DIFFERENTIAL AND INTEGRAL INEQUALITIES Theory and Applications Volume I1
FUNCTIONAL, PARTIAL, ABSTRACT, AND COMPLEX DIFFERENTIAL EQUATIONS
This is Volume 55 in MATHEMATICS I N SCIENCE AND ENGINEERING A series of monographs and textbooks Edited by RICHARD BELLMAN, University of Southern California A complete list of the books in this series appears at the end of this volume.
DIFFERENTIAL AND
INTEGRAL INEQUALITIES Theory and Applications Volume I1
FUNCTIONAL, PARTIAL, ABSTRACT, AND COMPLEX DIFFERENTIAL EQUATIONS V. LAKSHMIKANTHAM and S. LEELA University of Rhode Island Kingston, Rhode Island
A CAD E MI C P R E SS
New York and London
1969
COPYRIGHT 0 1969,
BY
ACADEMIC PRESS,INC.
ALL RIGHTS RESERVED. NO PART OF THIS BOOK MAY BE REPRODUCED I N ANY FORM, BY PHOTOSTAT, MICROFILM, RETRIEVAL SYSTEM, OR ANY OTHER MEANS, WITHOUT WRITTEN PERMISSION FROM THE PUBLISHERS.
ACADEMIC PRESS, INC. 111 Fifth Avenue, New York, New York 10003
United Kingdom Edition published by ACADEMIC PRESS, INC. (LONDON) LTD. Berkeley Square House, London W l X 6BA
LIBRARY OF CONGRESS CATALOG CARDNUMBER: 68-8425 AMS 1968 SUBJECT CLASSIFICATIONS 3401, 3501
PRINTED I N THE UNITED STATES OF AMERICA
Preface
T h e first volume of Differential and Integral Inequalities: Theory and Applications published in 1969 deals with ordinary differential equations and Volterra integral equations. It consists of five chapters and includes a systematic and fairly elaborate development of the theory and application of differential and integral inequalities. T h i s second volume is a continuation of the trend and is devoted to differential equations with delay or functional differential equations, partial differential equations of first order, parabolic and hyperbolic types respectively, differential equations in abstract spaces including nonlinear evolution equations and complex differential equations. T o cut down the length of the volume many parallel results are omitted as exercises. We extend our appreciation to Mrs. Rosalind Shumate, Mrs. June Chandronet, and Miss Sally Taylor for their excellent typing of the entire manuscript. V. LAKSHMIKANTHAM S. LEELA Kingston, Rhode Island August, 1969
V
This page intentionally left blank
Contents
V
PREFACE
FUNCTIONAL DIFFERENTIAL EQUATIONS Introduction Existence Approximate Solutions and Uniqueness Upper Bounds Dependence on Initial Values and Parameters Stability Criteria Asymptotic Behavior A Topological Principle Systems with Repulsive Forces Functional Differential Inequalities Notes
3 4 9 13 18 21 24 29 32 34 42
7.0. 7.1. 7.2. 7.3. 7.4. 7.5. 7.6. 7.7.
Introduction Stability Criteria Converse Theoreins Autonomous Systems Perturbed Systems Extreme Stability Almost Periodic Systems Notes
43 43 49 58 62 66 72 80
8.0. 8.1. 8.2. 8.3. 8.4. 8.5. 8.6. 8.7.
Introduction Basic Comparison Theorems Stability Criteria Perturbed Systems An Estimate of Time Lag Eventual Stability Asymptotic Behavior Notes
81 81 87 97 100 101 105 110
Chapter 6.
6.0. 6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 6.7. 6.8. 6.9. 6.10.
Chapter 7.
Chapter 8.
vii
...
CONTENTS
Vlll
PARTIAL DIFFERENTIAL EQUATIONS Chapter 9.
9.0. 9. I . 9.2. 9.3. 9.4. 9.5. 9.6. 9.7.
Introduction Partial Differential Inequalities of First Order Comparison Theorems Upper Bounds Approximate Solutions and Uniqueness Systems of Partial Differential Inequalities of First Order Lyapunov-Like Function Notes
Chapter 10. 10.0. lntroduction
10. I . Parabolic Differential Inequaliies in Bounded Domains 10.2. Comparison Theorems 10.3. Bounds, Under and Over Functions 10.4. Approximate Solutions and Uniqueness 10.5. Stability of Steady-State Solutions 10.6. Systems of Parabolic Diffcrential inequalities in Bounded Domains 10.7. Lyapunov-Like Functions 10.8. Stahility and Boundedness 10.9. Conditional Stahility and Boundedness 10.10. Parabolic Differential Inequalities in Unbounded Domains 10.11. Uniqueness 10.12. Exterior Boundary-Value Problem and Uniqueness 10.13. Notes
Chapter 11. I 1.0. Introduction 1 1.1, 1 1.2. 11.3. 11.4.
Hyperbolic Diflerential Inequalities Uniqueness Criteria Upper Bounds and Error Estimates Notes
113 113 118 127 134 136 144 148 149 149 155 163 170 174 181 186 190 200 205 210 21 3 219
22 1 22 1 223 229 233
DIFFERENTIAL EQUATIONS IN ABSTRACT SPACES Chapter 12. 12.0. Introduction 12. I . 12.2. 12.3. 12.4. 12.5. 12.6. 12.7. 12.8. 12.9. 12.10.
Existence Norilocal Existence Uniqueness Continuous Dependence and the Method of Averaging Existence (continued) Approximate Solutions and Uniqueness Chaplygin’s Method Asymptotic Behavior Lyapunov Function and Comparison Theorems Stability and Boundedness 12.11. Notes
237 231 24 1 243 247 249 254 258 264 267 269 272
CONTENTS
ix
COMPLEX DIFFERENTIAL EQUATIONS Chapter 13. 13.0. Introduction 13.1. 13.2. 13.3. 13.4. 13.5.
Existence, Approximate Solutions, and Uniqueness Singularity-Free Regions and Growth Estimates Componentwise Bounds Lyapunov-like Functions and Comparison Theorems Notes
215 215 219 284 286 288
Bibliography
289
Author Index
315
Subject Index
318
This page intentionally left blank
DIFFERENTIAL AND INTEGRAL INEQUALITIES Theory and Applications Volume I1
FUNCTIONAL, PARTIAL, ABSTRACT, AND COMPLEX DIFFERENTIAL EQUATIONS
This page intentionally left blank
FUNCTIONAL DIFFERENTIAL EQUATIONS
This page intentionally left blank
Chapter 6
6.0. Introduction
T h e future state of a physical system depends, in many circumstances, not only on the present state but also on its past history. Functional differential equations provide a mathematical model for such physical systems in which the rate of change of the system may depend on the influence of its hereditary effects. T h e simplest type of such a system is a differential-difference equation x’@) = .f(G x ( t ) , x(t
- T)),
where T > 0 is a constant. Obviously, for T = 0, this reduces to an ordinary differential equation. More general systems may be described by the following equation: x ’ ( 4 = f ( t , 4,
where f is a suitable functional. T h e symbol xi may be defined in several ways. For example, if x is a function defined on some interval [to- T , to a), a > 0, then, for each t E [ t o ,to a ) ,
+
+
(i) x i is the graph of x on [t - T, t] shifted to the interval (ii) x i is the graph of x on [to - T, t ] .
[-T,
01;
In case (ii), f is a functional of Volterra type which is determined by t and the values of x(s), to - T s t. Systems of this form are called delay-differential systems. I n what follows, we shall, however, consider the functional differential equations in which the symbol x L has the meaning described by case (i) and study some qualitative problems by means of the theory of differential inequalities. I n the present chapter, we consider existence, uniqueness, continuation, and continuous dependence of solutions and obtain a priori bounds
< <
3
4
CHAPTER
6
and error estimates. Asymptotic behavior and stability criteria are included. An extension of topological principle to functional differential equations, together with some applications, is discussed. Finally, we dcvclop the theory of functional differential inequalities, introduce the notion of maximal and minimal solutions, prove comparison theorems in this setup, and give some interesting applications. 6.1. Existence
Given any T > 0, let V r L= C[[-T,O], R"] denote the space of continuous functions with domain [ - T , 01 and range in Rn. For any element $ E P,define the norm
R"]. For any t >, 0, we shall let x L denote Suppose that x E C[[--7, a),
a translation of the restriction of x to the interval [t - T , t ] ; more
specifically, x t is an element of V n defined by
I n other words, the graph of x t is the graph of x on [t - T, t] shifted to the interval [ - T , 01. Let p > 0 be a given constant, and let
With this notation, we may write a functional differential system in the form (6.1.1)
x'(f) = f ( t , XJ.
DEFINITION 6.1.1. A function x ( t , ,$0) is said to be a solution of (6. I .l) with the given initial function do E C, at t = t, >, 0, if there exists a number A > 0 such that (i) x(t, , $ J is defined and continuous on [to- T , to x,(t, , $0) E c, for t o to t o -tA ;
(4 X&o ,$0)
< <
= $0 ;
+ A]
(iii) the derivative of x(t, ,$") at t , x'(t, ,#,)(t) exists for t E [ t o ,to and satisfies the system (6.1.1) for t E [ t o ,t, A).
+
We now state the following well-known result.
and
+ A)
6.1.
5
EXISTENCE
SCHAUDER'S FIXEDPOINT THEOREM. A continuous mapping of a
compact convex subset of a Banach space into itself has at least one fixed point. T h e following local existence theorem will now be proved. L e t f E C [ J x C, , R"]. Then, given an initial function C, at t = to >, 0, there exists an a > 0 such that there is a solution ~ ( t,+o) , of (6.1.1) existing on [to- T, to a).
THEOREM 6.1.1.
+o
E
Proof.
+
Let a
> 0 and
define y
E
C[[to- 7 , to
doe - to),
to to
r(t)= Id"(O),
-7
+ a ] , Rn] as follows:
< t < to
e t < to f
a
'1,
+
T h e n f ( t , y t ) is a continuous function of t on [ t o ,to a] and hence IIf(t ,yt)ll Ml . We shall show that there exists a constant
<
b E (0, P
such that
+
Ilf(t
9
-
$) - f ( t
II do(0)lI) >
rt)
I/ < 1
<
a ] , $ E C, and 11 $ - y 1/lo b. Suppose that this whenever t E [to , to is not true. Then for each k = 1, 2, ..., there would exist t, E [to , to a] and +k E such that 11 $, - y t , 1l0 < 1/k and yet
c,
llf(tk
Y
#k)
-f(&
,YtJ
+
2 1.
We now choose a subsequence {tkJ such that limp+%tkD= t , exists, and we have a contradiction to the continuity o f f at (tl ,y t l ) . I t now follows that Ilf(t, $)/ M = Ml 1 whenever t E [ t o , to a ] , $ E C, and I/ $ - y t /lo b. Choose 01 = min(a, b/M). Let B denote the space of continuous functions from [to - 7 , to a] into R". For an element x E B, define the norm
<
<
+
+ +
Since the members of S are uniformly bounded and equicontinuous
6
CHAPTER
6
on [to- T , t , 4-011, the compactness of S follows. A straightforward computation shows that S is convex. We now define a mapping on S as follows. For an element x E S, let
(4
m,)
$0
;
+
(ii) T ( x( t ) )= $do) For every x E S and t
11 ~
E
&J(s,
[ t o ,to
( t-&(o)lI )
xs) ds, to
+ a],
< MI t
<
-
< t < to + a. to
1
< Ma < b.
T h u s x t E C, and /I xl- y l 11" b. Hencef(s, xs) is a continuous function of s and IIf(s, x,)ll M for to s to 01. It therefore follows that the mapping I' is well-defined on S. It is readily verified that T maps S into itself and is continuous. An application of Schauder's fixed point theorem now yields the existence of at least one function x E S such that
<
mu)
6) = xt, ; (ii) T ( x ( t ) )= x(t), t,
< < +
< t < to + a,
which implies that
(9 xt, = $0
(ii) x(t)
=
;
+
~ ( 0 ) J!J(S,
xs>ds, to
< t < to +
01.
Since 3c" E S, the integrand in the foregoing equation is a continuous function of s. Thus, for t, t < to 4-01, we can differentiate to obtain
<
x ' ( t ) = f ( t , x,),
t"
< t < to + a.
+
It follows that x(t, , $J is a solution of (6.1. l ) , defined on [to - T, to a ) , and the proof is complete. We consider next the existence of solutions of (6.1.1) for all t 3 t o . T h e following lemma is needed before we proceed in that direction. Let rn
LEMMA 6.1.1.
E
C [ [ t ,- T , a), R,], and satisfy the inequality
D-mft)
< R(t, I m, lo),
t
->
t" ,
whereg t C [ J x R , , R,]. Assume that r ( t ) = r ( t , to , u,)is the maximal solution of the scalar differential equation u'
= g(t, u),
u(to) = U ( ,
existing for t >, t o . Then, provided 1 m,"
lo
< u0
m(f) .=, 4th ~
t
2 t"
7
20
(6.1.2)
6.1.
Proof.
7
EXISTENCE
T o prove the stated inequality, it is enough to prove that m(t) < u(t, f,
>
u"
, €1,
t
2 to,
where u(t, t o , u,, , c ) is any solution of u' = g(t, E
> 0 being
4 + E,
u(t0) = uo
+
€9
an arbitrarily small quantity, since lim u(t, to , uo , 6) e-30
r(t, to , uo).
=
T h e proof of this fact follows closely the proof of Theorem 1.2.1. We assume that the set
Since g ( t , u ) 3 0, u(t, t o ,u,, , c ) is nondecreasing in t , and this implies, from the preceding considerations, that
I mtl
lo = 4
4
7
to
, uo
9
4 = +l).
(6.1.4)
Thus, we are led to the inequality D-m(tl> G
>
I mtl I")
= R(tl
, U ( t , t, , uo , E ) ) , 7
which is incompatible with (6.1.3). T h e set 2 is therefore empty, and the lemma follows. With obvious changes, we can prove
COROLLARY 6.1.1. Let m E C[[t,- T , m), R,],and, for t D-m(t) 3 -g(t,
min
t&-TKS
m,(s)),
> to ,
8
CHAPTER
where g E C [ J x R, , R,]. solution of
Let p ( t ) = p(t, t o ,uo) >, 0 be the minimal
u' = -g(t, u),
existing for t
6
u(t0) = uo
>0
2 to . T h e n uo < min+GsGto mto(s)implies P(t)
< W), t 2 to -
T h e following variation of Lemma 6.1.1, whose assumptions are weaker, is also useful for later applications.
LEMMA 6.1.2. Let m E C[[to- 7 , co),R,], and let, for every t, for which I m t l lo = m(tl), the inequality D-Wd
> to
be verified, where g E C [ J x R, , R,]. Then, the conclusion of Lemma 6.1.1 is valid. T h e proof follows from Lemma 6.1.1 in view of the relation (6.1.4).
THEOREM 6.1.2. Let f E C [ J x V", R"], and, for (t,4) E J x Yn, Ilf(4 4111 < g(t7 II d Ilo),
(6.1.5)
where g E C [ J x R4. , R,] and is nondecreasing in u for each t E J . Assume that the solutions u(t) = u(t, to , uo) of (6.1.2) exist for all t to . Then, the largest interval of existence of any solution ~ ( t ,+o) , of (6.1.1) is [ t o , m).
Proof. Let x(to , $,) be a solution of (6.1.1) existing on some interval [t,,,p), where to i p < co. Assume that the value of /3 cannot be increased. Define, for t E [to, p), m(t) = / / ~ ( t,do)(t)I/, , so that m,: j j x l ( t 0 ,$o)ll. Using the assumption (6.1.5), it is easy to obtain, for t > t o , the inequality
o-44 < s k 1 m, lo). Choosing I m," lo
=
/I
+,)
Ijo
I/ " ( t o , +o)(t)ll
< uo , we obtain, by Lemma 6.1.1, < r ( t , to , u,),
t"
< t < P.
(6.1.6)
Since the function g(t, u) 3 0, r(t, to , uu) is nondecreasing, and, hence, it follows from (6.1.6) that
11 % ( to ,+")ll"
< r ( t , t o , U"),
to
P.
(6.1.7)
6.2.
APPROXIMATE SOLUTIONS AND UNIQUENESS
For any t, , t, such that to
9
< t, < t, < p, we have
which, in view of (6.1.7) and the monotonicity of g(t, u ) in u, implies
-
r(t, > to , uo) - +I
>
t o , uo).
Letting t , , t, -+ p-, the foregoing relation shows that limi+o- ~ ( t, ,+ J ( t ) exists, because of Cauchy's criterion for convergence. We now define .(to ,+o)(p) = limL+o-.(to ,+O)(t)and consider lCIo = xs(to , do) as the new initial function at t = /3. An application of Theorem 6.1.1 shows that there exists a solution x(p, &) of (6.1.1) on [/3, p 011, 01 > 0. This means that the solution a f t o ,+,J can be continued to the right of p, which is contrary to the assumption that the value of /3 cannot be increased. Hence, the stated result follows.
+
COROLLARY 6.1.2. T h e conclusion of Theorem 6.1.2 remains true even when the condition (6.1.5) is assumed to hold only for t E J and 4 satisfying I1 4 I10 = I1 +(O)ll. 6.2. Approximate solutions and uniqueness
DEFINITION 6.2.1. A function x(to ,4, , 6) is said to be an €-approximate solution of (6.1.1) for t 2 to with the initial function +0 E C, at t = to if (i) x(to , + o , E) is defined and continuous on [to - T, a) and %(to 3 $0 ). E c, ; ( 4 .,(to 90 ) . = $0 ; (iii) .(to , +,, , E ) is differentiable on the interval [ t o , a),except for an at most countable set S and satisfies 9
9
for t
E
[ t o , CO)
7
II XYtO I 4 0 - S.
9
.>
( t ) - f @ , %(to
,40 .)I1 I
e
E
(6.2.1)
I n case E = 0, it is to be understood that S is empty and .(to , 40) is a solution of (6.1.1). We shall now give some comparison theorems on r-approximate solutions of (6.1.1).
10
6
CHAPTER
THEOREM 6.2.1.
x C, , R"],and, for ( t ,+), ( t , 4) E J x C, , Ilf(6 4) - f ( t , 4)ll < At, I1 4 4 llo), (6.2.2)
Letfg C[J
-
where g E C [ J x [0, 2 p ) , R,]. Assume that r ( t ) = r ( t , t o ,uo) is the maximal solution of u' = g(l, u )
+ + €1
€2
,
u(t0) = uo
20
existing for t >, t, . Let % ( t o+o , , el), y(t, , & , e n ) be proximate solutions of (6.1.1) such that
II 4 0
Then,
I1 4 f n do , ~i)(t)
-~
9
Proof.
d(t)ll
7
e2-ap(6.2.3)
< ~ ( tto,
7
2 t o . (6-2-4)
t
uo),
Consider the function
m(4
so that
=
I1 X ( t , 4, I
-
9
2 t o . Then, we have, for O-m(t)
d ( t ) - Y(t0
7
II %(to do > €1)
mt
for t
( f o40
< uo.
