinear problems. J. Math. Phys., vol. 30, No 7, 1989, 14471455. [50] Bensoussan A., Lions J.-L., Papanicolaou G., Asymptotic method in periodic structures. North-Holland Publ. Co., New York, 1978. [51] Berger H.M., A new approach to the analysis oflarge deflections of plates. J. Appl. Mechs, vol. 55, No 4, 1955, 465-472. [52] Birkhoff G., Numerical fluid dynamics. SIAM Review, vol. 25, No I, 1983, 1-24. [53] Bodenschatz E., Pesch W., Kramer L., Structure and dynamics of dislocations in anisotropic patterns-forming systems. Physica D, vol. 32, 1988, 135-145. [54] Boettcher S., Bender C.M., Nonperturbative square--well approximation to a quantum theory. J. Math. Phys., vol. 31, No 11, 1990,2579-2585. [55] Bogoliubov N.N., Mitropolsky Yu.A., Asymptotic method in the theory of nonlinear oscillations. Gordon and Breach, London, 1985. [56] Bolotin V.V., An asymptotic method for the study of the problem of eigenvalues of rectangular regions. Problem of Continuum Mechs. SIAM, Philadelphia, 1961, 56-58. [57] Bruning J., Seeley R., Regular singular asymptotics. Advances in Maths, vol. 58, 1985, 133-148. [58] Bucco D., Jones R., Masumdar J., The dynamic analysis of shallow spherical shells. TI-ans. ASME, J. Appl. Mechs, vol. 45 No 3, 1978, 690-691. [59] Bush A.W., Perturbation methods for engineers and scientists. CRC Press, Boca Raton, 1992. [60] Caillerie D., Thin elastic and periodic plates. Math. Methods in App. Sc., vol. 6, 1984, 151-191.
294
General References
[61] Chalbaud E., Martin P., Two-point quasifractional approximant in physics: Method improvement and application to In(x). J. Math. Phys., vol. 33, No 7, 1992, 2483-2486. [62] Chen G., Goleman M.P., Zhon J., The eigenvalence between the wave propagation method and Bolotin's method in the asymptotic estimation of eigenfrequencies of a rectangular plate. Wave Motion, vol. 16, No 3, 1992, 285-297. [63] Christensen R.M., Mechanics of composite materials. John Wiley & Sons, New York, 1979. [64] Ciarlet P.G., Plates and junctions in elastic multi-structures. Masson, Paris, 1990. [65] Ciarlet P.G., Mathematical shell theory. A linearly elastic introduction. Computional Mechs Advances, North-Holland, Amsterdam, 1994. Cole J.D., Perturbation methods in applied mathematics. Ginn (Blaisdell), [66] Boston, 1968. [67] Datta S., Large amplitude free vibration of irregular plates placed on elastic foundations. Int. J. Nonlin. Mecks, vol. 11, 1976, 337-345. [68] Destuynder P., A classification of thin shell theories. Acta Appl. Math., vol. 4, 1985, 15-63. [69] Donnell L.H., Beams, plates and shells. McGraw-Hill, New York, 1976. [70] Dowell E.H., On the nonlinear fLectural vibration of rings. AIAA J., No 5, 1967, 1508-1509. Dowell E.H., Ventres C.S., Modal equations for nonlinear fLectural vibration [71] of a cylindrical shell. Int. J. Solid Str., No 4, 1968, 975-991. [72] Draux A., On two-point Pade-type and two-point Pade approximants. Ann. Mat. Pures et Appl., No 158, 1991, 99-124. [73] Elishakoff I., Bolotin's dynamic edge effect method. The Shock and Vibr. Digest, vol. 8, No 1, 1976, 95-104. [74] Elishakoff I., Sternberg A., Van Baten T.J., Vibrations of multispan all-round clamped stiffened plates by modified dynamic edge effect method. Compo Methods in Appl. Mechs and Engn., vot. 105, 1993, 211-223. [75] Erofeyev V.I., Potapov A.I., Longitudinal strain waves in nonlinearity-elastic media with couple stresses. Int. J. Nonlinear Mecks, vol. 28, No 4, 1993, 483-488. [76] Estrada R.P., Kanwal R.P., Asymptotic analysis: a distributional approach. Birkauser, Boston, Basel, Berlin, 1994. [77] Evensen D.A., Some observations on the nonlinear vibration of thin cylindrical shell. AIAA J., vol. 1, 1~3, 2857-2858. [78] Evensen D.A., A theoretical and experimental study of nonlinear fLectural vibration of thin circular ring. 1hms. ASME, J. Appl. Mechs, vol. 4, No 3, 1966, 553--560. [79] Fomenko A.T., Mathematical impressions. AMS, Providence, 1990. [80] Fredrichs K.O., Asymptotic phenomena in mathematical physics. Bull. Amer. Math. Soc., vol. 61, 1955, 485-504. [81] Frost P.A., Harper E.Y., Extended Parle procedure for constructing global approximations from asymptotic expansion: an explication with examples. SIAM Rev., vol. 18, 1976, 62-91. [82] Gol'denveizer A.L., Theory of elastic thin shells. Pergamon Press, New York, Oxford, London, Paris, 1961. [83] Gol'denveizer A.L., Lidsky V.B., Tovstik P.E., Free vibmtion of thin elastic shells. Nauka, Moscow, 1979 (in Russian). [84] Goodier J.N., McIvor I.K., The elastic cylindrical shell under nearly uniform radial impulse. Trans. ASME, J. Appl. Mecks, vol. 31, No 2, 1964, 259-266. [85] Grundy R.E., Laplace transform inversion using two-point rational approximants. J. Inst. Matks Applics., vol. 20, 1977, 299-306.
General References
295
[86] Grundy R.E., The solution of Volterra integral equation of the convolution type using two-point rational approximants. J. Inst. Matks Applics., vol. 22, 1978, 147-158. [87] Grundy R.E., On the solution of nonlinear Volterra integral equations using two-point Pacle approximants. J. Inst. Maths Applics., vol. 22, 1978,317-320. [88] Guckenheimer J., Multiple bifurcation problems of codimension two. SIAM J. of Math. Anal., 15, 1, 1981, 1-49. [89] Guckenheimer J., Holmes P.J., Nonlinear oscillations, dynamical systems and bifurcation of vector fields. Springer, Berlin, Heidelberg, 1983. [90] Han S., On the free vibration of a beam on a nonlinear elastic foundation. Trans. ASME, J. Appl. Mecks, vol. 32, No 2, 1965, 445-447. Hassard B.D., Kazarinoff N.D., Wan Y.H., Theory and applications of the [91] Hopf bifurcation. Cambridge University Press, Cambridge, 1980. [92] Heutemaker M.S., Frankel P.N., Gollub J.P., Convection patterns: time evolution of the wave--vector field. Physical Review Letters, vol. 54, No 13, 1985, 1369-1372. [93] Handbook of special functions with formulas gmpks and mathematical tables. Abramowitz M., Stegun LA. (eds.). Appl. Matks Ser., vol. 55, Dover, National Bureau of Standards, 1964. [94] Hinch E.J., Perturbation methods. Cambridge University Press, Cambridge, 1991. [95] Holms M.H., Introduction to perturbation methods. Springer, Berlin, Heidelberg, 1995. [96] Humphreys D.S., Bodner S.R., Dynamic buckling of shallow shells under impulsive loading. ACSE J. Engn. Mech. Div., vol. 88, No 2, 1962, 17-36. [97] Humphreys D.S., Roth R.S., Zalters J., Experiments on dynamic buckling of shallow spherical shell under shock loading. AIAA J., vol. 3, No 1, 1965, 33-39. [98] Iooss G., Joseph D.D., Elementary stability and bifurcation theory. Springer, Berlin, Heidelberg, 1980. [99] Jones W.B., Thron W.J., Analytic theory and applications. Addison-Wesley Publ. Co., London, 1980. [100] Kaas-Petersen C., Continuation methods as the link between perturbation analysis and asymptotic analysis. SIAM Rev., vol. 29, No I, 1987, 115-120. [101] Kalamkarov A.L., Composite and reinforced elements of construction. Wiley, Chichester, New York, 1992. [102] Kamolphan D.