$" I10
-
el,
40
YdtO 3
7
%)(t)/l,
40 >
4 1
> to,
t
< I/ 4" 4 t < I/ "(to do 4 t
%)(t)ll ) - A t , %(to > 4 0 > 4)ll ll Y'(t0 , 4 0 > &> - J P , YdtO > 4 0 > %>>II
XYtO
9
9
)
-
r'(to
40
7
f
+ + /lf(t!%(to
40
4- f k
9
YdtO
$0
?
.2))ll.
Now, making use of the assumption (6.2.2) and the fact that ~ ( t, , , el), y(tO, I/J~ , 6.) are e l , €,-approximate solutions of (6.1.1), respectively, it follows, from the preceding inequality, that D-m(t)
<
I mt
lo)
+ + €1
€2
Choose uo such that (6.2.3) holds. Then, on the basis of Lemmas 1.2.2 and 6.1.1, the estimate (6.2.4) results immediately. COROLLARY 6.2.1. T h e function g(t, u ) = Lu, L > 0, is admissible in Theorem 6.2.1. In fact, Corollary 6.2. I implies the well-known inequaIity for approximate solutions
a" < I / dn
I1 4 t o
I
9
E d f )
-
-
40
Y(t"
/lo
4" 4(t)ll
eL(t--tl,)
>
I-
€1
1
€2 [eL(t--to)
7
--
I],
t
2 t o . (6.2.5)
6.2.
11
APPROXIMATE SOLUTIONS AND UNIQUENESS
It is possible to weaken the assumptions of Theorem 6.2.1 in that we need not assume the condition (6.2.2) for all +, $ E C, . T o do this, we require the subset C, C C, defined by
THEOREM 6.2.2. Let the assumptions of Theorem 6.2.1 hold except that the condition (6.2.2) is replaced by
for t E J ,
4,$ E C, . Then, (6.2.3)
implies the estimate (6.2.4).
Proof. Suppose that, for some t, > to , 1 mil ,1 = m ( t l ) , where m ( t ) , m , are the same functions as in Theorem 6.2.1. Setting = x,,(t, , 4,, el), ICI = Y#o $0 , 4, we see that $ E c, and 4 t l ) = II +(O) - +(O)Il. Hence, using (6.2.6), we get
+
$7
9
D-m(t1)
< g(t1
7
m(t1))
+ + €1
€2
,
as previously. T h e assumptions of Lemma 6.1.2 are verified, and therefore the conclusion follows.
A uniqueness result of Perron type may now be proved.
THEOREM 6.2.3.
Assume that
< <
(i) the functiong(t, u ) is continuous, nonnegative for t, t to + a, 0 u 2b, and, for every t, < t, < to a, u ( t ) = 0 is the only differentiable function on to t < t, which satisfies
< <
<
u' = g(t, u),
+
u(to) = 0
(6.2.7)
<
for to t < t, ; (ii) f E C [ R , , R"],where
Then, the functional differential system (6.1.1) admits at most one to a such that x,, = + o . solution on to t
< < +
12
CHAPTER
6
Pmof. Suppose that there exist two solutions ~ ( t,+@), , y(to,+o) of (6.1.1) with the same initial function +o at t = t, . An argument similar to that of Theorem 6.2.2 shows that
II 4 t o
+dt)
-
to , +o)(t)II d
r ( t , to , O),
to
,
where r ( t , t o ,0) is the maximal solution of (6.2.7). By assumption, however, we have r ( t , t o ,0) = 0, and hence
to
to A)(t>, tn G t < ti for every t , such that to < t , < t, + a . T h e uniqueness 4o)(t> =
is therefore proved.
> 0 is admissible in
COROLLARY 6.2.2. Theorem 6.2.3.
T h e function g(t, u )
COROLLARY 6.2.3. for u > 0, and
T h e functiong(t, u ) = g(u), whereg(0)
= Lu,L
of solutions
= 0, g(u)
>0
i,,&j ds
=
is another admissible candidate in Theorem 6.2.3. T h e next uniqueness result is analogous to Kamke’s uniqueness theorem.
THEOREM 6.2.4.
Suppose that
< + <
(i) the functiong(t, u ) is continuous, nonnegative for t, < t to a, < t , < t, a, u(t) = 0 is the only function differentiable on to < t < t , and continuous on to t < t, , for which
0
< u < 2b, and, for every t , , t,
+
and (ii) the hypothesis (ii) of Theorem 6.2.3 is satisfied except that the condition (6.2.8) holds for ( t ,+), ( t , +) E R, , 4, $ E C, , and t # to . Then, the conclusion of Theorem 6.2.3 is true.
6.3. Proof.
13
UPPER BOUNDS
We define the function
< < + < <
< <
t t, a, 0 u 26. Since f ( t , 4) is continuous on R, , for t, it follows that gf(t, u) is a continuous function on t, t t, a and 0 u 26. Moreover, because of the condition (6.2.8),
< < +
4 < At, 4
gAt>
< +
< <
for t , < t t, a, 0 u 2b. Theorem 2.2.3 is therefore applicable with gl(t, u ) = g f ( t , u), and hence the conclusion follows from Theorem 6.2.3.
6.3. Upper bounds We give, in this section, some a priori bounds for the solutions ~ ( t,4,) , of the system (6.1.1).
THEOREM 6.3.1. Let j ' C[J ~ x C, , R"], and let x(t, , 4,) be any u,, . Suppose that solution of (6.1.1) such that 11 4, , 1
<
IIf(44111 < g ( t , /I #(O)Il> for t
E
(6.3.1)
J and 4 E %?" such that
I1 d /In = I/ d(O)ll,
(6.3.2)
where g E C[J x [0, p), R,]. Then,
I1 4 t n do)(t)lI G r ( t , to , Un),
(6.3.3)
on the common interval of existence of ~ ( t, ,4,) and u ( t , t, , uo), where r(t, t, , u,) is the maximal solution of (6.1 .2). O n the basis of the proof of Theorem 6.2.2, this theorem can be demonstrated.
14
6
CHAPTER
for a certain E~ > 0, r ( t , t o ,uo) being the maximal solution of (6.1.2) existing for t 2 to . Then, if ~ ( t,+,J , is any solution of (6.1.1) such that 11 bo \lo = uo , we have
/I .(to
< r ( t , to
+o)(t)lI
7
9
(6.3.4)
24")
as far as ~ ( t, &) , exists to the right of to . Proof. Let [ t o ,T] be a given compact interval. Then, by Lemma 1.3.1, the maximal solution r ( t , C) = r ( t , t o , uo , E ) of u' = g(t, u )
+
u(to) = uo
E,
exists on [ t o ,TI, for all sufficiently small
E
>0
+
E
and
lim r ( t , C) = r ( t ) r+n
TI.
uniformly on [ t o ,
I n view of this, there exists an
r(t, .)
< r(t)
+
€0
,
t E [to
7
co
>0
such that
.I.
Furthermore, we have, by Theorem 1.2.1, y(t)
< r(t, C),
t
[to , 71,
E
which implies that r(t)
< r ( t , ). < r(t>
+
€0
f
3
E [to
, .I.
(6.3.5)
T o prove (6.3.4), it is enough to prove that
11 "(to
7
I/ < r ( t , E),
do)(t)
t
E
[ t o , TI.
Assuming the contrary and proceeding as in the proof of Lemma 6.1.1 we arrive at a t , > to such that (i)
I/ X(t"
(4 /I " ( t o (iii)
I/ "*'&
Y
9
+o)(t1)il = Y ( t 1
$o)(t)ll
, +,,)l,
<-:
C);
r ( t , E), t o
= Y(t1
9
6)
=
' i t, ;
II 4 t "
Setting now $ = ~ ( ~ (, t+J, , , we see that of (6.3.5), it follows that, at t = t , , 4 t ) = ll to , +J(t)ii,
+
'j-m(t1)
7
+")(tl)ll.
I/ + E
= ~ ( t,,E), and so, because Q. Thus, we obtain, defining
< s(t1 I1 dJ 11") >
= R(tl
I
Y(t1
41,
6.3.
15
UPPER BOUNDS
which contradicts the relation D-m(t,)
2
+,, €1
= g(t1 , r(t,
,4 )
+
6,
resulting from cases (i) and (ii) above. T h e proof is therefore complete. T h e following theorem gives a more useful estimate.
THEOREM 6.3.3. Let f~ C [ j x C, , R"], g E C [ J x [0, p), R,], and
(6.3.6)
/I 9 /lo 1 4 lo where A(t) >, 0 is continuous on of (6.1.1) such that
=
/I +(O)Il
[-T,
4 t h
a).Let ~ ( t, ,
/I 4 0 110 I Ato lo d
(6.3.7) be any solution
*a
and r(t, t o , go) be the maximal solution of (6.1.2) existing to the right of t o . Then,
/I %(to
>
<
9o)(t)l/ 4) r ( t , t o
3
UO)?
as far as ~ ( t, $o) , exists to the right of to .
Proof.
Consider the function
4)= I1 XPO
so that
9
4So)(t)ll
at),
mi
=
/I %(to > 4o)lI 4 *
I mto lo
=
II 40 110 I Ato lo
By hypothesis, we have
Let u(t, E)
=
u(t, to , uo , c) be any solution of u'
for
E
>0
< uo .
= g(t, u )
+
E,
u(to) = U"
sufficiently small. Since lim u ( t , E ) S i O
=
r ( t , to , uo)
(6.3.8)
16
6
CHAPTER
and u(t, E ) exists as far as r(t, t o ,uo) exists, it is enough to show that m(t)
< u(t, 4
t
3 to *
(6.3.9)
If this inequality is not true, let t , be the greatest lower bound of
numbers t > to for which (6.3.9) is false. T h e continuity of the functions m(t) and u(t, E) implies that (i) m ( t ) (ii) rn(tl)
< u(t, E), =
< t < ti ;
to
~ ( t,,E ) , t
=
t, .
By the relations (i) and (ii), we have D-m(t,) >, U Y t ,
7
4
= g(t1 , 4 t l
?
6))
+
+
E.
(6.3.10)
Since g(t, u ) E is positive, the solutions u(t, E ) are monotonic nondecreasing in t , and hence, by relation (ii),
Setting
+
I mtl In =
=
4ti)
=
4ti
6).
x t l ( t o $,,), , it follows that
/I4 /In I At1 In = I1d(0)Il A(t1)Thus, at t = t , , (6.3.7) holds true with this 4. Hence, using (6.3.6), there results, using the standard computations, the inequality D-44)
< g(t1 , 4 t l ) ) ,
which contradicts (6.3.10). It therefore follows that (6.3.9) is true, and this proves the stated result.
As a typical result, we shall prove next a theorem for componentwise bounds.
THEOREM 6.3.4.
I f i ( t , +)I for each i
=
Let f
< gi(t, 1
C [ J x C, , R"],and let
E $1
1, 2 ,..., n, t
I4i I"
I"
E =I
I 4i-1
10
, I $i(O)l~ 1 4i+1 1"
j...)
I4n lo) (6'3.11)
J , and $ E C, satisfying
I d,(O)I
(i = 1,
z-., a),
whereg E C [ J x R+",H.,."],g(t, u ) is quasi-monotone nondecreasing in u for each t t J . Then, if .x(t,,, 4,) is any solution of (6.1.1) with the initial function +n = ,...,I$,,,), such that 1 +io lo uio , we have
<
I xi(tn ,$n)(t)I
< ~ i ( tto, , uo)
(i
=
1 , 2,***, n),
6.3.
17
UPPER BOUNDS
as far as ~ ( t,+o) , exists to the right of to , where r ( t , to , uo)is the maximal solution of the ordinary differential system u' = g(t, u),
to .
existing for t
Proof.
u(to) = uo
Define the vector function
so that
Then, since 1 (bio
lo
< uio , we have
As before, it is enough to show that mi(t) < u i ( t , E ) ,
where u(t, E)
=
t
2 to,
i
=
1, 2,..., n,
(6.3.12)
(ul(t,c), ..., un(t, c)) is any solution of u' = g(4 u )
for sufficiently small
E
+
E,
4,) = uo + t,
> 0. If (6.3.12) is false, let n
Z = (J [ t E J : mi(t) 2 u,(t, €)I i=l
be nonempty and t , = inf 2. Arguing as in Theorem 1.5.1, there exists an index j such that (i) mj(tl) = U j ( t l 9 E), (ii) mi(t) < uj(t,E), to d t (iii) and
"i(t,)
< %(t, ,
E),
D-w(t1)
< t, ,
# j,
2 "i'(t1 , €1 =
, U(t1 , .))
+
E.
(6.3.13)
Since u(t, E) is nondecreasing in t, it follows from (i), (ii), and (iii) that
18
CHAPTER
+
Setting == x,,(t, ,do),it results that (6.3.1 I ) , we arrive at
6
I +j lo
=
I +j(0)l, and hence, using
T h e quasi-monotone property of g(t, u) in u and the inequalities (6.3.15) yield
+
because of the definition of and (6.3.14). This inequality is incompatible with (6.3.13), and hence the set 2 is empty, which in turn proves the stated componentwise estimates. T h e theorem is proved.
6.4. Dependence on initial values and parameters We shall first prove the following lemma, which will be used subsequently.
LEMMA 6.4.1.
Let f~ C[J x C, , R"],and Iet, for t
E
J , 4 E C, ,
Suppose that r * ( t , to , 0) is the maximal solution of U' =
G(t,U)
through ( t o ,0). Then, if ~ ( t ,,+,J is any solution of (6.1.1) with the initial function bn at t = t o , we have
on the common interval of existence of ~ ( t, ,+,J and ~ * ( t o, , 0).
Pmof.
Consider the function
6.4.
19
DEPENDENCE ON INITIAL VALUES AND PARAMETERS
and hence by Theorem 1.4.1. T h e function G being nonnegative, r*(t, t,, , 0) is nondecreasing in t , and therefore there results the desired inequality (6.4.1).
THEOREM 6.4.1. Let f~ C [ J x C, , R"],and, for t E J , 4,4 E C, , llf(t, $1
f ( t 7
~
$)I1
< At, II d(0)
-
$(O)lI),
where g E C [ J x [0, 2p), R,]. Assume that u(t) = 0 is the only solution of the scalar differential equation (6.1.2) through ( t o ,0). Then, if the solutions u(t, t o ,uo) of (6.1.2) through every point ( t o ,uo) exist for t 3 to and are continuous with respect to the initial values ( t o ,uo), the solutions %(to, of (6.1.1) are unique and continuous with respect to the initial values ( t o ,+o). Proof. T h e uniqueness of solutions of (6.1.1) is a consequence of Theorem 6.2.3, and hence we need only to prove the continuity with respect to initial values. Let ~ ( t ,+o), , ~ ( t, ,$,) be the solutions of (6.1.1) with the initial functions q50 , $o at t = to , respectively, existing in some interval to t to + a. Then, an application of Theorem 6.2.2 yields that
< <
II 4 t o
1
do)(t)
"( to > $o)(t)ll
~
< r(t, to It 4%
- $0
llo),
to
< t < to + a,
+,,
(6.4.2)
where r ( t , to , 11 - $, ) ,1 is the maximal solution of (6.1.2) through the point (to, /I 4"- $, llo). By assumption, it follows that, given E > 0, there exists a 6 > 0 such that r(t, to
7
I/ do
~
$0 110)
< E,
to
< t < t o + a,
provided I/ c$,,- $" /lo < 6. This, in view of (6.4.2), assures the continuity , ,) of (6.1.1) with respect to initial functions 4o . of the solutions ~ ( t,+
20
CHAPTER
6
We now prove the continuity of solutions with respect to the initial time t,, . IAett , > t, and " ( t o, #,), x(t, , #,) be the solutions of (6.1.1) through (to, #,,) and ( t , , do), respectively, existing in some interval to the right. Define m(t)
~
I1 4 t n
7
Co)(t)
?4o)(t)II,
- "(ti
< t < t i + a.
ti
Since, by Lemma 6.4.1, we have mtl
=
/I X t , ( t n
we obtain m(t)
14n)
< f(t),
4" II
~
t,
< r*(ti
9
t o , O),
< t < t , + a,
(6.4.3)
where F(t) = F(t, t , , r * ( t , , t , , 0)) is the maximal solution of (6.1.2) through ( t l , r*(tl , tn , 0)). Now, lim
tl+to+
f(t, t,
, ~ * ( t,,t , , 0)) = f ( t , t , , 0).
Since lim,l,to r * ( t , , t, ,0) = 0 and by assumption, F ( t , t o , 0) is identically zero. This fact, in view of the relation (6.4.3), proves the continuity of solutions of (6. I . 1) with respect to initial time. T h e proof is complete.
COROLLARY 6.4. I . Theorem 6.4. I .
T h e function g(t, u)
= Lu,
L
> 0, is
admissible in
Using the arguments of Theorems 2.5.2 and 6.2.1, we can prove the following theorem on dependence on parameters. We merely state
THEOREM 6.4.2. Let f~ C [ J x C, x R", R"], and, for p xo(t) = xo(to, #o , pn)(t)be a solution of
= po,
let
x' = f ( t , X t Po), 1
with an initial function #o at t
=
t o , existing for t >, to . Assume that
limf(t2 4, P )
!-+PO
uniformly in ( t , #) F J x C, , and, for t
I/ f(t,4, P )
-
4, Pn)
=f(t,
E
J, #,
f(t, $>PII< I~
$J
E
C, , p
E
R",
( t/I ,4 - $ IIJ,
where g F C [ J x R, , R , ] . Suppose that u(t) = 0 is the maximal solution of (6.1.2) such that u(t,) = 0. Then, given E > 0, there exists
6.5.
21
STABILITY CRITERIA
a a(€) > 0 such that, for every p, 11 p - po I/ system x’ = f ( t , X t P)
<< a(<), the differential
7
admits a unique solution x ( t ) = ~ ( t, $, o , p ) ( t ) defined on some interval [ t o ,to 4-a] such that
II x ( t )
~
xo(t)lI
<
< t < t o + a.
to
€7
6.5+ Stability criteria Let us consider the functional differential system (6.1.1). We shall assume that f ( t , 0) = 0, so that the system (6.1.1) possesses the trivial solution. Let us also suppose that the solutions ~ ( t, +o) , of (6.1 . l ) exist in the future.