-U., Mei C., Finite element method for forced nonlinear vibrations of circular plates. Int. J. Math. Engn., vol. 29, No 9, 1986, 1715-1726. [103] Kovrigin D.A., Potapov A.I., Nonlinear resonance interactions of longitudinal and flexural waves in a ring. Soviet Physics Doklady, voL 34, No 4, 1989,330332. [104] Kovrigin D.A., Potapov A.I., Nonlinear vibration of a thin ring. Soviet Appl. Mechs, vol. 25, No 3, 1989, 281-286. [105] Kruskal M.D., Asymptotology: Mathematical Models in Physical Science. Prentice--Hall, Englewood Cliffs, N.J., 1963, 17-48. [106] Ladygina Ye.V., Manevitch A.I., Free oscillations of a non-linear cubic system with two degrees of freedom and close natural frequencies. PMM, J. Appl. Maths Mechs, vol. 57, No 2, 257-266. [107] Lau S.L., Cheung Y.K., Chen Chuhui, An alternative perturbation procedure of multiple scales for nonlinear dynamics systems. Trans. ASME, J. Appl. Mechs, vol. 56, No 3, 1989, 587-605. [108] Lighthill M.J., A technique for rendering approximate solutions to physical problems uniformly valid. Phil. Mag., voL 40, 1179-1201.
296
General References
[109] Lin S.S., Segel L.A., Mathematical methods applied to deterministic problems in the natural sciences. SIAM, Philadelphia, 1988. [110] Lindberg H.E., Stress amplification in a ring caused by dynamic instability. nuns. ASME, J. Appl. Mechs, vol. 41, No 2, 1974,392-400. [111] Litvinov G.L., Approximate construction of rational approximants and effect of self-correction of error. Maths and Modeling, Puschino, 1990, 99--141 (in Russian). [112] Lock M.H., Okubo S., Whittier J.S., Experiments on the snapping of a shallow dome under a step pressure load. AIAA J., vol. 6, No 7, 1968, 1320-1326. [113] Luke Y.L., Computations of coefficients in the polynominals of Pade approximants by solving systems of linear equations. J. Compo Appl. Maths, vol. 6, No 3, 1980, 213-218. [114] McCabe J.H., Murphy J.A., Continued fractions which correspond to power series expansions at two points. J. Inst. Maths Applics., vol. 17, 1976, 233-247. [115] Manevitch L.L, Pavlenko A.V., Koblik S.G., Asymptotic methods in the theory of elasticity of orthotropic body. Visha Shkola, Kiev-Donezk, 1979 (in Russian). [116] Marsden J.E., McCraken M., The Hopfbifurcation and its applications. Appl. Mathem. Sci. 19, Springer, Berlin, Heidelberg, 1976. [117] Martin P., Baker G.A. Jr., Two-point quasifractional approximant in physics. Truncation error. J. Math. Phys., vol. 32, No 6, 1991, 1470-1477. [118] Mikhlin Yu.V., Matching of local expansion in the theory of nonlinear vibrations, J. Sound Vibr., vol. 182, No 4, 1995, 577·588. [119] Nayfeh A.H., Perturbation methods. John Wiley & Sons, New York, 1973. [120] Nayfeh A.H., Introduction to perturbation techniques. John Wiley & Sons, New York, 1981. [121] Nayfeh A.H., Problems in perturbations. John Wiley & Sons, New York, 1985. [122] Nayfeh A.H., Mook D.T., Nonlinear oscillations. John Wiley & Sons, New York, 1979. [123] Nemeth G., Paris G., The Gibbs phen
General References