DEFINITION 6.5.1. T h e trivial solution of (6.1.1) is said to be stable, E > 0 and t o e J , there exists a 8 > 0 such that ~ \ + o ~< ~ 6o
if, for any implies
/I %(to
>
d0)Iln
<
t
€7
>, t” *
DEFINITION 6.5.2. T h e trivial solution of (6.1.1) is said to be asymptotically stable if it is stable and, in addition, for any > 0, to E J , there exist positive numbers S o , T such that / / $, < 6, implies
II xt(to > do)lIo < E ,
t
3 to
+ T.
Simple criteria for stability and asymptotic stability of the trivial solution of (6. I . 1) are given in the following theorems.
THEOREM 6.5.1. LetfE C [ J x C, ,R ” ] , gE C [ J x [0, p ) , R ,],g(t, 0) and, for t E J, E C, such that
+
~
0,
(6.5.1)
holds. If the trivial solution of the scalar differential equation (6.1.2) is stable, then the trivial solution of (6.1.1) is stable.
Proof.
By Theorem 6.3.1, we have
I1 4 t n > dn)(t)ll < ~
( ttn,
9
II do llo),
t
2 tn ,
22
CHAPTER
6
where r ( t , to , 11 $, I ,J is the maximal solution of (6.1.2) through ( t o ,I( $o I l o ) . T h e fact g 3 0 implies that r ( t , to , (1 $o )1, is nondecreasing in t , and therefore it follows that
< r(t, t n
I1 "dto ,4o)l&
3
/I $0
(6.5.2)
2 to .
t
Iln),
Assume that the trivial solution of (6.1.2) is stable. Then, given E -. p, to c J , there exists a 6 > 0 such that 0 r ( t , to
I1 (fu !In) < 6 ,
9
t
2 to
< 6. T h e conclusion is immediate from (6.5.2).
provided
11 do /I,
TIiEoRm
6.5.2. 1,et f~ C [ J x C,, Rn], g E C [ J x [O, 0. Assume that
g(t, 0)
... :
a t >lim inf Wl d(0) h+O-
+ hf(t> $)I1 - II +(O)lll
p),
R,], and
-t- 11 4(0>llD W t )
< s(t2 I1 C(0)lI J J ( t ) ) for t
> to and $ E C,, satisfying II 4 1In I At
I1 $(o)il A ( t ) ,
lo
(6.5.3)
where A ( t ) 2 1 is continuous on [to- 7,a)and A(t)+ GO as t + co. Then, the stability of the trivial solution of (6.1.2) implies the asymptotic stability of the trivial solution of (6.1 .I). Proof.
By Theorem 6.3.3, it follows that
II "(ti]
4o)(t)!l -d(t> ,< r ( t , to
>
ll (fu Ilo I Ato lo),
t
2 tn ,
and, arguing as in Theorem 6.5.1, we have
11 .yt(t,
I
a,,,;',,1
24,
I"
< r ( t , to , I/
$0
/lo I
t B to.
lu),
Since A ( t ) 23 I , the stability of the trivial solution of (6.1.1) is a consequence of 'I'heorem 6.5.1. Now let E = p and designate by 6,, the S-obtained corresponding to p. Suppose that (1 $(,1Io S,, . Assume that, if possible, there exists a sequence {t,J, t,. co as h ca and a solution (lo : : S,, such that x(t,,, +,,) of (6. I . 1 ) with 11
<
--f
---f
II "/,(to 4")lIl~2 6 . ?
7'hen, there results the inequality E
I L4tk
Iu
r(t,s 9 t n , I1 +o /lo I A,,, lo)
% <
< P.
6.5.
23
STABILITY CRITERIA
Since A(t)-+ GO as t + 03, the foregoing inequality leads to an absurdity for sufficiently large k. As a result, the asymptotic stability follows. We now extend the preceding results to the perturbed systems. Corresponding to (6.1.1), let us consider x’ = f(t,
Xt)
+ R(t, 4,
(6.5.4)
where R(t,4) is a perturbation such that the solutions x(t0 , (bo) of (6.5.4) exist in the future.
THEOREM 6.5.3. Let
for t > to and 4 E C, satisfying (6.5.1). Let the perturbation R(t,4) verify the condition
II R(t7 4111 < g(4 II d(0)ll)
(6.5.6)
for t E J , Q E C, such that (6.5.1) holds. Then, the stability of the trivial solution of (6.1.2) implies the stability of the trivial solution of the perturbed system (6.5.4). Proof.
Let
4 E C,
be such that (6.5.1) is satisfied. Then, we obtain
using the conditions (6.5.5) and (6.5.6). If x(t0 , (bo) is any solution of the perturbed system (6.5.4), we get, by Theorem 6.3.3, with A(t) = I , the inequality
I/ 4
t O 4o)(t)ll 7
< r ( t , to
>
II $0
IlO)?
t
2 to ,
where r ( t , t o , 11 4o ] l o ) is the maximal solution of (6.1.2). T h e stated result follows, arguing as in Theorem 6.5.1. THEOREM
6.5.4. Let
24
CHAPTER
and g(t, 0)
6
0. Assume that
. I ( t ) l i m infh-l[Il +(O) A
,n
1- h f ( t , +)I1
- ll4(O)lII -1-
I1
n-A(t) < 0 (6.5.7)
+
for t ' ' t,, and F C,,such that (6.5.3) holds, A(t)being the same function as in 'I'heorem 6.5.2. Suppose that R(t,d) verifies the condition
4t)ll R(t?4)ll
+
< g(t>/I d(0)lI 4 t ) )
(6.5.8)
for t E J and t C, satisfying (6.5.3). Then, the stability of the trivial solution of (6.1.2) implies the asymptotic stability of the trivial solution of (6.1.1).
Pmof. Let follows that
4 E C,
be such that the relation (6.5.3) holds. Then, it
because of the assumptions (6.5.7) and (6.5.8). It is now easy to establish the asymptotic stability, following the proof of Theorem 6.5.2. 6.6. Asymptotic behavior Let us begin with a result that gives sufficient conditions for every solution of (6.1.1) to tend to a finite limit vector as t + CO. 'I'IIEOREM 6.6.1.
Let f
E
en,R"] and, llf(t, +)I1 < At, II 4 ll")? C[J x
for ( t , +)
E
J x Vn, (6.6.1)
where g t C [ J x R , , R,] and g(t, u ) is monotonic nondecreasing in u for each t E J . Assume that all the solutions u ( t ) of (6.1.2) are bounded on [t, , m). 'Then, every solution of (6.1.1) tends to a finite limit 6 as t Go. --f
Proof. By 'I'heorem 6.1.2, every solution ~ ( t, ,&) of (6.1.1) exists on [t,,, m). 1,et x(t,,, +,)) be any solution of (6.1.1). Then, on account of I m n m a 6. I . 1, it follows that
provided 11
+ 1,)
II .$t,,, 4")(t)li 7
< r(t),
t
2 t" ,
(6.6.2)
u,,, where ~ ( t is) the maximal solution of (6.1.2).
6.6.
ASYMPTOTIC BEHAVIOR
25
Since, by assumption, every solution of (6.1.2) is bounded on [ t o ,a), we see from (6.6.2) that every solution of (6.1.1) is bounded on [to, m). Furthermore, for any t , > to and t > t , , we have
Because g(t, u ) 3 0 and consequently r ( t ) is monotonic nondecreasing in t, it results from the inequality (6.6.2) that
This, together with (6.6.3) and the monotonicity of g in u, implies that
Moreover, by the assumption of boundedness of all solutions of (6.1.2), we deduce that r ( t ) tends to a finite limit as t + m. This means that, given an E > 0, it is possible to find a t , > 0 sufficiently large such that
0
e r(t) -
Y(tJ
< E,
t
> t, .
Consequently, we obtain, as a result of (6.6.4),
This completes the proof. T h e next theorem deals with the asymptotic equivalence of two functional differential systems.
THEOREM 6.6.2.
Let u(t) be a positive solution of u'
> g(t, u )
26
6
CHAPTER
for t 2 to such that limt-,mu(t) = 0, where g E C [ J x R, , R]. Suppose further thatf, ,f 2 E C [ J x V n ,R"] and lim inf h h-n-
for t
"I1 440)
ItfO) iWdt,4
~
< d t , I1 4(0)
-
-
f & 4lll
-
/I d(0) - lcr(0)tll (6.6.5)
lcr(0)ll)
> t o and a, II, E Q, where l2
=
[+, 4 E V": 11 &(O)
$(O)ll
~
=
Then, if the existence of solutions for all t
4th t
2 to].
3 t, of
the systems
(6.6.6) (6.6.7) is assured, the systems (6.6.6) and (6.6.7) are asymptotically equivalent. Proof. Let us first suppose that y(to , II,J is a solution of (6.6.7) defined for t 2 ; t , . Let ~ ( t, , be some solution of (6.6.6) such that
/I $0
-
lcro
110
< 4to)-
Define m(t) = II 4 t " > a d t )
Then,
m(t) < u ( t ) ,
-
Y(t" lcr")(t)ll. 9
t 3 t".
If this is false, let t , be the greatest lower bound of numbers t >, to for which m ( t ) z i ( t ) is not satisfied. T h e continuity of the functions m(t ) and ~ ( tguarantees ) that, at t = t , , < .
(6.6.8) and
(6.6.9)
This implies that D_m(t,) 3
On the other hand, defining $
444
= N(t,
1
‘Go)(
and therefore ??&(ti)
=
)I(
6.6.
ASYMPTOTIC
BEHAVIOR
27
because of (6.6.8). Thus, 4, It E Q at t = t, , and hence, using the condition (6.6.5), it is easy to deduce that D-m(t,)
< g(t1
m(t1)).
>
This being incompatible with (6.6.10), we conclude that m ( t ) < u(t), t 3 t o . T h e assumption that limt+mu(t) = 0 now implies that
pt 4 t o
9
+ow
-
Y(to M t ) ?
= 0-
(6.6.11)
If x(t, ,+,) is a given solution of (6.6.6), defined on [ t o , a), arguing as before, we can assert that there is a solution y ( t o ,Ito) of (6.6.7) such that (6.6.11) is satisfied. It therefore follows that the systems (6.6.6) and (6.6.7) are asymptotically equivalent. We shall now give an analog of Theorem 2.6.3 with respect to the following two systems: x' =f&, Y'
4,
=f d t , Y d .
(6.6.12) (6.6.13)
THEOREM 6.6.3. Let fi E C [ J x 'P,R"] and y(to,#o) be any solution of (6.6.13) defined for t 3 to . Assume that fi E C [ J x R", R"], afl(t, x)/ax exists and is continuous on J x Rn.If x(t, t o ,+,,(O)) is the solution of (6.6.12) such that %(to, to ,+,,(O)) = #,,(O) existing for t 3 to , then y(to ,+,J satisfies the integral equation Yt" = 4 0
where @(t,to , xo) = ax(t, to , x o ) / a x o and y ( t ) = y(to , #o)(t). T h e proof of this theorem is very much the same as that of Theorem 2.6.3. It is important to note, however, that (6.6.12) is an ordinary differential system, whereas (6.6.13) is a functional differential system. As an application of Theorem 6.6.3, we have
THEOREM 6.6.4. Assume that (i) fi E C [ J x R", R"], 8f1(t,x)/ax exists and is continuous on E C [ J x Vn, R"];
J x R", and f2
28
CHAPTER
6
(ii) @(t,t o , xu) is the fundamental matrix solution of the variational system %'
=
q-dt, x(t, t o , xo)) ax
such that @ ( t o ,t o , x,,) = identity matrix I ; (iii) for a given solution y ( t ) = y ( t O+,,)(t) , of (6.6.13), existing on [t" a), I
J=q t ,
$7
Y ( W f & , YS) -fi(s?
Y(4)l
ds
+
0
as
t
-j
a*
Then, there exists a solution x ( t ) of (6.6.12), on [to , a), satisfying the relation lim [ x ( t ) - Y(to ,M t ) l = 0. tim T h e proof can be constructed using the arguments employed in Theorem 2.10.3. Finally, we may mention a result parallel to Theorem 2.14.10.
THEOREM 6.6.5.
Assume that
(i) f E C [ J x K", R"], f ( t , 0) = 0, and fz(t, x) exists and is continuous on J x R"; -0, 0 > 0, t E J ; (ii) p[fX(t,O)] (iii) R E C [ J x V n ,R n ] ,R(t, 0) = 0, and there exists an 01 > 0 such that, if / / (b /lo Lt, t E J ,
<
<
where y
t
/I R(t,d)ll e At),
C [ J ,R,] and
0 such that, if t,) 3 T o ,the trivial solution of
Then, there exists a T,,
27'
f ( t , x)
+ R(t,
Xt)
is asymptotically stable.
Pmof. T h e proof is almost similar to that of Theorem 2.14.10. W e only indicate the major changes. Proceeding as in Theorem 2.14.10, we arrive at the following step: and
I1 4 t " a")(tl)ll 3
/I 4
t O
>
d1Xt)Il < S ( E ) ,
=
S(E)
t fE[tl) , tll.
6.7.
29
A TOPOLOGICAL PRINCIPLE
This implies also that
+
+ <
Hence, letting = xl(t,, +,,), we see that 11 /I, N , and therefore the assumption (iii) can be used. It only remains to follow the rest of the proof of Theorem 2.14.10, with necessary changes to complete the proof.
+ F ( t , x), where (i) A is a n x n constant matrix such that p ( A ) < (ii) F E C [ J x R", R"] and, given any > 0, there exist S ( E ) , T ( E )> 0 such that 11 F ( t , x)ll < I/ x I/ provided 11 x 11 < S ( E ) and t 3 T ( E ) ; COROLLARY 6.6.1.
T h e function f ( t , x)
=
Ax
-0;
E
E
is admissible in Theorem 6.6.5.
6.7. A topological principle I n this section, the topological principle discussed in Sect. 2.9 is extended to functional differential system (6.1.1). We assume, in what follows, the uniqueness of solutions of (6.1.1). More specifically, it is assumed thatfE C [ j x Co, R"] andf(t, +) satisfies a Lipschitz condition in for a constant L = L ( p ) , so that we have the following estimate:
+
II to ,4o)(t>- A t n
#o)(t)lI
G I/ 4 n
-
#n IIo
exp[L(t - to)]
on the common interval of existence of the solutions xft, , +,) 7 $0) of (6.1.1).
(6.7.1)
and
y(to
REMARK6.7.1. I n general, we cannot extend the solution x ( t o ,+,) to the left of t , , that is, we cannot guarantee the existence of a S > 0 such that the system (6.1.1) is satisfied with x(t, ,+&t) extended to [to - S - 7,t f ) . If t+ = GO, the solution x(t, ,+") is defined in the future. Also, for every point P, = ( t o ,x,) E Rn+l,there are, in general, infinitely many solutions of (6.1.1) satisfying x(t, , +,)(to)= x, . Let E be an open set in Q = J x S o ,where So = [x E R": I/ x I/ < p ] , and i3E be the boundary of E in Q. DEFINITION 6.7.1. A point Po = ( t o ,x,) E aE is said to be a point of egress if there is at least one solution x(t) = x(t, , +,)(t) of (6.1.1) with x(t,) =: x,,defined on [to T - 6, t,], S > 0, such that ( t , x ( t ) )E E for ~
t , - S < t i t , .
We define
S = [P E i3R: P is a point of egress].
30
CHAPTER
6
DEFINITION 6.7.2. A point Po ( t o , x,,) E aE is said to be a point of strict egress if (i) for every solution x(t) of (6.1.1) with x(to) = x,,, defined on [to 7,t + ) , there is a S > 0 [t+ and S may depend on x(t)], such that ( t , x ( t ) )E E*, where E* is the exterior of E with respect to Q, for t,, t t,, 6 ; and (ii) whenever x ( t ) can be extended to some [t,, T - u, t ') [u > 0 depending on ~ ( t )there ] , is an E > 0,0 < E < u, such that ( t , x(t)) E E for to - E t r, t o . We shall denote by S* the set of points of strict egress. We see from the preceding definitions that Po = ( t o ,x,,)E S* implies that Po E S if there is at least one do E WL, $,,(0) = x,,, such that the solution ~ ( t,,$,,)(t) can be extended to the left. These definitions coincide with the definitions given in Sect. 2.9, if r = 0. ~
< +
<
~
DEFIKITION 6.7.3. A solution x(t, , $,,) of (6.1.1) is said to be asymptotic with respect to E if ( t , x ( t ) )E E for t,, t < t b. Let x(t) = x ( t , , d,,)(t), with x(t,) = +,,(O) E E. If x ( t ) is not asymptotic with respect to E, then there exists a t , > to such that Q = (tl , x ( t l ) ) E S and ( t , x ( t ) )E E for to t < t , . We denote the point Q by ~ ( t ,,,+,,). For every to > 0, let E(t,,) be the set of points (to, x) E E. Again, for t, > 0, $,, E E7L, define
<
<
If G(t,,,$,,) is not empty, we define a mapping taking G(t, , +,,) into S , - h ( 0 ) $,J, where Po = ( t o ,yo). Let S C S* by ~ ( p , )=) ~ ( t , yo and w be any set satisfying
+
sc w c s*. We then define a mapping K : G(t0, +,) u w (i) K(P0) = 4P">if Po E G(to (ii) K(P,,) = Po if P,, E w .
LEMMA 6.7.1. into w.
9
+w
as follows:
$0);
T h e mapping K is a retraction from G(t0 ,+,) u w
Proof. Since K(Po)= Po if P,,E w , it is enough to prove that K is continuous. This can be shown by using the estimate (6.7.1) and following the standard techniques. We can now prove a theorem analogous to Theorem 2.9.1.
6.7.
THEOREM 6.7.1. conditions hold:
Let
31
A TOPOLOGICAL PRINCIPLE
and to
E
> 0.
Assume that the following
+
(i) for every ( t o , y o )E E(to), it follows that y o - C$~(O) E C, ; (ii) S C w C S*; (iii) there exists a set 2, 2 C E(to)LJ w such that 2 n w is a retract of w and Z n w is not a retract of 2. Then, there exists at least one point xo E R" such that ( t o I xo
and x ( t o ,xo
+
+O(O))
E
2 - w ,xo
+ 4" c, , E
+ $o)(t) is asymptotic with respect to E.