297
[134] Rogers R.J., Pick R.J., Factors associated with support plate forced due to heat-exchanger tube vibratory contact. Nucl. Engn. Design, No 44, 1977. [135] Rosenberg RM., On nonlinear vibrations of systems with many degrees of freedom. Advances Appl. Mechs, vol. 9, 1966, 156-243. [136] Ross E.W. (Jr .), Asymptotic analysis of the axisymmetric vibrations of shells. 1hms ASME, J. Appl. Mechs, vol. 33, No 1, 1966, 85-92. [137] Rubanik W.O., Vibration in complex quasi-linear systems with delay. University Press, Minsk, 1985 (in Russian). [138] Rutten H.S., Theory and design of shells on the basis of asymptotic analysis. Rutten & Kruisman, Consulting Engns, Rijswijk, 1973. [139] Sanchez-Palencia E., Non-homogeneous media and vibration theory. SpringerVerlag, Berlin, Heidelberg, 1980. [140] Sanders J.-L., Nonlinear theory for thin shell. Quart. Appl. Meehs, vol. 21, 1963, 21-36. [141] Sanders J.A., Verhulst F., Averaging methods in nonlinear dynamical systems. Springer, Berlin, Heidelberg, 1985. [142] Segel L.A., The importance of asymptotic analysis in applied mathematics. Amer. Math. Monthly, vol. 73, No 1, 1966, 7-14. [143] Segel L.A., Distant side-walls cause slow amplitude modulation of cellular convection. J. of Fluid Mechanics, vol. 38, No 1, 1969, 203-224. [144] Shanks D., Nonlinear transforms of divergent and slowly convergent sequences. J. Maths £3 Phys., vol. 34, 1955, 1-42. [145] Shinbrot T., Grebogi C., Ott E., Yorke J.A., Using small perturbations to control chaos. Nature, vol. 363 (June), 1993, 411-417. [146] Slotine J.-J.E., Li W., Applied nonlinear control. Prentice--Hall, Englewood Cliffs, New Jersey, 1991. [147] Smith D.A., Ford W.F., Acceleration of linear and logarithmic convergence. SIAM J. Numer. Anal., vol. 16, No 2, 1979, 223-240. [148] Srinivasan R.S., Thiruvenkatachar V., Free vibration of annual sector plates by an integral equation technique. J. Sound Vibr., vol. 89, No 3, 1983, 425432. [149] Strickling J.A., Martinez J.E., Tillerson J.R., et al., Nonlinear dynamic analysis of shells of revolution by matrix displacement method. AIAA J., vol. 9, No 4, 1971, 629-636. [150] Surkin R.G., Zuev B.M., Stepanov S.G., Experimental investigation of dynamical stability of spherical segments. Soviet Appl. Meeks, vol. 3, No 8, 1967, 124-132. [151] Szubshcik L.S., Stolyar A.M., Tsybulin B.G., Asymptotic integration of nonlinear equations of cylindrical panel vibrations. PMM, J. Appl. Maths Mechs, vol. 52, No 1, 1988, 511-518. [152] Tables of integral transformation, vol. 1. McGraw-Hill, New York, 1954. [153] Toda M., Theory of nonlinear lattices. Springer, Berlin, Heidelberg, 1981. [154] Van Dyke M., Perturbation methods in fluid mechanics. The Parabolic Press, Stanford, California, 1975. [155] Vasil'eva A.B., Butuzov V.F., Kalachev L.V., The boundary function method for singular perturbation problems. SIAM, Philadelphia, 1995. [156] Vedenova E.G., Manevitch LJ., Nisichenko V.P., Lysenko S.A., Solitons with a substantially nonlinear one--dimensional chain. J. Appl. Mechs Technical Physics, vol. 25, No 6, 1984,910-914. [157] Vol'mir A.S., Nonlinear plates and shells dynamics. Nauka, Moscow, 1972 (in Russian). [158] Volos N.P., Korol' LYou., Nonlinear vibrations of rectangular plates with various boundary constraints. Soviet Appl. Mechs, vol. 12, No 7, 1976, 101106.
298
General References
[159] Wah T., The normal modes of vibration of certain nonlinear continuous systems. 1Tans ASME, J. Appl. Meeks, vol. 31, No 1, 1964. [160] Whitham G., Linear and non-linear waves. John Wiley & Sons, New York, 1974. [161] Wong R., Distributional derivation of an asymptotic expansion. Proc. Amer. Math. Soc., vol. 80, No 2, 1980, 266-270. [162] Zarutsky V.A., Oscillations of ribbed shells. Intern. Appl. J. Meeks., vol. 29, No 10, 1993, 837-841. [163] Zeitounian R., Asymptotic modeling of atmospheric flows. Springer, Berlin, Heidelberg, 1990. [164] Zemanian A.H., Distribu.tion theory an.d tmnsform analysis. McGraw-Hill, New York, 1965. [165] Zhuravlev V.F., Klimov D.M., Applied methods in the theory of oscillations. Nauka, Moscow, 1988 (in Russian).
Detailed References (d)
[1] Amiro LYa., Zarutsky V.A., Studies of the dynamics of ribbed shells. Soviet. Appl. Mech3, vol. 17, No 11, 1981, 949--962. [2] Andrianov LV., Asymptotic solutions for nonlinear systems with high degrees of nonlinearity. PMM, J. Appl. Maths Mechs, vol. 57, No 5, 1993, 941-943. [3] Andrianov LV., A new asymptotic method of integrating the equations of quantum mechanics for strong coupling. Physics Doklady, vol. 38, No 2, 1993, 56-57. [4] Andrianov LV., Two-point Parle approximants in the mechanics of solids. ZAMM, vol. 74, No 4, 1994, 121·122. [5] Andrianov I.V., Danishevskyi V.V., Asymptotic investigation of the nonlinear dynamic boundary value problem for rod. Technische Mechanik, vol. 15, No 1, 1995, 53-55. [6] Andrianov LV., Kholod E.G., Exact solution in the nonlinear theory of reinforced with discrete stiffness plates. Facta universitatis, University of Nis, vol. I, No 4, 1994, 389--400. [7] Andrianov LV., Kholod E.G., Chernetsky V.A., Asymptotic-based method for a plane elasticity mixed boundary eigenvalue problem. J. Theor. Appl. Mechs, vol. 3, No 32, 1994, 701-709. [8] Andrianov LV., Lesnichaya V.A., Manevitch LJ., Homogenization methods in statics and dynamics of reinforced shells. Nauka, Moscow, 1985 (in Russian). [9] Andrianov LV., Manevitch L.I., Homogenization method in the theory of shells. Advances in Mechs, vol. 6, No 3/4, 1983, 3-29 (in Russian). [10] Andrianov LV., Manevitch L.L, Asymptotology: ideas, methods, results. AsIan, Moscow, 1994 (in Russian). [11] Andrianov I.V., Sedin V.L., Composition of simplified equations of nonlinear dynamics of plates and shells on the basis of homogenization method. ZAMM, vol. 67, No 7, 1988, 573-575. [12] Andrianov LV., Shevchenko V.V., Perforated plates and shells. Asymptotic Methods in the System Theory, Irkutsk, 1989, 217-243 (in Russian). [13] Andrianov I.V., Starushenko G.A., Application of the averaging method for the calculation of perforated plates. Soviet Appl. Meehs, vol. 24, No 4, 1988, 410-415. ' [14] Awrejcewicz J., Vibration system: rotor with self-excited support. Proceedings of Int. Conf. on Rotor Dynamics, Tokyo, Sept. 14-17, 1986, 517-522. [15] Awrejcewicz J., Hopf bifurcations in Duffing oscillator. PAN American Congress of Appl. Mech., Rio de Janeiro, 1989, 640-643. [16] Awrejcewicz J., The analytical method to detect Hopf bifurcation solutions in the unstationary nonlinear systems. J. Sound Vibr., vol. 129, No I, 1989, 175-178.