Note: If p = 00, so that So = R", we can drop the condition (i). Furthermore, in the proof given below, it will be assumed that $o(0) = 0, since there is no loss of generality in doing so.
Proof. By Lemma 6.7.1, the mapping K is a retraction from G(to, LJ w into w. Since 2 n w is a retract of w ,there is a retraction K * from w into 2 n w. Then, the mapping T = K* . K is a retraction from G(to, LJ w into Z n w. By the assumption (i), it follows that xo E C, for every ( t o , xo) E 2 - w. Suppose that x(t0 , xo d o ) ( t )is not asymptotic with respect to E for every ( t o , xo>E 2 - w. Then, r(to, xo + $,) exists for every (to , xo) E 2 - w , and hence
+
+
2 = (2 n W )u (2 - W ) C w
LJ
G(t, ,do).
This implies that the restriction of the mapping T to 2 is a retraction from Z into w n Z , which is a contradiction. T h e proof is therefore complete. Corresponding to Theorem 2.9.2, we now state the following result, which gives sufficient conditions to ensure that S C S".
THEOREM 6.7.2. Let (i) E
=
LetfE C [ J x C,, R"], u E C[B, Rp],and ZI E C[Q,Rq].
[ ( t , x) EQ: uj(t,x)
=
[ ( t , x)
EQ:
I<j
=
[ ( t , x)
E
1 <j
< 0,
0 and u3(t,x)
< 0,
1
l<j
vk(t,x)
< 0 and
u,(t, x)
=
l
Q: v,(t, x)
=0
I
and
u j ( t ,x)
< 0,
vk(t,x)
< 0,
vUlc(t, x)
< 0,
32
CHAPTER
6
Assume that, for each P,, = ( t o, xu)E L , , M B and every solution x(t) of (6.1.1) satisfying x(t,,) x,,, it follows that
> 0, < 0.
%'(t, x(t))t=t, %'(t,x(t))t=t,
Then, if w = Li - Uj"=l M j , we have S C w C S". T h e proof can be constructed by making necessary changes in the proof of Theorem 2.9.2.
6.8. Systems with repulsive forces Let F E C [ J x V" x V n ,R"] and F ( t , 4, $) be locally Lipschitzian in C$ and $. We consider the second-order functional differential equation x"
=F(t,Xt
7
(6.8.1)
Yd,
where y ( t ) = x ' ( t ) . This can be written in an equivalent form: (6.8.2)
DEFINITION 6.8.1. A solution x(t) of (6.8.1), defined for t 3 t o ,is said to be asymptotic in the sense of Wintner if u 2 ( t ) = x(t) . x(t) is nondecreasing for t 2 to . 6.8.1. Let E' E c[]X V n X Vn,RTL]and F ( t , $, $) be locally 1,ipschitzian in 4,$, and, for all +, $ E W , $(O) # 0, t 3 T ,
'rmoREM
0) .F ( 4 A 4 + +(O) . #(O) > 0. Then, for every to > T , & , $o E Y", there exists a family of points (x,,,y o )E RZn,depending at least on n parameters, such that the solutions "(to , x,, + , y o + $,,) of (6.8.1) are asymptotic in the sense of Wintner.
+,,
Proof. T o prove the theorem, we have to show the existence of a family of solutions x(t) of (6.8.1) satisfying [u2(t)]' = 2x(t) . .x'(t)
Let Q
=
<0
[ T , co) x R27L. For some b E,
=
for
t
3 to > T.
> 0, T < t , < t o , +o , $o E Vn,let
[ ( t ,X, y ) E R2nt1:~ ( tX,, y)
< 0, ~
( tX,, y )
< 01,
6.8.
By Theorem 6.7.2, we can conclude that the sets S, , strict egress points obey s b
c c Wb
s,*of egress and
sb*,
where Wb
33
SYSTEMS WITH REPULSIVE FORCES
= [ ( t , x, y ) E l P + l :x * y = 6, t
> 4.
Let us take 2, C Eb(to)u w, , where 2, is a segment connecting two points ( t o , , qI) and ( t o , qz) located in two distinct components of w, with ( t o , 0, 0 ) 4 2, . As a consequence of Theorem 6.7.1, it follows that there is at least one point ( t o ,f , q) with
e2,
<,
such that the solution
+ & , r ] + &,),
where 7, = ([
is asymptotic with respect to E, , that is,
x(t) . y ( t ) < b,
t
2 to.
Let a sequence {bll,}, b,, > 0, 6, + 0 as m + co, and the sets .Zba be chosen such that Z,nt+lC Z,, . Then, for every m, there is (xm ,yrr8), with (to , x,, +,,(O), ytn $o(0))E Z,, - w,, , such that the solution ( 4 t o 7 Y " , ) ( t ) ,Y(t0 Y , / J ( t ) ) , where Y n t = ( X n t $0 7 Ym +o>, is asymptotic with respect to E,,. By taking a convergent subsequence of {xrtt, ynt>with limit (xo ,yo),it follows that x(t) . y ( t ) 0, t >, to , where x(t) = X ( t 0 Y ) ( t ) , Y ( t ) = Y(to 7 Y ) ( t ) , and Y = (Xo $0 Yo $0). Since 5,n Zbl is empty for b, # b, , we see that it is possible to take (xo,yo) depending on n parameters. T h e proof is complete.
+
+
+
9
DEFINITION 6.8.2.
forces if
+
+
9
<
9
+
Equation (6.8.1) is said to be a system with repulsive
d(0) . F ( t ,4,$1
> 0,
t
2 T,
474 E wl-
COROLLARY 6.8.1. If Eq. (6.8.1) is a system with repulsive forces, then the conclusion of Theorem 6.8.1 is valid.
6.9. Functional differential inequalities
It is natural to expect that a comparison theorem for functional dif-f'erential equations, analogous to Theorem 1.4.1, would be equally useful in some situations. T o prove such a result, we need to show the existence of maximal solution for functional differential equations. We are thus led to the study of functional differential inequalities. With this motive, we shall consider a functional differential equation of thc form (6.9.1) x' = f ( t , x,4, which is convenient for some later applications. T h e existence theorems 6.1.1 and 6.1.2 are valid for such an equation, with obvious changes. Let us begin with the following basic theorem on fundamental differential inequalities, recalling that V = C[[-T, 01, R].
'I'HEOHEM 6.9.1. Let f~ C [ J x R x Y , R] and f ( t , x,+) be nondecrcasing in d, for each ( t , x). Let x,y E C[[to T , m), R] and ~
Xi"
(6.9.2)
Yt, '
Assume further that (6.9.3)
(6.9.4)
Pyoof.
If the assertion (6.9.4) is false, then the set
is nonempty. Let t , Moreover, .?(tl)
z
=
=
inf Z. It is clear from (6.9.2) that t ,
[ t E [t,, , 02): x ( t ) 3 y ( t ) ]
and
=Y(tJ
x(t)
< Y(t),
t E [ t o , tl).
>to. (6.9.5)
'I'hus, we obtain for small h <<0
which, in turn, implies that D-x(t,) b D-y(t,).
(6.9.6)
6.9.
35
FUNCTIONAL DIFFERENTIAL INEQUALITIES
From the relations (6.9.2) and (6.9.5), we deduce that Xt,
>
which, in view of the monotonic character of f ( t , x, +) in f(t1 t x(tl)?Xi,)
G f(t1
?
+, yields
x(tl), Yt,).
(6.9.7)
On the other hand, the relations (6.9.3) and (6.9.6) lead to the inequality f(tl > "-(tl>?%,>
>f(tl
Y(tlhY t J
which is incompatible with (6.9.7) because of (6.9.5). T h u s the set 2 is empty, and the result follows. REMARK6.9.1. T h e conclusion (6.9.4) remains valid even when the inequalities (6.9.3) are replaced by D-JC(t)
) s
Xi),
D-YW >, f(4Y ( Q Yi).
R ] , and let x+'(t) exist for DEFINITION 6.9.1. Let x E C[to- T, a), t E [ t o ,a).If x(u) satisfies the differential inequality x+'(t> < f ( t ,44, 4
7
t
E
[to , a>,
it is said to be an under function with respect to (6.9.1). On the other hand, if x+'(t>> f ( 4 x ( t > ,Xi>,
t
E
[to , a),
x(u) is said to be an over junction with respect to (6.9.1). As in the case of ordinary differential equations, the next theorem shows that any solution of (6.9.1) can be bracketed between its under and over functions.
THEOREM 6.9.2. Let y(u), z(u) be under and over functions with respect to (6.9.1). Let f~ C [ J x R x V , R] and f ( t , x, +) be nondecreasing in +, for each ( t , x). Suppose that x(t, ,+,) is any solution such that of (6.9.1) defined on [ t o ,a), Yt,
Then, Y(t)<
3
< do < Z t O .
doP> < 4 t h
t E [to a>1
T h e proof follows by a repeated application of Theorem 6.9.1.
36
6
CHAPTER
We shall now consider the existence of maximal solution of (6.9.1).
DEFINITION 6.9.2. Let ~ ( t,,+(J be a solution of (6.9.1) defined on [t, , t, a). For any other solution " ( t o ,4,) of (6.9.1) defined on the same interval, if
+
"(t,
< Y(t,
7
t t "0
(bo)(t),
>
> f"
+4
then r(t,, ,$(J is said to be the maximal solution of (6.9.1). A similar definition may be given for the minimal solution by reversing the preceding inequality.
THEOREM 6.9.3. Let f E C [ J x R x Y , R] and f ( t , x, 4) be nondecreasing in 4 for each ( t , x). Then, given an initial function 4,E Y at t t,, , there exists an m1 > 0 such that Eq. (6.9.1) admits a unique maximal solution r(t0 , 4") defined on [to , to q).
+
~
Proof. Following the proof of Theorem 6.1.1, we obtain a , b Suppose that 0 :E b/2. Consider the equation
<
X'(t) =f ( t , " ( t )
Let +o
+
E
be the initial function at t
, xt)
2
t,
+
> 0.
(6.9.8)
E.
. Observing that
+
whenever t E [t, , to a ] , x E S, , 4 E C, , I x - y ( t ) - E 1 < b and 1 4 - y t - E ,1 <, b, we deduce, on the basis of Theorem 6.1.1 that (6.9.8) has a solution x(t,, , 4, , E ) on the interval [t, - 7,t , 4, where cyI = min[a, (6/(2M t h ) ] . For 0 < : c2 E , we have
+
<
"t,(to
3
a"
2
"dt"
€2)
a"
!
E J t )
$0
9
El)>
(b(J
f(t>
>
" f ( t ? "(f"
,
> $0 >
"'(t" ,
>
' 2 ) ( t ) , Xt(t,l
a, , ' , ) ( t ) , "dt,
!
$0 > '2))
, a0 , )1'
+ +
'2
,
€2
.
We can now apply Theorem 6.9.1 to get .$to
14"
>
4 t )
< "(t"
7
$,I
1
E,)(f),
t
6
[t" , to
+
011).
Since the family of functions x(t, ,4" , ~ ) ( t are ) equicontinuous and uniformly bounded on [ t o ,t, (Y~),it follows by Ascoli-Arzela's
+
6.9.
37
FUNCTIONAL DIFFERENTIAL INEQUALITIES
theorem that there exists a decreasing sequence ( e n } , E , ~ such that
---f
lim
n+m X ( t 0
7
do > E d t )
= r(t0
1
0 as n
-+
00,
d")(t)
+
uniformly on [to, to al). Clearly, ~ ( , ( t,,+o) = . T h e uniform continuity of f implies that f ( t , ~ ( t,, , ~ , ) ( t ) , xl(t, , , en)) tends uniformly to f ( t , r(to , +o)(t),y l ( t 0 , $o)) as n 00, and, thus, term-byterm integration is applicable to
+,
---f
4 t o Id0
7
%At)= dO(0) + E n t
.cI [ f h
do
X(t0
to
9
E
m
, Xdt"
> $0
7
4 )+ %I ds,
which, in turn, shows that the limit ~ ( t ,$o)(t) , is a solution of (6.9.1) 4. on [ t o 7 t o We shall now show that ~ ( t, $o) , is the desired maximal solution be any solution of (6.9.1) defined of (6.9.1) on [to, to q).Let "(to, on [ t o ,t, al). We then have
+
+
+
4") < Xf,(tO > d o , €1, do E ) ( t ) 2 f ( t > X(t0 40 , E ) ( t ) ,
%,(to XYtO
7
XYt,
for 0
7
7
40x4 < f ( h "(to
7
%(to , $0 , 4)
do)(th %(to
> $0))
+
+
6,
c,
< E < 6/2. By Remark 6.9.1, it follows that X(t0
Since
9
9
40)(t)
< X ( t 0 do 7
?
lim r+U X ( t " , do , .)(t)
uniformly on [ t o ,to
t t
E)(t),
=
[to to ?
+ 4.
'(to ,d">(t)
+ al), the theorem is proved.
REMARK 6.9.2. Under the assumption of Theorem 6.9.3, we can show the existence of the minimal solution also. T h e proof requires obvious changes. We are now in a position to prove the following comparison theorem for functional differential inequalities. Let %+ denote the set of all nonnegative functions belonging to %'.
THEOREM 6.9.4. Let m E C[[to-r>-m(t)
T,
a), R+],and satisfy the inequality
< j ( t , m(t), m,),
t
>- t" ,
38
CHAPTER
6
where f E C[J x R, x V+ , R]. Assume that f ( t , x, 4)is nondecreasing in 4 for each ( t , x) and that ~ ( t ,,do), , E Y+ , is the maximal solution of (6.9.1) existing for t 3 t o . Then, mto $(, implies
<
m( t ) G
Pyoof, +o
+
E
Y(t"
, do)(t),
2 to .
t
(6.9.9)
Let ~ ( t,4,, , , c) be a solution of (6.9.8) with an initial function at t = t,, . Then, we have
< do -I-
mto
< f ( t , m(t), 4,
DWt) "'(to
1
do
9
€7
> f ( t ,X ( t , > do >
E)(t)
w,x * ( h,do
>
4).
These inequalities imply, by Theorem 6.9. I , the relation m( t )
< X(t,
>
t
d o > f)ft),
>, t o *
T h e conclusion (6.9.9) results from the fact that uniformly for t 2 to ,
lim "(to , do , ~ ) ( t= ) r(t, ,&)(t) <+O
T h e following theorem provides an estimate for the difference between a solution and an approximate solution of a functional differential system.
THEOREM 6.9.5. Let g E C [ J x R+ x Y , , R+]and g ( t , u, u) be nondecreasing in a for each ( t , u). Assume that f t C [ J x Rn x Yn,R"], and, for t E 1,x,y E R", and 4,$I E %PL,
411 < g(tt I/ "
llf(t? x, 4 )
~
Y
/I>/I 4
~
(6.9.10)
1cI 11).
Let "(to , +,)(t),y ( t ) = y(to, 4", 8 ) ( t ) be a solution and a &approximate solution of the system (6.9.1 1) x' = f ( t , "0 "1
defined for t
3 to . Let
.' .,
~ ( t,,a) be the maximal solution of = g(t,
with an initiaf function whenever / / rj,) - I/
+,, <
(7
.t)
k 8,
E %?+, at t == t o , existing for t
(7,
we have
2 t o . Then,
6.9.
FUNCTIONAL DIFFERENTIAL INEQUALITIES
39
P?f. Define m ( t ) = II X ( t , ,+o)(t)- Y(t)ll, so that m , = II x,(t, ,4,)- y1 II, for t >, t,, . Then, as in Theorem 6.2.1, it is easy to obtain the inequality Km(t)
< g(t, m(t), m,)+ 6,
t
> t"
*
T h e desired result follows by Theorem 6.9.4. A uniqueness theorem of Perron type is an immediate consequence of Theorem 6.9.4.
THEOREM 6.9.6. Let g E C [ J x R, x V+ , R,] and g ( t , u, u) be nondecreasing in u for each ( t , u). Assume that f E C [ J x R" x V n ,R"] and that (6.9.10) holds. If the maximal solution of u' = g(t, 21,
4,
with the initial function u = 0, is identically zero, then there is atmost one solution of the system (6.9.11) on [ t o , t,, u). T o give yet another interesting application of Theorem 6.9.4, let us consider the differential-difference system
+
Y ' ( t ) = f ( t ,Y(t>,y ( t
- T>),
7
> 0.
(6.9.12)
If T is small, it is natural to expect that it can be neglected, and we can consider the ordinary differential system x' = f ( t , x,x).
(6.9.13)
THEOREM 6.9.7. Assume that (i) f E C [ J x R" x R", R"], fz(t, x, x) exists and is continuous for
t E J , x E R", and
x,4 1
P[f&
(ii) for t
E
J , x,y , z
E
<
-01,
t 3 0,
R",
llf(t>.>Y) - f ( t , x,z)ll
(iii) for t
E
J and y
E
x R";
< L IIY
-
I/;
C[[t,,- T, a), R"],
and @ ( t , t,, , x,,) is the solution of the variational system (6.9.14)
40
CHAPTER
6
such that @(to, to , x,,)is the identity matrix I , where x(t) = x ( t , to , xo) is any solution of (6.9.13). 7 n/I,N, every solution of (6.9.12), defined on [ t o ,a), l'hcn, if 0 -, tends to zero exponentially, as t + CO. ~
Pro($ I,et y ( t ) y ( t o ,+,))(t)be any solution of (6.9.12), defined on [ t o , CO). Then, by Theorem 6.6.3, it follows that :
4") 4" At" 4"&) - x ( t , to ,4"(0)) Y/"(t"
9
~
>
7
t J'I@(hs, V ( N f ( S , Y ( 4 Y ( S - 7)) (0
for t
3 t o . Moreover, on the basis
-fh Y(+ Y ( 4 l 4 (6.9.15)
of Theorem 2.6.4, we deduce that
Also, using assumption (i), we have
Thus, we obtain, on account of (6.9.15), (6.9.16), and (6.9.17),
because of condition (ii). Observe that
6.9.
FUNCTIONAL DIFFERENTIAL INEQUALITIES
41
Hence,
where M
= LNr.
We now define
so that This, together with the fact that
yields the functional differential inequality u’(t)
< -au(t) + M
sup
Ut
-27 G.9GO
.
By Theorem 6.9.4, we therefore get u(t)
provided u l , = defined on -27 of
< r(to
t
u)(t),
9
> to ,
(6.9.19)
< u, where is the initial function at t = t o , < s < 0, and r ( t ) = r(to , u ) ( t ) is the maximal solution
11 &,I/
(r
+M
U’ = - a ~
SUP -27<
e a
u!.