300
Detailed References (d)
[17] Awrejcewicz J., Parametric and self-excited vibrations included by friction in a system with three-degree-of-freedom. KSME Journal, vol. 4, No.2, 1990, 156-166. [18] Awrejcewicz J., Analysis of double Hopf bifurcations. Nonlinear Vibration Problems, vol. 24, 1991, 123-140. J., Analysis of the biparameter Hopf bifurcation. Nonlinear ViAwrejcewicz [19] bration Problems, vol. 24, 1991, 63-76. [20] Awrejcewicz J., On the Hopf bifurcation. Nonlinear Vibration Problems, vol. 24, 1991, 15-31. [21) Awrejcewicz J., Nonlinear oscillations of the string caused by the electromagnetic field. In: Proceedings of the Third Polish-Japanese Seminar on Modelling and Control of Electromagnetic Phenomena, Kazimierz, Poland, April 19-21, 1993, 125-128. [22] Awrejcewicz J., Analytical and numerical study of nonlinear behaviour of the electromechanical systems. J. Tech. Phys., vol. 34, No 1, 1993, 57-68. [23] Awrejcewicz J., Nonlinear dynamics of machines. Technical University Press, L6dz, 1994 (in Polish). [24] Awrejcewicz J., Strange nonlinear behaviour of the electromechanical system. Machine Dynamics Problems, vol. 7, 1994,7-10. [25] Awrejcewicz J., Strange nonlinear dynamical behaviour of a string type generator with a time delay amplifier. Proc. Ninth World Congress on the TMM, Milan, Italy, August 30-31/Sept 1-2, 1995,913-916. [26] Awrejcewicz J., Vibration of discrete deterministic systems. WNT Publishers in Science and Technology, Warsaw, 1996 (in Polish). [27] Awrejcewicz J., Tomczak K., Lamarque C.-H., Controlling system with impact. Proceed. of the Int. Conf. on Nonlinearity, Bifurcation and Chaos, L6dzDobieszk6w, Sept. 16-18, 1996,73-76. [28] Babich V.M., Buldirev V.S., The art of asymptotic. Vestnik Leningrad Univ. Maths, vol. 10, 1982, 227-235. [29] Bauer S.M., Fillipov S.B., Smirnov A.Il., Tovstik P.E., Asymptotic methods in mechanics with applications to thin plates and shells. Asymptotic methods in Mechanics. CRM Proc. and Lect. Notes, vol. 3, 1994,3-140. [30] Belov M.A., Tyrulis T.T., Asymptotic methods of integral transforms inverse. Zinatne, Riga, 1985 (in Russian). [31] Bender C.M., Milton K.A., Boettcher S., A new perturbative approach to nonlinear partial differential equations. J. Math. Phys., vol. 32, Noll, 1991, 3031-3038. , [32] Bogdanovich A.E., Zarutsky V.A., Oscillations of ribbed shells. Soviet Appl. Mechs, vol. 27, No 10, 1991, 1001-1006. [33] Elsgolc L.E., Introduction to the theory of differential equations with a delay. Nauka, Moscow, 1964 (in Russian). [34] Estrada R.P., Kanwal R.P., A distributional theory for asymptotic expansions. Proc. R. Soc. Lond. A, vol. 428, 1990, 399-430. [35] Estrada R.P., Kanwal R.P., Taylor expansions for distribution. Math. Methods in Appl. Sc., vol. 16, 1993, 297-304. [36] Evkin A.Yu., A new approach to the asymptotic integration of the equations of shallow convex shells in the post-critical stage. PMM, J. of Appl. Maths Mechs, vol. 53, No 1, 1986, 92-96. . [37] Filatov A.N., Sharova LV., Integral inequalities and theory of nonlinear oscillations. Nauka, Moscow, 1976 (in Russian). [38] Fok V.A., Problems of difmction and spreading of electromagnetic waves. Nauka, Moscow, 1970 (in Russian). [39] Gonchar A.A., About uniform convergence of the diagonal Pade approximants. Matemat. Sbornik, vol. 118, No 4, 1982, 535-556.
Detailed References (d)
301
[40] Grigoluk E.I., Phylshtinsky L.A., Perforated plates and shells. Nauka, Moscow, 1970 (in Russian). [41] Grybos R., Theory of impact in discrete mechanical systems. PWN, Warsaw, 1969 (in Polish). [42] Hamada M., Ota T., FUndamental frequencies of simply supported but partially clamped square plates. Bull. Japan Soc. Mech. Eng., vol. 6, No 23, 1963, 397-403. [43] Hara S., Yamamoto Y., Ornata T., Nokano M., Repetitive control system: a new type servo system for periodic exogenous signals. IEEE 7Tansactions on Automatic Control, vol. 33, No.7, 1988, 659-668. Huang N.C., Axisymmetric dynamic snap-through of elastic clamped shallow [44] spherical shells. 'Thms. ASME, J. Appl. Meehs, vol. 7, No 2, 1969, 215-220. [45] Huseyin K., Atadan A.S., On the analysis of Hopf bifurcation. Int J. Engng. Sci., vol. 21, No 3, 1983, 247-262. [46] Huseyin K., Atadan A.S., On generalized Hopf bifurcations. J. Dyn. Sys., Measur. and Contr., vol. 106, 1984, 327-334. [47] Infeld E., Rowlands G., Lenkowska-Czerwinska T., On introducing dynamics into the theory oj nucleation of coherent structures (to be published). [48] Kantorovitch L.V., Krylov V.I., Approximate methods of higher analysis. Nauka, Moscow, 1949 (in Russian). [49] Karmishin A.V., Zhukov A.I., Kolosov V.G., et al., Methods of dynamic calculation and testing for thin-walled structures. Mashinostroyenie, Moscow, 1990 (in Russian). Kayuk Ya.F., Some problems of the methods splitting by parameters. Naukova [50] Dumka, Kiev, 1991 (in Russian). [51] Koiter W.T., On the nonlinear theory of thin elastic shells. Proc. Kon. Ned. Ak. Wet., ser B, vol. 69, No I, 1966, 1-54. [52] Krodkiewski J.M., Faragher J.-S., Stability of the periodic motion of nonlinear systems. Proceedings of the International Conference on Vibration and Noise, Venice, 1995. [53] KryloffV.I., Skoblya N.S., Methods of approximate Fourier and Laplace transform inverse. Nauka, Moscow, 1974 (in Russian). [54] Lyapunov A.M., The general problem of the stability of motion. Gostehizdat, Moscow, 1950. [55] Manevitch A.I., Stability and optimal design of reinforced shells. Visha Shkola, Kiev-Donetzk, 1972 (in Russian). [56] Manevitch L.I., Mikhlin Yu.V., Pilipchuk V.N., The method of normal oscillations for essentially nonlinear systems. Nauka, Moscow, 1989 (in Russian). [57] Marchuk G.I., Agoshkov V.I., Conjugate operators and algorithms of perturbation in non-linear problems: 1. Principles of construction of conjugate operators. Soviet J. Num. Anal. and Math. Modeling, No I, 1988, 21-46. [58] Marchuk G.I., Agoshkov V.I., Conjugate operators and algorithms of perturbation in non-linear problems: 2. Perturbation algorithms. Soviet J. Num. Anal. and Math. Modeling, No 2, 1988, 115-136. [59] Mitropol'sky Yu.A., Homa G.P., Gromak M.I., Asymptotic methods of investigation of quasiwave equations of hyperbolic type. Naukova Dumka, Kiev, 1991 (in Russian). [60] Nayfeh A.H., Numerical-perturbation methods in mechanics. Comput. and Struct., vol. 30, No 1-2, 1988, 185-204. [61] Newell A.C., Whitehead J.A., Finite bandwidth, finite amplitude convection. J. of Fluid Mechanics, vol. 38, No 2, 1969, 279-303. [62] Obraztsov I.F., Nerubaylo B.V., Andrianov LV., Asymptotic methods in the theory of thin-walled structures. Mashinostroyenie, Moscow, 1991 (in Russian).