(6.9.20)
All that is required is to find the explicit form of r ( t ) . Let r ( t ) = u(0) e-v(t-tJ.
Then, we get, from (6.9.20), the equation - y = -a
+ MeVT.
It is now easy to see from this that r ( t ) is the solution of (6.9.20) with y > 0, provided a > M , which implies that 0 <7
< a/LN.
T h e conclusion of the theorem is now immediate from the inequality
(6.9.19), noting that
T h e proof is complete.
I1 Y ( t 0
7
+o)(t)ll
< 49.
42
CHAPTER
6
6.10. Notes
Theorem 6.1.1 is adapted from the work of Driver [2, 31. See also Lakshmikantham and Shendge [l] and Yoshizawa [3]. T h e remaining results of Sect. 6.1 are due to Lakshmikantham and Shendge [l]. See also Driver [2, 31 and Sugiyama [ l , 51. Theorems 6.2.1, 6.2.2, and 6.2.3 are taken from Lakshmikantham [7]. Theorem 6.2.4 is new. Section 6.3 contains results adapted from Lakshmikantham [7]. T h e results of Sect. 6.4 are new. See also Driver [4], Sugiyama [4] for particular cases. Concerning linear functional differential equations, see Hale [4] and also Sugiyama [6]. T h e stability criteria discussed in Sect. 6.5 are taken from Lakshmikantham [7]. All the results of Sect. 6.6 are new. T h e extension of Wazewski’s topological principle and applications given in Sects. 6.7 and 6.8 are due to Onuchic [l]. Section 6.9 consists of the results of Lakshmikantham and Shendge [I], except Theorem 6.9.7, which is new. For the notion of extremal solutions, see also Sugiyama [8]. Delay-differential equations have been discussed in a recent monograph of O&ztoreli [l], which also treats time-lag control systems. See also Driver [ l , 2, 31 and Lakshmikantham [7] concerning this area. A systematic presentation of fundamental theory in neutral functional differential equations may be found in the work of Driver [4]. T h e books of Bellman and Cooke [4] and Halanay [22] are famous contributions in this field. T h e survey papers by Bellman and Danskin [2], Bellman et al. [l], El’sgol’ts and Myshkis [l], El’sgol’ts, Zverkin, Kamenskii, and Norkin [ l , 21, Hale [7], and Kamenskii [5] give important information about the development of the theory. T h e books of El’sgol’ts [7], Krasovskii [5], Myshkis [6], Pinney [l], and Hahn [l] also discuss equations of this type. For related work, see also Bullock [I], Cooke [I-81, El’sgol’ts [l-61, El’sgol’ts, Myshkis, and Shimanov [l], Fodcuk [l], Franklin El], Hale [2, 3, 9, 101, Hale and Perello [I], Hastings [I, 21, Jones [2-51, Kakutani and Markus [l], Karasik [2], Myshkis [l-71, Perello [I], Shimanov [l-lo], Stokes [l], Wright [l-41, and Zverkin [l-31.
Chapter 7
7.0. Introduction I n extending the second method of Lyapunov to the differential equations with time lag, one has a choice of treating the solutions as the elements of a function space or as elements of Euclidean space, for all future time. Each approach has certain advantages, although it appears natural that the proper setting for the study of functional differential equations is a function space. If we choose to study from the point of view that the solutions define curves in a function space, the concept of Lyapunov functional may be used to discuss many problems including stability and boundedness of functional differential systems. Moreover, if Lyapunov functionals are employed for direct theorems on stability, the converse problem of showing the existence of Lyapunov functionals can be solved in a manner analogous to that in ordinary differential equations. T h e main advantage in using Lyapunov functionals consists, however, in applying the converse theorems for the study of perturbed systems. I n this chapter, we study a variety of problems of functional differential systems by means of Lyapunov functionals and the theory of differential inequalities.
7.1. Stability criteria We shall consider the functional differential system (7.1.1)
where f E C[J x C,, , R"], C,, being, as before, the set
44
CHAPTER
7
We shall assume thatf(t, 0) 0, so that the system (7.1.1) possesses a trivial solution. Suppose also that f ( t , #) is smooth enough to guarantee the existence of solutions of (7.1.1) in the future. Let ~ ( t ,, + ,,) be any solution of (7.1.1) with an initial function 4"E C, at t = to . DEFINITION 7.1.1. T h e trivial solution of (7.1.1) is said to be (S,) equistable if, for each t > 0, to €1,there exists a positive function S = S ( t , , E ) , which is continuous in to for each E, such that the inequality
ll4ollo
implies
/I "dt,
4o)Ilo
?
<
<6 t
€7
3 to.
On the strength of this definition and that of Definition 3.2.1, we can formulate thc other notions of stability (S,) to (Sl,,)of the trivial solution of the functional differential system (7.1.1). T h e same is true concerning other various types of stability and boundedness definitions. DEFINITION 7.1.2.
For any V E C [ J x C,, , R,], define
L)+ 1 (1, x , ( t , , $0))
=
lim sup W V ( t h-O+
V t ? %(to
>
+ h,
Xt+h(&
,do))
4o>)1*
(7.1.2)
Sometimes, we also define D+V(t,4) = lim sup h-l[V(t h+O+
+ h, ~ ~ + ~ ( t , V+ ()t , + ) ] . -
(7.1.3)
where it is understood that x(t,#) is any solution of (7.1.1) with an initial function # at time t.
REMARK7.1.1. If the uniqueness of solutions of (7.1.1) is assured, both the definitions (7.1.2) and (7.1.3) are identical, since letting # xt(t, , &) and noting that ~ ~ + ~,A)) ~ (= t , , ,,,(t,41, h 2 0, because of uniqueness, (7.1.2) reduces to (7.1.3), and vice versa. We now state certain fundamental propositions regarding Lyapunov stability of the trivial solution of (7. I . I). ~
THEOREM 7.1.1. Assume that there exist a functional V ( t ,4) and a function g(t, u ) fulfilling the following properties: (i) V EC [ ] x C,, R , ] , and, for t 3 t o , I)+ G-(t, X , ( t ,
9
$0))
-<s(4 L'(4
%(to
,$0)));
7.1.
45
STABILITY CRITERIA
(ii) g E C [ J x R, , R], and g ( t , 0) = 0; (iii) there exist functions b E Z and a E C [ J x [0, p), R,],a(t, u ) E X for each t E J , such that
Wld 110)
< V(4d) < 4 4 II d llo),
(4d) E J x
c,
*
Then, the trivial solution of the functional differential system (7. I. 1) is (1") equistable if the trivial solution of (6.1.2) is equistable; (2") equi-asymptotically stable if the trivial solution of (6.1.2) is equi-asymptotically stable.
Proof. Suppose that the trivial solution of (6.1.2) is equistable. Let 0 < E < p , to E J be given. Then, given b(c) > 0, to E J , there exists a positive function 6 = S ( t , , c) that is continuous in to for each E such 6 implies that uo
<
u(t, t o , uo)
< b(c),
t 2 to
(7.1.4)
*
<
Choose uo = a(t, , I/ +o ] l o ) so that V(to, uo . By the assumption on a(t, u), it follows that there exists a 6, = Sl(to, E), which is continuous in to for each E, such that
/I do I10
< 8,
and
a(t0
9
I/ do 110)
<8
hold simultaneously. T h e choice of uo , the condition (i), and Theorem 1.4.1 imply the inequality V ( t ,%(to
>
do)) < r ( t , to
1
4
,
2 to
t
7
(7.1.5)
where r(t, t o ,uo) is the maximal solution of (6.1.2). This, together with (7.1.4) and assumption (iii), yields, for t 3 t o , b(l/ 4
t O > d0)Ilo)
< v(4 %(to
,401)
e r ( t , t o , uo) < b(c),
which implies that
II %(to
<
d0)llo
<
€9
t 3 to
9
provided /I +o 1l0 6, . This proves (1'). T o prove (2"), it is enough to show that (S,) holds. For this purpose, let 0 < E < p and to E J be given. Then, since (S,*) is satisfied, given b(c) > 0, to E J , there exist positive integers So = So(to) and T = T(to, E ) such that (7.1.6) u(t, t o 1.0) < b ( ~ ) , t 3 to -t T ,
46
<
provided uo So . Choosing u,, a = s,(t,) such that
so
7
CHAPTER
=
a(to , /I $,,
and
I1 4 0 I10 ,< $0
4to>
[lo),
we obtain, as previously,
It 4 0 110)
<
80
<
are satisfied at the same time. Moreover, it follows that V(to, $o) uo . Let 6",(t,, p ) be the number obtained by taking E = p and let So* = min[S0 , Assume that 11 +o jjo 8,". Then, using (7.1.5), (7.1.6), and V ( t ,+), we get the fact that b(l/+ilo)
so].
<
<
b(lI %(to
?
4O)llO)
G V(t7 xdto
< b(4,
9
t
$0))
2 to
G r(4 to > uo)
+ T,
which, in turn, yields that
II %(to
>
4o)llo
< 6,
t
<
2 to + T ,
whenever 11 I/o So*. This establishes (2"), and the proof of the theorem is complete.
COROLLARY 7.1 .l. T h e function g ( t , u) = 0 is admissible in Theorem 7.1.1, to give equistability of the trivial solution of (7.1 .I). COROLLARY 7.1.2. Let g ( t , u)= X ( t ) w(u), where X E C [ J ,R],w(u) 3 0 is continuous for u 0, ~ ( u> ) 0 for u > 0. If
>
<
for some uo > 0, every to 3 0 and to t < m, theng(t, u ) is admissible in Theorem 7.1. I to assure equistability of the trivial solution of (7.1.1). If, on the other hand, there exists a T , to T < co,verifying the property
<
then g(t, u ) is admissible in Theorem 7.1.1 to guarantee equi-asymptotic stability of the trivial solution of (7.1.1).
THEOREM 7.1.2. Assume that there exist a functional V(t,4) and a function g(t, u ) satisfying conditions (i) and (ii) of Theorem 7.1.1. Suppose further that there exist functions a, b E X such that, for ( t ,4)E J x Co 9
b(ll4
110)
G V ( t ,4) < 44 110)-
7.1.
47
STABILITY CRITERIA
Then, if one of the notions (S,*) through (S,*) holds, the corresponding one of the notions (S,) through (S,) holds for the system (7.1.1). It is not difficult to construct the proof corresponding to each of the statements of this theorem, on the basis of the arguments used in Theorem 7.1.1 and respective theorems in Sect. 3.3. and 3.4. We leave the details. COROLLARY 7.1.3. T h e functiong(t, u) = -C(u), C E X , is admissible in Theorem 7.1.2 to assure that the trivial solution of (7.1.1) is uniformly asymptotically stable. Proof. By Corollary 3.4.2, it follows that the trivial solution of the scalar differential equation (6.1.2) is uniformly asymptotically stable, and hence Theorem 7.1.2 guarantees the stated result. REMARK7.1.2. Notice that, in Theorems 7.1.1 and 7.1.2, we have asked that V(t,4) satisfies the condition
T h e same results can be proved, with minor modifications, even when we assume a weaker condition, namely,
This latter assumption is more convenient in applications. Observe, however, that in the next section, when we consider the converse problem of showing the existence of Lyapunov functions, we do obtain functionals verifying the former condition. T h e following theorem, which is parallel to Theorem 3.4.9, is of interest in itself. Its proof can be constructed on the basis of the proofs of Theorem 3.4.9 and the foregoing theorems in this section.
THEOREM 7.1.3. Assume that there exists a functional V ( t , $ ) satisfying the following properties:
(ii) o+V(t,~ i ( t o7 40))
< -C(ll
to
9
4o)(t)ll),
t 3 to , C E X .
Then, the trivial solution of (7.1.1) is uniformly asymptotically stable.
48
7
CHAPTER
Let us consider the scalar differential-difference equation
where
T, u
> 0 and b are constants. For each t > t n + T, we have X ( f -- T )
=z
0
x(t) -
x'(t
+ s) ds.
-7
Equation (7.1.8) can therefore be written as x'(t)
+ b ) x ( t ) - ab r"
= -(u
x(t
+ s) ds (7.1.9)
Let us take as a Lyapunov functional c.(t,+) = p ( 0 )
+a
4"s)
ds.
--7
Observe that V(t,+)fulfills the condition (7.1.7), where b(u) = u2 and (1 C Z T )u2. If ~ ( t, &)(t) , is the solution of (7.1.8) corresponding to a given initial function + o , then V(t,x l ( t 0 #,,)) , is a differentiable function of t , and, therefore, a simple calculation yields the inequality u(u) =
V'(4
+
Xdt"
Id"))
< -4.
-
I b I)[X2(t0 do)(r) I
+
X2(t0
9
do>(t
-
<
.>I.
I t is now easy to see from Theorem 7.1.3 that I b I a, I b I < a imply uniform stability and uniform asymptotic stability of the trivial solution of (7.1.8), respectively. It is observed that the particular Lyapunov functional just used has given a region of stability which is independent of the lag T and the sign of b. If, on the other hand, we take the functional
and assume that b 3 0, we see that this functional verifies condition (i) of Theorem 7.1.3. T o verify condition (ii), we compute V ( t ,x l ( t , , 4,)) using (7.1.9). After some calculations, we obtain the following relation:
b"(4 .,(to
,$0))
<
-TP1
+ b(1
J'"
[a(l
-7
- b.>{X2(t0
-
+
b.){x2(t0 9 do)(t) >
60)(t)
X2(t, >
+ x2(t0 $o)(t + 9
do>(t s
-
+ 4)
.)>I
ds.
7.2.
49
CONVERSE THEOREMS
<
It follows from Theorem 7.1.3 that 0 < b~ 1, 0 < b~ < 1 yield uniform stability and uniform asymptotic stability of the trivial solution of (7.1 A), respectively. T h e foregoing discussion shows that, by choosing a suitable Lyapunov functional, it is possible to get the information about the qualitative structure of the region of stability in different ways. Finally, we shall state a result whose proof is analogous to that of Theorem 3.4.10.
THEOREM 7.1.4. Assume that there exist a functional V ( t , + ) and a function g ( t , u ) enjoying the following properties: (i) V E C [ J x C, , R,],and, for t 3 t o , %(to ,40))
D’W
< g(t, V ( t ,
4
0
9
40)));
(ii) there exist functions a , b E X such that, for ( t , 4) E J x C, ,
w 9 110) 0
< w, 4) < 4 4
110);
(iii) g E C [ J x R, , R ] , and, for every pair of numbers 01, p such that < 01 p < p , there exist 0 = 0(a, p) 3 0, k = k(a, p) > 0 satisfying
<
g(4 4 ,< --k,
(iv) for a X
E
t 3 8,
,< u
< 8;
C [ J ,R + ] , f ( t ,4) satisfies the condition
Ilf(t, $111 < A(t) /I 4 110 ,
(4 4) E J x
c,
*
Then, the trivial solution of (7.1 .l) is uniformly asymptotically stable. 7.2. Converse theorems
As in the case of ordinary differential equations, we may define the notion of generalized exponential asymptotic stability. DEFINITION 7.2.1. T h e trivial solution of (7.1.1) is said to be (Sll) generalized exponentially asymptotically stable if
II %(to ,+o)llo
<
W O )
II 40 110 exP [P(tCJ - fJ(t)l,
where K ( t ) > 0 is continuous for t as t .+ Go.
E
J,p
EX
for t
t E
3 to
7
(7.2.1)
J , and p ( t ) -+
00
50
CHAPTER
7
T h e particular case when K ( t ) = K > 0, p ( t ) = at, 01 > 0 is referred to as the exponential asymptotic stability of the trivial solution of (7.1.1).
THEOREM 7.2.1. Assume that f ( t , +) is linear in 4 and that the trivial solution of the system (7.1.1) is generalized exponentially asymptotically stable. Suppose further that p ( t ) is continuously differentiable on J . Then, there exists a functional V ( t ,4)satisfying the following properties: (I") V E C[J x C, , R,], and V is Lipschitzian in 4 with the function K(t); (2") I/ ,$ I10 < V(t,$1 < K ( t ) II 4110 t 6 J , 4E c, ; (3") D+V(t,4) < -p'(t) v(t,+), t E J , 4 E C, . 9
Proof.
Consider the functional
Clearly, from (7.2.1), it follows that V ( t , $ )verifies the property (2"). Moreover, D+V(t,4 )
=
lim
SUP h-O+
h-"Qt
+ h,
Xt+h(4
4))
-
V(4 +)I
This proves (3"). Let us note that, in proving the foregoing inequality, we have used the uniqueness of solutions of (7.1.1).
7.2.
51
CONVERSE THEOREMS
Now let dI , d2 E CD. Then, using the fact that f (t,#) is linear in # and the inequality (7.2.1), we get
I Q,41)
-
w 42)I
=
I o>o S U P I1 Xt+o@, 41)lIo .XP{P(t - SUP 020
a>O
= SUP o>o
-
ll X t + o ( 4 ddlo .XP{P(t
< S U P I1 X t + u ( 4 41) II X t + o ( 4
< K ( t )I1 4,
- 4 2 110
P(t)>
- P(t>)l
42)llo exP{P(t
- Xt+o(i,
4 1 - 42)llo
+4 +4
+4
..P{P(t
+4
-
$(t>l
~
P(t>>
.
T h e continuity of V(t,#) may be proved as in Theorem 3.6.1, with minor changes. This completes the proof of the theorem. A similar result is true, even whenf(t, #) is nonlinear in +, provided f ( t ,#) satisfies a Lipschitz condition.
THEOREM 7.2.2.
Assume that the trivial solution of the system (7.1.1) is generalized exponentially asymptotically stable. Let the function p ( t ) occurring in (7.2.1) be continuously differentiable for t E J. Suppose that f ( t , #) verifies a Lipschitz condition in # with a constant L > 0. Let the function K ( t ) be bounded, and, for some q, 0 < q < 1, let there exist a T > 0 such that K(t)exp{-qp(t
+ T )-$(t))
< 1,
(7.2.3)
t€J.
Then, there exists a functional V(t,#) possessing the following properties: ( ") I/'
'[J
I JV, 41) -
'Po
7
R+], and>for
w, +,)I < eLT
SUP
0
J>4 1
.XP{P(t
7
42
+4
-
>
'Do
P ( t ) > II 41 - 4 2
/I d 110 < v(t,4) G K ( t ) It # 110 > t E J, # E Coo; (3") D+V(t,#) < -(l - q) p'(t) V ( t ,#), J,#
I10
;
(2") Proof.
'Do
'
Let q, T be given satisfying (7.2.3). Define
Since K ( t ) is assumed to be bounded, let po = p / M , where = supteJK(t). Then, it is clear that V E C[J x C o o R,]. , The relations (2") and (3") can be proved, following the proof of Theorem 7.2.1.