302
Detailed References (d)
[63] Peterka F., Laws of impact motion of mechanical systems with one degree of freedom. Part I and II. Acta Technica CSAV, No.5, 1974, 462-473, 569-580. [64] Potier-Ferry M., Bensaadi M.H., Computation of periodic solutions by using Pade approximants. 1st European Nonl. Osc. Conf. Abstracts, 1993, 14. [65] Proskuriakov A.P., Poincare method in the theory of nonlinear oscillations. Nauka, Moscow, 1977 (in Russian). Pyragas K., Continuous control of chaos by self-controlling feedback. Physics [66] Letters, A vol. 170, Iss. 6, 1992,421-428. [67] Semerdjiev Kh., Trigonometric Pade apP'roximants and Gibbs phenomenon. Rep. United & Inst. Nuder Res., NP5 - 12484, Dubna, 1919, 10 (in Russian). [68] Slepyan A.I., Yakovlev U.S., Integral transforms in the nonstationary mechanical problems. Shipbuilding Publ., Leningrad, 1980 (in Russian). [69] Therapos C.P., Diamessis J.E., Approximate Pade approximants with applications to rational approximation and linear-order reduction. Proc. of the IEEE, vol. 72, No 12, 1984, 1811-1813. [70] Tzeitlin A.I., Glikman B.G., Elastic rectangular plate bending under mixed boundary conditions. Proceed. Central Res. Inst. Civil Engn. Str. Res. Str. Dynamics, vol. 17, Moscow, 1971, 123-142 (in Russian). [71] Youncef-Toumi K., Wu S.-T., Input/output linearization using time delay control. Journal of Dynamics Systems, Measurement and Control, vol. 114 (Mar), 1992, 10-19.
Index
Airy stress function 211 Amplitude 12,13,17-19,26,27,30,
31,36-38,41,43,44,50,82,85,86,96, 100,116-118, 125 t 128, 135,136,142, 144, 145, 147, 152,. 161, 177, 185, 186, 197,213,219,220,.227-229,236,242, 245,246,248,259-263,266,269,272, 280-282,284-289
Averaging
1,5-7,155,165,166,280,
282 -
method 6,7,176,280 operator 164, 169, 177 ordinary 7 7,171,174,177,228, procedure
280,284 - technique
280
Amplitude-frequency dependence
129,185,211,213,214,230 Amplitude-phase portrait 110 Antiphase 111 Approach - asymptotic 1,3,5,8,9,11,12,14,
118,129,132,139,228 - asymptotic-numerical 12 - quasi-linear 136 - weak-coupling 120 Approximant 136,146, 149 - quasifactional 148 Approximation 1,3,5,9, 11, 16,24,
25,27,63,67,75,80,81,95,101,109, 119,121,129,131,134,136-138,140, 142,146,150-155,171,172,176,186, 192,193,195,196,220-222,228,240, 280,285 - asymptotic 3,4 - continual 151 - equivalent linear 25 - higher-order 11, 132 197 - high-frequency continuum 154 - linear 14 - long-frequency 154 - long-wave 236,242,244,246 - short-wave 237 - short-wavelength continuum 154 - successive 19, 95, 137 Asymptotic reduction 4, 13 Asymptotology Attractor 101
1
Ball bearing assembly 101 Be8lll 8,151,152,236-238,240,261, 262 Berger 171, 172 - equation 158,171-176 - hypothesis 171, 172, 175, 176 - plate 175 Bessel - equation 187, 189 - function 132, 164, 186-188 Bifurcation 93-95, 100,260 - par8llleter 94, 95, 97 Body 8,11,48,57 - deformable rigid 1 - elastic 8 236 - rigid - stre8llllined 1, 8 Bolotin 177,185 - approach - method 178,184 Boundary 112, 166, 172, 183, 190,
216-219,221,233,234 203 1,2,8,11,129,159,202-204 207,209-211,221,224 Bubnov-Galerkin method 170,171, 189
- effect - layer
Case - isotropic 197,205,225 - limit 231,234,236,238 - limiting 1,2,4,5,146,184