M
52
7
CHAPTER
T o show that V ( t ,+) satisfies the stated Lipschitz condition, notice that
'I
exp[U
+)llo
Xtta(4
-
4)Mt
+
:k ' ( t ) exP[-dP(t
+4
0) -
P(W P ( W II C I10 > -
and, because of (7.2.3), b7(t,+)
=
Consequently, for
I
f (t7
SUP
O o T
+,,
$2
II rt+a(t! 4)IIn e x ~ [ ( l- q ) { P ( t E
+
0)
-
Cooand t E J ,
91) v(t!(bdl ~
< oSUP II 9 t o r eLT
SUP
I
~(t)>l-
"(t>41) - X t + o ( 4
exp[(l
-
4 d O exp[(I
+
q ) { ~ ( t u)
-
-
4){P(t
+4
-
P(t))l
~ ( t ) ) I/l (61 - C z /In ,
on the basis of the estimate (6.2.5). This proves the stated result.
T h e next theorem is a result similar to Theorem 7.2.2, whose conditions and the arguments of proof are slightly different.
THEOREM 7.2.3. Assume that (i) for any two solutions x(0, $0), x(0, &) of (7.1 .I), the lower estimate
!I +o
-
$0 110
cxp [-Jlpl(s) ds]
i II Xt(07
40)
-
Xt(O,
+o)llo
t
2 0,
holds, where p , E C[/, R ] ; (ii) there exists a p E .K for t E J , p ( t ) 4 co as t + CO, $ ( t ) exists, and
11 .rt(O,4o)ll(,
K
11 4 0 110 e-"(t),
t
2 0,
K > 0,
+,,
where x(0, +(J is a solution of (7.1.1) with an initial function at t = 0; (iii) the system (7.1.1) is smooth enough to ensure uniqueness and continuous dependence of solutions. Then, there exists a functional V ( t ,+) satisfying the following properties:
(1 ') I/ t C[/ x C, , R,], V ( t ,4) is Lipschitzian in function K ( t ) >, 0;
<
(2Y II 4 110 v(t,4),< K ( t ) II 4 110 ; (3") D+V(t,+) = -p'(t) V ( t ,$).
+ for a continuous
7.2.
53
CONVERSE THEOREMS
Proof. Let us denote C$ = X~(O,C$~),so that, in view of assumption (iii), we have &, = xo(t,4).We now define the functional V ( t ,4)by V(4 4)= Ke-P‘t)I1 xo(4 9)lIo ’
Then, it is clear that VVEC [ J x C, , R+], and, because of condition (ii), there results
I/ d I10 G
V ( t >4).
I n view of uniqueness, we have, for small h V(t
+ h, xt+&(t,9))
=
> 0,
Ke-p(t+h) II xo(t
- Ke-p‘tfh)
+ h,
/I xo(4 4)llo
4))llo
~t+h(t, 9
and, consequently, it follows that D+V(t,4) = -p’(t)V(t, 4).
Letting
4 = x,(O, (b0), 1c, = ~
/I xo(4 4)- xo(t, $)I10
< I1 4
~ (lc,o), 0 ,we obtain the inequality -
$ /lo exp
[fp) 4,
t
3 07
because of uniqueness and assumption (i). Thus, we have
I V(t,4)- V(t,+)I
=
Ke-”‘t’I1 xo(4 4 ) - xoP, 4)llO
+
provided we define K ( t ) = K exp[-p(t) JApl(s) ds]. Finally, the upper estimate in (2”) results, by setting 1c, of the theorem is therefore complete.
=
0. T h e proof
<
p(t), in addition to the assumptions COROLLARY 7.2.1. If J:pl(s) ds of Theorem 7.2.3, then K ( t )is to be replaced by K > 0 in the conclusion of Theorem 7.2.3. THEOREM 7.2.4.
Assume that
(i) condition (i) of Theorem 7.2.3 holds; (ii) the system (7.1.1) is smooth enough to ensure the uniqueness and
54
CHAPTER
7
continuous dependence or solutions, and there exist functions A, , A, E 3? such that (7.2.5) t 3 0; &(I1 4 0 110) < 11 4 0 , +O)llO < X,(lI 4 0 IIO), (iii) g E C [ J x R , , R ] , and, for t E J , u1
2 u, ,
At, 4 - g(t7 .2) G PdtKu, where p , estimate
E
4,
~
(7.2.6)
C [ J ,R ] , and the solutions u(t, 0, uo) of (6.1.2) verify the
< u(t, 0, uo) ,< Y*(%),
Y&")
(7.2.7)
2 0,
t
for y l ,y z E X . Then, there exists a functional V ( t ,4) satisfying the following conditions:
(I") I.'E C [ / x C, , R,], and V ( t ,4)is Lipschitzian in tinuous function K ( t ) 0;
> (2")@I1 4 < q t , 4)< 4 14llo),
110) a, b E -x, ( t ,4)E (3") D+V(t,4)= g(t, q t , 4)),( t ,+) E J x c, .
4 for
a con-
J x c, ;
P u ~ o f . By condition (7.2.6), the uniqueness of solutions of (6.1.2) is assured. Let u(t, 0, u0)and x(0, 40)be the solutions of (6.1.2) and (7.1.1), respectively, satisfying the assumptions of the theorem. Denote 4 == x1(0,4,,), so that, by uniqueness, we have +o = xo(t,4).Let us now define V(t> 45) = u(t, 0, /I %(t, 4) 1'0 ).
I t is clear that V E C [ / x C, , R c-(t
i
11, .%+h(4
$1)
= u(t ~
u(t
Furthermore, for small h
> 0,
1- h, 0, I1 .Y"(t -t hr%+h(t, a,,ii0>
t h, 0, /I X"(4 +)ll").
Thus, for ( t ,4)E J x C,, , Dl
i ? ( t ,+)
~-
lim sup h-'[ I ' ( f h-O+
lim sup I/ ' [ ~ ( t t 12, 0, 11 r,,(t,+)llo I1-0
+
2 ~ ' ( 1 ,0, Ii -
d f$)lid ,
d r , 4 4 0, 'I % ( t , $)ll",) g(t,
proving (3").
+ h, x,+,[(t,+))
E
(f,
a,>,
['(t,+)]
~
~
~ ( t0,, 11 xo(f,d)ll,)]
7.2.
Since
+
=
55
CONVERSE THEOREMS
x,(O, b0) and do = xo(t,+), the relation (7.2.5) yields
A,,' A;' being the inverse functions of A, , A, , respectively. Hence, using this inequality and (7.2.7) successively, we obtain
and
Evidently a, b E X , and hence (2") is verified. Finally, for t E J , ,d2 E C , ,
I
w
$1)
-
v(4d2)l
=
I 4 4 0, /I xo(4 c1)lIo)
< /I xo(4 41) - xo(4
-4
d2)lIo
4 0, II xo(4 d2)llo)l
exp
[St
0
P2(4
4,
using the condition (7.2.6). Furthermore, as observed in the proof of Theorem 7.2.3, we have, as a consequence of assumption (i),
These considerations imply that
I v(441) - J+, where K ( t ) = exp[lh [p,(s)
42)l
< K ( t )I1
$1 - $2 110 Y
+ p,(s)] ds]. T h e proof is complete.
REMARK 7.2.1. We note that, since p2(t)need not be nonnegative, there is a possibility that K ( t ) may be bounded by a constant. On the basis of Theorem 7.2.4, it is possible to state and prove other converse theorems involving differential inequalities, parallel to certain theorems in Sect. 3.6. We shall only state two converse theorems with respect to uniform asymptotic stability.
56
CHAPTER
7
THEOREM 7.2.5. Assume that (i) the system (7.1.1) is smooth enough to ensure the uniqueness and continuous dependence of solutions; (4 P1 /I do I10 U l ( t - t o ) II X L t O 7 d0)lIo G P 2 /I do I10 0 2 ( t - to), t b t o 7 where P1 ,P2 > 0 are constants.
<
Then, there exists a functional V(t,4) verifying the following properties: (1 ") V E C [ / x C, , R,],and there exist two functions a, b that b(/ld 110) < q 4 4)< 4 9 Ilo), (4 4) E J x c, ;
<
E
X such
(2") d/dt [ q t ,%(to , +O))l -,V(t, %(to do)), t 3 t o . Proof. For some fixed T > 0, which we shall choose later, define 9
Since, by assumption (ii), we have
it follows that
(7.2.8)
Moreover, similar arguments yield
We have thus proved (lo). T o prove the validity of (2"), notice that
7.2.
57
CONVERSE THEOREMS
Hence,
Let us now fix T so that u2(T ) < (2Pz)p1. This choice is possible, since uz E 2. It then results that
This, together with (7.2.8), yields, setting (2~x-l = ~ B z u z 2 T ( 0, )
and proves the theorem.
THEOREM 7.2.6. Let the trivial solution of (7.1.1) be uniformly asymptotically stable. Suppose that
for ( t , +), ( t , 4) E J x C,, , where L(t) >, 0 is continuous on J and t+u
t
L(s)ds
< Ku,
u
2 0.
Then, there exists a functional V(t,+) with the following properties: (1") V E C[J x C, , R+],and V ( t ,+) satisfies
I for t
E
J , 4,4E
CS(S,)
< <
w, 4) ;
- V ( t ,#>I
< M II c
~
$110
?
+
(2") b(ll+ 110) V ( t ,4) d 4 Ilo), 4 b E 3? ; (3") D+V(t,4) -C[V(t, 411, c E 3Y. T h e proof of this theorem can be constructed parallel to Theorem
3.6.9 with essential changes. We leave the details.
58
CHAPTER
7
7.3. Autonomous systems
I n this section, we consider some stability and instability results for autonomous systems of the form x ' ( t ) =f ( 4 , t E
I,
(7.3.1)
where f~ C[C,, Rn] and f(+) is locally Lipschitzian in 4. It is quite natural to consider the system (7.3.1) as defining motions or paths in $9. I n fact, we can define a motion through 4 as the set of functions in %?n given by UteJxt(O,+), assuming that the solutions xt(O,+) exist on J. We shall, in what follows, abbreviate xi(O, +) by xi(+).
DEFINITION 7.3.1. An element $ E V nis said to be in the w-limit set of [-T, co) and there is a sequence of nonnegative real numbers {tn}, t , + co as n ---t co,such that
+, Q(+) if ~ ~ ( is4 defined ) for
DEFINITION 7.3.2. A set M C W Lis said to be an invariant set if, for any 4 in M , there exists a function x(+) depending on 4, defined on (-00, a), xi(+)E M for t E (- co, co), xo(+) = +, such that, if x*(u, xu) is the solution of (7.3.1) with the initial function xu at u, then x*(cr, xD) = xt(+) for all t 3 u. We notice that to any element of an invariant set there corresponds a solution that must be defined on (- co, a).
LEMMA 7.3.1. Let x(+) be a solution of the system (7.3.1) with an initial function at t = 0, defined on [ - T , a), and let
+
/I 4 d ) l l n
< pi < P
1-
t~
for
Then, the family of functions {xi(4))},t E J , belongs to a compact subset of Vn,that is, the motion through 4 belongs to a compact subset of en. T h e proof of this lemma follows from the fact that, for any p1 < p, there exists a constant L > 0 such that Ilf(+)ll L for all satisfying
I1
+
110
G
P1
<
.
+
LEMMA 7.3.2. Let 4 E C,, be such that the solution x(+) of the system (7.3.1) is defined on [ - T , co) and 11 xt(+)llo \< p1 < p, t E J. Then, the w-limit set Q(+) is nonempty, compact, connected invariant set, and
4x,(d), Q(4l
-
0
as
t
-
a.
(7.3.2)
7.3.
AUTONOMOUS SYSTEMS
59
Proof. By Lemma 7.3.1, the family of functions x,(+), t E J , belongs to a compact subset S C Vn,and, furthermore, S could be chosen to be the set of $ E V n such that 11 $)lo < p l , 11 $' (lo < k, for some constant k.
This proves that Q(+) is nonempty and bounded. If $ E Q(+), then there exists a sequence {t,}, t, -+ co as n -+ m, such that 11 x,n(+) - $ Ilo-+O as n-+ co. For any integer N , there exists a subsequence of {t,}, which we keep the same designation, and a function gm(+) defined for - N N N , such that 11 x,,+~(+) - gm(+)llo-+O as n -+ co uniformly for 01 E [ - N , N ] . By the diagonalization process, we -+ 0 as n -+ co uniformly can choose the t, so that 11 x~,+~(+)- gb(+)(/,, I n particular, the sequence {x,,+~(+)} on all compact subsets of (- 03, a). defines a function go,(+),for -a < N < co. It is easy to see that gm(+) satisfies (7.3.1). Since go(+) = $, it follows that the solution x,($) of (7.3.1) with initial value $ at t = 0 is defined for t E (-00, co) and, furthermore, is in Q(+), since
< <
for any fixed t. This shows that Q(+) is invariant. It is clear that Q(+) is connected. T o show that Q(+) is closed, suppose $, in Q(+) approaches $ as n-+ co. There exists an increasing sequence {t,} = {tn($,)} such that t, -+ co as n -+00, and 11 x,,(+) -$, 1l0 -+ 0 as n -+ co. Given any E > 0, choose n so large that
for sufficiently large n, which shows that $ ~ 5 2 ( + ) , and hence a(+)is closed. But, clearly, Q(+) C S, and, since S is compact, it follows that Q(+) is compact. T o prove (7.3.2), suppose that there is an increasing sequence {t,}, t, -+ 03 as n -+ co, and an a > 0 such that
Since x,%(+) belongs to a compact subset of Vn,there exists a subsequence that converges to an element $ in en,and thus $ is in Q(+). This is a contradiction to the foregoing inequality and completes the proof of the lemma.
60
CHAPTER
7
REMARK7.3.1. I n the proof of Lemma 7.3.2, we have only used the fact that xl(+) is continuous in t, and that xt(+) belongs to a compact subset of Vn.Therefore, the Lipschitz condition on f could have been p1 < p implies Ilf(+)ll L for some L. replaced by I/ /lo
+
+ <
THEOREM 7.3.1.
<
Let
(i) V E C[C, , R] and Q, = [+ E C, : V(+)< a ] ; (ii) there exist a constant K such that II+(O)ll K , V(+)3 0, and, for E Q ~ D+V(+) , 0; (iii) E be the set of all points in Qe, where D+V(+) = 0, and M be the largest invariant set in E.
+
<
<
Then, every solution of (7.3.1) with initial value in Q, approaches Mast4co.
Proof. T h e conditions on V imply that V(xl(+))is a nonincreasing function of t and bounded below within Qw . Hence, E Q, implies xt(+)E L?, and 11 x(+)(t)lI K for all t 3 0, which shows that I/ xt(+)ilo K for all t 3 0; that is, xt(#) is bounded, and Lemma 7.3.2 yields that a(+) is an invariant set. But V(xl(+))has a limit a. < 01, as t co and V = a,, on Q(+). Hence, Q(+) is in Qa and D+V(+) = 0 on Q(+). Consequently, the fact that Q(+) is invariant implies that Q(+) is in M , and, by Lemma 7.3.2, xt(+) tends to M as t -+ 00. This completes the proof. T h e conditions 11 +(O)il K and V 3 0 of the foregoing theorem may be replaced by the assumption that the region where V(+)< 01 is compact.
+
<
<
--j
<
COROLLARY 7.3.1. If the conditions of Theorem 7.3.1 are satisfied and D+V(+) < 0, for all # 0 in 8,, then every solution of (7.3.1) with
+
initial value in Q, approaches zero as t
THEOREM 7.3.2. Assume that
-+
CO.
< +
(i) V E C [ q r LR,] , and D+V(+) 0, E V n ; (ii) E is the set of all points in Wn for which D+V(+) = 0, and M is the largest invariant set in E.
+
Then, all solutions of (7-3.1), which are bounded for t 3 0, approach Mast+a. If, in addition, there exists a continuous, nonnegative function b(u) on the interval [0, a) such that b(u) 4GO as u 4 co and 4l4(0)ll)
< V(+), c E vn,
then all solutions of (7.3.1) are bounded for t 3 0.
1.3.
61
AUTONOMOUS SYSTEMS
Proof. T h e first part of the theorem proceeds essentially as in Theorem 7.3.1. For the second part, let E gn. Then, there is a constant N such that V(+)> V(+,), for II+(O)l/ 2 N . Since V ( X ~ ( +is~a) )nonincreasing function of t, it follows that 11 x(+,)(t)ll < N , for t >, 0, which implies II ~ l ( + O ) I l O < N , t 3 0.
+,,
COROLLARY 7.3.2. Iff (0) = 0, all the conditions of Theorem 7.3.2 are satisfied, and V(0) = 0, D+V(+) < 0 for # 0, then all solutions of (7.3.1) approach zero as t -+ CQ, and the origin is globally asymptotically stable.
+
We next give a theorem on instability of the trivial solution of (7.3.1).
THEOREM 7.3.3. Suppose that V(+) is a continuous, bounded scalar function on C,, and that there exists a y and an open set E in %?n such that the following conditions are satisfied: (i) V(+)> 0 on E, V(+)= 0 on that part of the boundary of E in CY ; (ii) 0 belongs to the closure of E n C, ; (iii) V(+) a(][+(O)ll), on E n C,, , where, a E A? ; (iv) D+V(+) = lim inf,,,, k1[V(xh(+)) - V(+)]>, 0, on the closure of E n C, and the set U of in the closure of E n C, such that D+V(+) = 0, contains no invariant set of (7.3.1) except = 0.
<
+
+
Under these conditions, the trivial solution of (7.3.1) is unstable, and the trajectory of each solution of (7.3.1), with initial value in E n C, , intersects C, at some finite time. Proof.
Suppose +o
E
E n C, . By hypothesis (iii),
I1 4o(O)Il b a - l ( W o ) > > and (iii) and (iv) imply that xi(+,) satisfies
I1 x(+o)(t)ll b a-'(J+d+o))) b u-Y W O > > P as long as xt(+,) E E n C,, . If xl(+,) leaves E n C,, , then it must cross the boundary aC,, of C,, . I n fact, it must cross either aE or aC, , but it cannot cross aE inside C, since V = 0 on that part of aE inside C, and V(x,(+,,)) 3 V(+,,)> 0, t >, 0. Now, suppose that xt(+,) never reaches aC, . Then, xi(+*) belongs to a compact subset of the closure of E n C,, , for t 3 0. Consequently, xl(+,) approaches Q(+,), the w-limit set of
+,,
62
CHAPTER
7
and .Q(dO) C closure of E n C,, . Since V(xl(b0))is nondecreasing and bounded above, it follows that V(xl(+,))+ B, a constant, as t + co,and, implies that Since 4 E thus, D+V(xl(+))= 0 for 4 E a(&).
/I $(O)ll
< .-'(V(40>> > 07
we have a contradiction to hypothesis (iv). Consequently, there is a t , > 0 such that jl x((bo)(tl)ll = y. Hypothesis (ii) implies instability, since do can be chosen arbitrarily close to zero. This completes the proof of the theorem. 7+4. Perturbed systems
We shall be interested, in this section, in the perturbed functional differential system (7.4.1) 4 9 = f ( t , 4 R(4 4,
+
where f , R E C [ J x C, , R"] and f ( t , +), R(t, (6) satisfy a Lipschitz condition in 4 for each t E J.
THEOREM 7.4.1. Suppose that the trivial solution of (7.1.1) is exponentially asymptotically stable andf(t, 4)is linear in 4. Assume further that
/I R(t?4111 < 711 c 110 >
t6
/t
4E c,
(7.4.2)
9
7 being a sufficiently small positive number. Then, the trivial solution of the perturbed system also enjoys the exponential asymptotic stability.
By Theorem 7.2.1, there exists a V ( t ,4) such that, for d) E J x C,
Proof. ( t 7
7
(i) I/ E C [ J x C, , R,], and V is Lipschitzian in K>O; ( 4 II d 110 v(t,4) K 114 110 ; (iii) D+V(t,(b) < --(xV(t,(b), a > 0.
<
4 with
a constant
<
Let y(t, ,40)be any solution of (7.4.1) such that J j 4"jJO 4t)
=
V t , YdtO
9
< p / 2 K . Define
do)).
Whenever 11 do /lo <
=
p/2,
m(tl) = p ,
and
m(t)
> p/2,
t E [t., tJ.
7.4. PERTURBED
63
SYSTEMS
We see from these relations that D+m(t,) >, 0.
(7.4.3)
On the other hand, it follows, from (ii), that IIYt(t0 ,do)llo
for
to
< t < tl
*
Let 4 = yt,(to,do), and, because of uniqueness, Ytz+h(to> do)
Thus,
D+"(t2) = lim SUP h-l[V(t, h-O+
= lim SUP h-l[V(t, h 4 +
+ Vt2 +k
= Ytz+h(t27
h 3 0.
41,
+ h, Ytz+h(to,do>> + h, Yt,+h(h
V ( t , Yt,(to > do))]
-
-
I d > )
!
V(t2
+ h,
,4 4 - V t 2 +)I,
Xt,+h(f,
9
Xt,+h@,
,4))
+
where ~ ( t,4) , is the solution of (7.1.1) with an initial function at t = t, . Using the Lipschitzian character of V , the assumption (7.4.2), and the preceding relation, we obtain (7.4.4)
D+m(t2) G K.111d 110 - ""(t2).
Since q > 0 is sufficiently small, there exists a y > 0 such that Kq < a - y , and, hence, the fact that 11 (I, V ( t ,d), together with the inequality (7.4.4), implies that
<
+ K.11
< &!)[-a < -Ym(tz)
Df"(t2)
< 0,
since m ( t 2 )= p / 2 > 0. This contradicts (7.4.3) and proves that m(t) < p , t 3 to . Consequently, by ii, it follows that
II YdtO > do)llo <
P2
whenever 11 c $ ~ (lo < p/2K. Thus, setting before, we arrive at the inequality
D+v(4Y d t o
9
do))
3 to
t
4
9
and arguing as
= y l ( t o ,+o)
< -Y w, Yt(fo do)), >
and therefore, by Theorem 1.4.1,
w, Y&O
>
do))
<w
o
9
do) exP[- Y ( t
-
toll,
t
2 to *
64
7
CHAPTER
Because of ii, from this inequality results a further inequality
proving the stated result.
THEOREM 7.4.2. Assume that the trivial solution of (7.1.1) is generalized exponentially asymptotically stable and thatf(t, 4) is linear in 4. Suppose that g E C [ J x R, , R,] and g(t, 0) = 0, g ( t , u) is nondecreasing in u for each t E J , and
/I R ( t , +)I1
< At, I1 d Ilo),
t
E
4E c, -
f,
(7.4.5)
If y(to , +o) be the solution of (7.4.1) such that uo = K(t,)j/#JO /IO
I1YdtO > 4 O ) l l O G
r ( t , to
>
UO),
t
3 to
, then (7.4.6)
9
where r ( t , t o , uo) is the maximal solution of 24’
existing for t ug
< p.
=
-p’(t).
+ K(t)g(t,
U),
3 to and satisfying
u(t,) = U o
r(t, t o ,uO)< p, t
,
(7.4.7)
to , whenever
Proof. Lety(to ,#Jo) be any solution of (7.4.1) such that /I & ( ( o Setting #J = yt(t, , #Jo), we have Yt+h(tO >
d o ) = Yt+*(t,
d),
h
< p/K(to).
2 0,
because of uniqueness of solutions. Suppose now that ~ ~ + #J), ~ (h t2, 0, is the solution of (7.1.1) through ( t ,4). If 11 yl(to, $0)110 < p, t 3 to , we should have
Thus, since #J
= yl(t,
,#JO), it follows that
7.4.
65
PERTURBED SYSTEMS
By Theorem 1.4.1, it now results that
V t ,Y t ( t 0 4 0 ) ) < r ( t , t o uo), t b t o choosing uo = K(to)/I$, I(o , r(t, to , uo) being the maximal solution of 7
7
9
(7.4.7). Furthermore, from (ii) in the proof of Theorem 7.4.1,
II Yt(t0 > 4o)llo
e v, Yt(t0
9
$0)).
Also, by assumption, r(t, t o ,uo) < p, whenever uo < p. Our choices II do I10 < p / K ( t o ) and uo = W t o ) II $0 I10 imply that a0 < P. Hence,
I/ Yt(t0
9
4o)llo
< P,
t
2 to .
Thus, the estimate (7.4.6) holds.
THEOREM 7.4.3.
Under the assumptions of Theorem 7.4.2, the stability properties of the trivial solution of the scalar differential equation (7.4.7) imply the corresponding stability properties of the trivial solution of the perturbed system (7.4.1). T h e proof of this theorem is immediate from the relation (7.4.6). However, the following special cases are of importance. COROLLARY 7.4.1. T h e function g ( t , u ) = h ( t ) u, where X E C [ J ,R,], is admissible in Theorem 7.4.3, provided there exists a continuous function q(t) > 0, t E J , such that exp
[p(ro)- p ( t >
+
Jt to
W) ~
s ds] )
< q(to),
t
> to.
(7.4.8)
Proof. It is enough to show that, under the assumptions of the corollary, the trivial solution u = 0 of (7.4.7) is equistabIe. For, the general solution u(t, t o , uo)of u' = -p'(t)u
+ K(t)X(t)u,
is given by u(t, t o , uo) = uo exp
[p(to)- p ( t )
u(to) = uo ,
+ Jt W W ds] to
9
t
2 to
9
and, hence, equistability follows from (7.4.8). COROLLARY 7.4.2. T h e functions g(t, a ) = h(t)u, h E C [ J , R,], p ( t ) = cut, 01 > 0, and K ( t ) = K > 0 are admissible in Theorem 7.4.3, provided (7.4.9) lim sup ( t - to)-l ~ ( s )dsl < c u j ~ t-w
[
jt
to
66
CHAPTER
Proof.
7
I n this case, the general solution u(t, to , uo) of (7.4.7) is given by
and, therefore, the condition (7.4.9) shows that the trivial solution of (7.4.7) is uniformly asymptotically stable. Hence, by Theorem 7.4.3, the trivial solution of (7.4.1) is also uniformly asymptotically stable. T h e foregoing results can be extended to the case when f ( t , $ ) is nonlinear, on the basis of Theorem 7.2.2. Furthermore, as in the case of ordinary differential equations, we can show that, if the trivial solution of the unperturbed system is uniformly asymptotically stable, then it has certain stability properties under different classes of perturbations. For example, the concepts of total stability may be formulated parallel to the definitions of Sect. 3.8, and corresponding results may be proved. Likewise, boundedness, Lagrange stability, integral stability, and partial stability can be discussed. We shall omit such results as exercises to the reader. I n the following sections, we shall only concentrate on extreme and perfect stability criteria and existence of almost periodic solutions.
7.5. Extreme stability Associated with the system (7.1.1), let us consider the product system x’ == f ( t ,
4
(7.5.1)
Y’= f ( 4 Yt).
7
DEFINITION 7.5.1. T h e system (7.1.1) is said to be extremely unqormly stable if, for every E > 0, to E J , there exists a S(E) > 0 such that implies
I/ do - $0
I10
<S(4
I1 XdtO do) - Y4tO > $o)llo < 7
€9
t 3
t0
;
extremely quasi-uniform asymptotically stable if, for every and to E J , there exists a T ( E7) , > 0 such that
implies
I/ 4 0 II X 4 t O , 4 0 1 - YdtO
9
- $0
#o)llo
/lo
<
€9
< ?1 t 3 to
E
> 0, 7) > 0,
+ w,7).
If the preceding two concepts hold simultaneously for $o , +o E 5 P , we shall say that the system (7.1.1) is extremely unqormly completely stable.
7.5.
67
EXTREME STABILITY
T h e following theorem provides necessary conditions for the system
(7.1 -1) to be extremely uniformly completely stable.
THEOREM 7.5.1. Assume that (i) f~ C [ J x %%,R"],and, for every llf(4
01
> 0, if II 4 1l0 <
< L(4114
4) - f ( 4 #)I1
-
*
110
<
II $ 110
01,
01,
;
(ii) the solutions of (7.1 . l ) are uniformly bounded; (iii) the system (7.1.1) is extremely uniformly completely stable. Then, there exists a functional V(t,4, $) satisfying the following conditions:
(1") V E C[J x
I V t ,dl ,$1)
x gfl,R+],and V(t,4, $) satisfies
%n -
V t ,42
< W?"I 4 1
42)l
9
- $2 110
+ II
$1 - *z 1101,
for di, lJli E C , , (i = 1,2), t 3 0, where M ( q ) is a positive continuous function; (2") there exist functions a, b E X such that
4d(3") D'V(4
4, + ) ( 7 . 5 . 1 )
=
*
< w #>4 < 414
110)
+ h,
lim sup h - l [ V h-Of
Xt+h(4
-
+
110);
41,Y t + h ( 4 *)
W d ,41
-
< -V(4 4, $1. Let 7 be an arbitrary nonnegative number. Consider the case C,, . Then, corresponding to each E > 0, to E J , there exists a T ( E9) , > 0 such that, if t >, to T ( E.I,), then
Proof.
do,$o
E
+
II %(to do) 9
-
We assume that, if E > 1, T(E,q) ness of solutions, it follows that
II %(to for all t
7
d0)llO
Yt(4l =
t
< E.
*o)llo
T(1,q). By the uniform bounded-
II Yt@O
?
~0)Ilo
< Y(71)
2 to . Furthermore, if 11 4 \lo < y(q), 11 $ /lo < y(v), Ilf(4 4)- f ( 4 9911 < L(Y(?))ll d - 9 110
we have
We can assume that T (Eq), , y(y), and L ( y ( 7 ) ) are continuous. Moreover, if we consider a positive function A(€,7) for E > 0 and 7 3 0, such that
4% 71) = exP[{w4?))
+ 11 %, ?)I + 2 Y ( d exp[%
d
1
9
(7.54
68
CHAPTER
7
there exist two continuous functions g ( c ) and M ( 7 ) such that for E > 0, g(0) = 0, M ( 7 ) > 0, and
d.1
A(%7) G
g(E)
Wd-
>0
(7.5.3)
We now define V J t ,4,$) as follows:
vdt, 4,$1 = g ( W 0SUP Gk(ll X t + o ( 4 4)- Yt+& 30 for k
=
(7.5.4)
$)llo)eo,
1 , 2, 3,..., where u Gk(U)
u 3 K-1,
- k-1,
0
10,
(7.5.5)
K-1.
I t is easy to see that
by (7.5.3). Moreover, fort& , PoE %%' , there exists a p
I1 X t ( h 4,) 9
-
Ydt,
>
c0)Ilo
< B(Ild0
- $0
Ilo),
EX
t >to
such that I
because the system (7.1.1) is extremely uniformly stable. We thus have
c, $1 < g ( W Gk(B(II c exp[W-l, < P(ll c $ M(lld - $
V7C(t,
- $110))
-
110)
+,
II d
~
$ Il0)l
(7.5.9)
110).
We shall next show that Vk(t, $) satisfies a Lipschitz condition with respect to 4 and $. For any 4, , $, E C, ,
7.5.
EXTREME STABILITY
69
By (7.5.10) and the continuity of x t ( t 0 , c $ ~ in ) t , it follows that V k ( t ,4, $) is continuous in ( t , 4,$). Let us now consider D+V,(t, 4,$) with respect to the product system (7.5.1). Since
T h e desired Lyapunov functional may now be defined by
I n view of (7.5.8), this V ( t ,4,$) can be defined for all t E J , 4,$ E P. From (7.5.7) and (7.5.91, it is easy to see that there exist functions a, b E 3-satisfying (2"). Furthermore, because of the inequality
we obtain, for 4,41 , #, t,bl E C, ,
which proves that V ( t ,4,$) satisfies the Lipschitz condition as described in (1'). Finally, we shall show that (3")is satisfied. By the definition,
70
CHAPTER
7
if h is small enough,
<
D+r(t, 9,
$)(7.5.1)
N (2k))-1
li
k=l
-
vk(t,
9,
and therefore, by (7.5.1 I),
As N is arbitrary, this implies
T h e proof is complete. Sufficient conditions for the extreme complete stability of the system (7.1.1) are given by the following result.
THEOREM 7.5.2.
Assume that
(i) j ' C~[ J x q n , R n ] ,and, for every +o E Vn,the solutions x ( t 0 ,$o) exist in the future; (ii) V E C [ J x V n x Vn,R,],and, for t 2 to , D'Vt,
XdtO
t
(iii) g E C [ J x R,
do), YdtO 7
40))
< g(t, w ,%(to
, R], and g ( t , 0) = 0;
7
do), yt(t0 ,$0)));
7.5.
71
EXTREME STABILITY
(iv) there exist functions a, b E X on the interval [0, co) such that, for t 0, +, E Vn,
+
w, 4, 1cI) < 411d
b(lld - 9 110) G
-
*
110).
Then, if the trivial solution of (6.1.2) is uniformly completely stable, the system (7.1.1) is extremely uniformly completely stable.
Proof. Assume that the trivial solution of (6.1.2) is uniformly completely stable. This means that (S,*) and (S,*) hold at the same time. If E > 0, to E J are given, there exists a 6 = 6 ( ~ > ) 0 such that uo 6 implies
<
u(t, t o , uo)
<
w,
3 to.
t
Choose 6, = ~ ' ( 8 ) and uo = V(t,, + o , +,). yields, because of conditions (i) and (ii),
q t , %(to > do), Ydt ,
9
$0))
< r(t, t o
?
Then, Theorem 1.4.1 t b to
uo),
9
where r(t, to , uo) is the maximal solution of (6.1.2). We therefore obtain b(ll %(to
do)
?
YdtO *o)llo)
< b(4,
Yt@O ? 1cIo)llo
< E?
9
~
t
3 to
9
which implies that
I1 %(to
?
40)
~
3 to
t
3
<
provided 11 +o - +o \lo 6,. This proves extreme uniform stability of the system (7.1.1). a. Let now 01 > 0, E > 0, and to E J be given. Suppose I/ 4, - +o 1l 0 Let 01, = a(a). Since (&*) holds, given 01, > 0, b ( ~ > ) 0, and to E J , there exists a positive number T = T(to, 01, E) such that, if uo a , ,
<
u(t, t o , uo)
< bk),
t
t to
<
+ T.
As previously, it results that
4
%(to
do)
9
~
Y4tO > 1cIo)llo)
and this shows that, whenever
I1 %(to
7
90) -
< b(+
/I +, - +, 1l 0
YdtO ,+o)llo
< E?
t 3 to
<
01,
t
+ T,
we have
3 to + T.
I t therefore follows that the system (7.1.1) is extremely uniformly completely stable. COROLLARY 7.5.1. in Theorem 7.5.2.
T h e function g(t, u) = -01u,
01
> 0,
is admissible
12
7
CHAPTER
7.6. Almost periodic systems We shall continue to consider the functional differential system (7.I. 1). For the purpose of this section, however, we take f €
C[(-Co,
03)
x
c,, R"].
All the results that follow are extensions of the results of Sect. 3.18 to functional differential systems.
DEFINITION 7.6.1. A functional f E C[(-co, CO) x C, , Rn]is said to be almost periodic in t unqormly with respect to q5 E S for any compact set S C C, if, given any q > 0, it is possible to find an Z(q) such that, in any interval of length [(q),there is a T such that the inequality
Ilf(t + 4 ) -f(4 4111 < 77 is satisfied for t E (- GO, co), q5 E S. 7 7
We shall first prove a uniqueness result.
THEOREM 7.6.1.
Assume that
(i) f~ C[(-co, co) x C,, , Rn] and f ( t , q5) is almost periodic in t uniformly with respect to E S, S being any compact set in C, ;
+
(ii) I/ E C [ J x C, x C, , R,], V(t, q5, q5) Lipschitz condition in q5, $ for a constant M b(l/ 4
-
# 110)
< q t , 4, $1,
= 0, V ( t ,q5, $) satisfies = M ( p ) > 0, and
a
b E .f;
(iii) g E C [ J x R, , R],g(t, 0) = 0, and, for t 3 0,
n+V t ,4 , 4)< g(4 V ( t ,4, #I), where D+V(t, +, 9) is defined with respect to the product system x' = f ( t , X t ) , y' = f ( t , Y t ) ; (iv) the maximal solution of (6.1.2), through the point (T,,, 0), T,, 0, is identically zero. Then, there exists a unique solution of the almost periodic system (7.1 . l ) , to the right of t,, E (- co, a).
Proof. Sincef(t, +) is continuous, there exists at least one solution for a given to E (- co, co) and a q5,, E C, . Suppose that, for some to E (- co,co) t C, , there exist two solutions ~ ( t, , and and y(t,, ,+,) of (7.1.1). Then, at some t, > t o , we should have
II %,(to
* 40)
-
Yt,(t,
>
4o)lIo
=
€9
(7.6.1)
7.6.
where we may assume E < p. For to constant p1 < p such that
II -%(to
9
73
ALMOST PERIODIC SYSTEMS
d0)llo
< P1
9
< t < t, , there exists a positive I/ Yt@o
7
4o)llo
< P1
*
These solutions are uniformly continuous functions and bounded by p1 on the interval to - T t t , , and, hence, there exists a compact set S C C, such that
< <
%(to
6
9
do) E s,
Yt(4J,do) E s
for
t E [ t o , tll.
By Lemma 1.3.1, given b ( ~ ) /and 2 a compact set 6(~> ) 0 such that
, TI, there is a
[ T ~
=
r ( t , 7 0 , 0,s)
<w / 2 ,
t
E [To
(7.6.2)
, TI,
where r(t, T~ , 0, 6) is the maximal solution of U' = g(t, U)
+ 6,
.(To)
=
0.
Let 8 be a 6/2M translation number for f ( t , 4) such that to + 8 3 0, that is, (7.6.3) Ilf(t 8,d) -f(4 4111 < V M , t E (-00, a),
+
provided
belongs to a compact set S C C, . We consider the function m(t) = v t
Then,
+
+ 8, %(to
+
f
do), Y&o > do)),
t
E
[ t o 7 tll.
where x * ( t 8, xt), y*(t 8, y t ) are the solutions of (7.1.1) such that 4++o(t 8, x t ) = X f = X d t , > 4 0 ) and YTfdt 8, rt) = Y t = Yt(& , 4 0 ) , respectively. We thus have
+
+
74
CHAPTER
because of (7.6.3). Defining inequality m(t)
T,,
=
< r(t + 6,
to
70
7
+ 8, we get, by Theorem 1.4.1, the
, 0, a),
t E [ t o , tll, (7.6.4)
Assumption (ii) and (7.6.1), on the other hand, lead to 4tl)
2 w,
contradicting (7.6.4). It therefore follows that there is a unique solution for the system (7.1.1) to the right of to . COROLLARY 7.6.1. 7.6.1.
T h e function g(t, u) = 0 is admissible in Theorem
DEFINITION7.6.2. If, for any p that l l f ( t , #)]I < M(p), whenever bounded.
> 0, there exists an M ( p ) > 0 such # E C, , we shall say that f ( t , 4) is
The next theorem gives the sufficient conditions for perfect stability criteria of the trivial solution of (7.1.1).
THEOREM 7.6.2. L
Suppose that
(i) V E C [ J x C, , R,],V ( t ,#) is Lipschitzian in > 0, and, for ( t ,#) E J x C, ,
# for a constant
= L(p)
b(ll9 110)
< q t , 41,
(ii) g E C [ J x R,, R],g(t,0)
b E .f;
(7.6.5)
= 0, and, for ( t ,4) E J x C, ,
D+V(t>9)(7.1.1) G A4 V ( t ,4));
(iii) f E C [ ( - CO, CO) x C, , R"],f ( t , 0) = 0, f ( t , #) is bounded, and 4)is almost periodic in t uniformly with respect to 4 E S, S being any compact set in C, .
f(t,
Then, the null solution of (7.1.1) is (I") perfectly equistable if the trivial solution of (6.1.2) is strongly equistable; (2") perfectly uniform stable if the trivial solution of (6.1.2) is strongly uniform stable;
7.6.
75
ALMOST PERIODIC SYSTEMS
(3") perfectly equi-asymptotically stable if the trivial solution of (6.1.2) is strongly equi-asymptotically stable; (4") perfectly u n i f o r d y asymptotically stable if the trivial solution of (6.1.2) is strongly uniformly asymptotically stable. We shall prove only the statement corresponding to (4"). Let us suppose that the trivial solution of (6.1.2) is strongly uniformly asymptotically stable. Let 0 i E < p and to E (- 03, 03) be given. Then, given b ( ~> ) 0, T,,E J , and any compact interval K = [T,,, t*], there exist an 7 = q ( ~> ) 0 and a 6 = S(T,, E ) > 0 such that Proof.
u(t, 70 uo I 7)
< b(t-),
t s[To
7
t*l,
(7.6.6)
whenever u,,< 6, where u(t, T,,,u o , q) is any solution of (3.18.6). Choose LS, = 6 and u,, = L 11 +o I/, , L being the Lipschitz constant for V(t,$). Consider a solution x(t0 ,#,,) of (7.1.1) such that to E (- 03, a), I/ +o I/, < 6, . Suppose that, at some t , we have
/I X d t , > 9o)llo
=
6.
Then, there exist t , and t, such that to < t , II x&o , +o)llo = and that
< t , , (1 x t , ( t , , +,,)\lo
=
a,,
€9
81
< II %(to 9o)llo < 6, 7
tl
< t < t2
*
Clearly, there exists a compact set S C C, such that xt(t, , +,) E S for t t, . Let 0 be an q/L translation number off ( t , +) for E S such that to 0 3 0, that is, to
< < +
Consider the function
Then we have
+
76
CHAPTER
where xl*,,(t
x *( t
+ 8, x l ( t , , &))
is a
7
(7.1.1) such
solution of
+ 0, x t ( t , , do)) = xt(t, , do).It, therefore, follows that
Since x , ( t o , +,J E S for to D+m(t)
that
< t < t, , we obtain, using
< g(t + 8, m ( t ) ) + 7,
t
E
[to,
bl,
and hence, by Theorem 1.4.1,
< r(t + 8, *, 4,
m(t>
letting T~ : t, of (3.18.6). At t
+ 0, =
h(6) s; q t ,
70,
t
9
E
+
(7.6.8)
[ t o , 421,
where r(t 0, T ~ a,, , 7) is the maximal solution t, , we are led to an absurdity
+ 8, %,(to
7
40))
< r ( t , + 8,
70
, *o , 7 )
< 44,
in view of relations (7.6.5), (7.6.6), and (7.6.8). Thus, the perfect uniform stability of the trivial solution of (7.1.1) is proved. By assumption, given b ( c ) > 0, T , >, 0, there exist positive numbers S o , 7 = T ( E ) and T = T ( E )such that u(t, 7 0 , * o , ~ )< b(c),
3 70
t
+ T,
(7.6.9)
whenever uo 6 6 , . Choose u, = L (1 4, )lo and L8, = 8, , and let 6,* min[S,, So], where 8, = 6(p). Suppose now that x(t, ,4,)is any solution of (7.1.1) such that 11 4,1, So*, t, E (-a, a). Since [ / x l ( t , ,+o) l o ,< p for all t 2 to a n d f ( t , 4)is bounded and consequently ~ ' ( t , +,)(t) , is bounded by some constant, there exists a compact set S such that xt(t,, E S for all t 3 t, . As before, let 6 be an v/l-translation number of f ( t ,+) so that (7.6.7) is satisfied. Considering the function
<
:
m(t) = F(t
+ 8, ",(to ,$J~))
for
t
3 to ,
it is easy to obtain, as previously, m(t)
-= 1-(t + 8,
70,
*o
, 7),
t
2 to *
This, together with (7.6.5) and (7.6.9), yields
411%(to ,4o)llo)
< m ( t ) G r(t + 8,
70
, *o
,v) < b ( 4 ,
7.6.
for t
77
ALMOST PERIODIC SYSTEMS
+ 6 3 + T , and, consequently, T~
<
whenever 11 +o So*. This completely establishes (4"). COROLLARY 7.6.2. T h e function g(t, u) = -mu, N > 0, is admissible in Theorem 7.6.2 to yield perfect uniform asymptotic stability of the trivial solution of (7.1.1). As remarked in Sect. 3.18, if the functional f ( t ,+) is not almost periodic and f E C [ J x C, ,R"],then, from the strong stability properties of the trivial solution of (6.1.2), we may deduce strong stability properties of the trivial solution of (7.1. l), on the basis of Theorem 7.6.2. Finally, the following theorem assures the existence of an almost periodic solution.
THEOREM 7.6.3.
Suppose that
(i) V E C [ J x C, x C, , R,],V ( t ,+, +) is Lipschitzian in for a constant L = L ( p ) > 0, and, for t E J , +, E C, ,
+
b(II 4 - 1cI IIo)
< v(t,4,$) < .(I1 4
-
(ii) g E C [ J x R, , R ] ,and, for t E J , +, t,h
D+q4 4,$1
a , b E S;
$ llo), E
+ and + (7.6.10)
C, ,
< g(t, U t ,4,$1);
(iii) f E C [ ( - co, 00) x C, , R n ] , f ( t +) , is bounded, almost periodic in t uniformly with respect to E S, S being any compact subset in C, , and f ( t ,+) is smooth enough to ensure the existence and uniqueness of solutions of (7.1.1); (iv) for any b(c) > 0, CY > 0, and 5, E I , there exist positive numbers 7 = q ( c ) , T = T ( E CY) , such that, if uo CY and 5 3 5, T,
+
<
4 5 , 50
9
+
(7.6.11)
uo ,.I)< b(c),
where u( 5 , to, uo , 7)is any solution of u' = g(5, ).
+ q,
450)
=%,
5, 2 0;
(7.6.1 2)
(v) there exists a solution x(to, +,J of (7.1.1) such that
II %(to
,40)llo
< B < P,
t
2 to >
to E (-a, 00).
Then, (7.1.1) admits a bounded almost periodic solution, with a bound B.
78
CHAPTER
7
Proof. T h e proof runs, naturally, parallel to the proof of Theorem 3.18.5. Hence, we shall only indicate necessary changes. Let %(to, + o ) be the solution of (7.1.1) such that 11 xt(t,, , 40)ilo B. Since f ( t , 4) is assumed to be bounded, we have, consequently, that (1 %'(to,40)(t)\lis bounded for a constant B, , for t 3 t o . Let S be the compact subset of C, consisting of functions that are bounded by B and are Lipschitzian for a constant B, . Lct { T ~ be } any sequence such that rk -+ co as k -+ a3 and
<
uniformly for t E (- co, co), 4E S. Let /3 be any number, and let U be any compact subset of [p, co). Let 0 < E < p, and choose 01 = a(2B). Then, let 7 and T be the numbers defined in assumption (iv), for this choice. Let k, = k,(P) be the smallest value of R such that
Choose an integer n,
=
llf(t
+
no(€,p) Tkl >
2 k, so large that, for k2 3 k, 2 no ,
4) - f ( t
+
T/c2>
4111 < 7 / 3 4
(7.6.13)
for all t E (- GO, co), q5 E S. Let 8 be an y/3Ltranslation number for f ( t , 6) such that t, 8 >, 0, that is,
+
a), 4 E S. for t E (-a, Consider the function, for t
where t ,
=
where x*(t
t
+
3 to,
and x t = xt(to, +o). Then,
T~~ - T ~ ,
+ 8, xt),y*(t + 0, xt,) are the solutions of (7.1.1) such that
7.6.
79
ALMOST PERIODIC SYSTEMS
+
+ 6, x t ) = x t ,y&(t 6, xt,) = x t , , respectively. Thus, in view of the Lipschitzian character of V ( t ,4, $) and assumption (ii), we get
x&(t
o+m(t) < g ( t + e, fl(t>>+
+ /I
lim sup h4’
< g(t + 0, 4 t ) ) + L[ll %’(to
+ II
X’(t0
9
+ 6,
< g(t + 8, d t > + L[llf(t, Tkz
+
T7cl
Xt,)(t
-f(4
+ e,
Xt)l10
+ 6, %)(t + 6)ll
+ e)lll
+ e,xt)ll + Ilf(4
-f(t
XtJ
?
do)(t>- x*’(t
7
+o)(G) - Y*’(t
+ llf(t + - f ( t + 6, %,)Il .
X t + h - X?+-ts+h(t
+ 6, ~t,>iioI
- Y?+e+h(t
Xtl+h
h-’[l l
XtJ
Xt,)
Since t rkl 2 to + T , for t E U , we obtain, using the relations (7.6.13) and (7.6.14), D+m(t
+ < g(t + + 0, m(t + Tkl
.k,>
TkJ)
+ 17,
which implies, by Theorem 1.4.1, if uo = m(to), m(t
+
< r(t +
Tkl)
Tkl
+ 6, + 6, uo, 4, to
where r(4, t o ,u,, ,7)is the maximal solution of (7.6.12). By assumption (iv), it follows that r(6, t o ,
But,
for
all
UO,
t E U, t
5 = t + Tk, + 6, to= to m(t
17)
+
< &)
> to + + T .
+ 0, we get +
t 3 t o + T.
if
T ~ ,
Tkl)
Consequently, for all t E U , k,
II Xt+.rkl
Hence,
T
< b(€),
t
E
identifying
u.
> k, 3 no , we have, in view of (7.6.10), - X t + T k Z /lo
< E,
which, in turn, leads to the inequality
II 4%> do)(t
+
T k J ~X(t0
9
+o>(t
+ %z)lI <
t
€9
u.
This proves the existence of a function w ( t ) defined on [/3, co) and bounded by B. Since is arbitrary, w(t) is defined for t E (-00, a), and we have x(t0 ,do)(t
+
7Rl)
-~
-
( t )0
uniformly on all compact subsets of (-
00,
as
a).
4
-
m,
80
CHAPTER
7
Following closely the rest of the proof of Theorem 3.18.5, we can show that w ( t ) satisfies (7.1 . l ) and is almost periodic. This completes the proof. COROLLARY 7.6.3. If, in addition to the hypothesis of Theorem 7.6.3, the trivial solution of (6. I .2) is strongly uniformly asymptotically stable, then the system (7.1.1) admits an almost periodic solution that is perfectly uniformly asymptotically stable. I n particular, g(t, u ) = --NU, 01 > 0, is admissible.
7.7. Notes T h e results of Sect. 7.1 are adapted from the work of Driver [3]. See also Halanay [22] and Krasovskii [5]. Theorem 7.1.4 is new. Theorems 7.2.1 and 7.2.2 are taken from Hale [l]. See also Yoshizawa [3]. Theorems 7.2.3 and 7.2.4 are new. Theorems 7.2.5 and 7.2.6 are based on Halanay [221T h e results on autonomous systems in Sect. 7.3 are taken from the work of Hale [S], which may also be referred to for a number of illustrative examples. For the results on perturbed systems of Sect. 7.4, see Corduneanu [2], Halanay [22], and Hale [l]. Theorem 7.5.1 is due to Yoshizawa [I], whereas Theorem 7.5.2 is new. Section 7.6 contains the work of Lakshmikantham and Leela [3]. See also Hale [6] and Yoshizawa [2, 31. For closely related results, see Driver [3], Halanay [22], Hale [5], J. Kato [l], Krasovskii [5], Lakshmikantham and Leela [2], Liberman [I], Miller [l], Razumikhin [2, 61, Reklishkii [l-51, Seifert [I], Sugiyama [S], and Yoshizawa [3].
Chapter 8
8.0. Introduction I n what follows, we wish to treat the solutions of the functional differential system (7.1.1) as elements of euclidean space for all future time except at the initial moment. Our main tool, in this chapter, is therefore a Lyapunov function instead of a functional. T h e derivative of a Lyapunov function with respect to the functional differential system will be a functional, which may be estimated either by means of a function or a functional. While estimating the derivative of the Lyapunov function in terms of a function, a basic question is to select a minimal class of functions for which this can be done. Thus, by using the theory of ordinary differential inequalities and choosing the minimal sets of functions suitably, several results are obtained. If, on the other hand, the estimation of the derivative of the Lyapunov function by means of a functional is considered, the selection of a minimal set of functions is unnecessary. Nevertheless, this technique crucially depends on the notion of maximal solution for functional differential equations and the theory of functional differential inequalities. This method also offers a unified approach, analogous to the use of general comparison principle in ordinary differential equations. Moreover, it is important to note that the knowledge of solutions is not demanded in either case.
8.1. Basic comparison theorems
+
Let V E C[[-T, co) x S o , R+], and let E C,, . We define D+V(t,+(O), +), D-V(t, #(O), +) with respect to the functional differential system (7.1.1) as follows:
+ h, 4(0) + w, 4)) lim inf W V ( t + h, 4(0) + hf(t, 4))
D+V(t,4(0),4) = lim SUP h - V ( t
D-V(t, +(0),4) =
~
h-O+
-
h-0-
81
V ( 44(0))1, V ( t ,d(0))l.
(8.1.1)
82
8
CHAPTER
We need, subsequently, the following subsets of qn,defined by Q, =
[+E c, : I
Qo = [+ E
and Q,
=
v,lo = w, +(ON,
t E J1,
(8.1.2)
c, : v(t + s, W)) < W V ,+(O))),
[d E c, : I VtA, lo =
where A(t) > 0 is continuous on
+
w d(O)A(t)),t
t E 11,
(8.1.3)
J1,
(8.1.4)
E
a),
[-7,
(i) I Vt lo = sup,-<,Go V(t s,4(s)); (ii) L(u) is continuous on R , , nondecreasing in u, and L(u) u > 0; and (iii) I VtAt ( 0 = sup V(t s, +(s))A(t s).
+
-T<s
+
> u, for (8.1.5)
We now state a few fundamental comparison results.
THEOREM 8.1.1.
Let V E C[[-T, co) x So , R,] and V(t,3) be locally Lipschitzian in x. Assume that the functional D-V(t, +(O), +), defined by (8.1. I), verifies the inequality
o-v(t,+(O), 4) < g(t, l’(t, +(O))),
t
> to
9
4E Q,,
(8.1.6)
where g E C[J x R , , R,], and r ( t , to , uo) is the maximal solution of the 0. Let scalar differential equation (6.1.2), existing to the right of to %(to,+o) be any solution of (7.1.1) defined in the future, satisfying vto
SUP
--7<s
Then,
v,%(to
9
90)(t))
7
+0(4>
< r(t, t o
>
< uo uo),
(8.1.7)
t 2 to
-
(8.1.8)
Proof. Let %(to, +o) be any solution of (7.1.1) with an initial function E C, at t = to . Define the function
49 For
E
=
V(t,
4 t O 1 #O)(t)).
> 0 sufficiently small, consider the differential equation u‘
= g(t, u)
+
E,
u p o ) = uo 2 0,
(8.1.9)
whose solutions u(t, E) = u(t, to , uo , E ) exist as far as ~ ( t o, , uo) exists,
8.1.
83
BASIC COMPARISON THEOREMS
to the right of t o . Since lim u(t, e )
= r(t, to ,uo),
c-0
the truth of the desired inequality (8.1.8) is immediate, if we can establish that t 2 to. m(t) < u(t, E ) , Supposing that this is not true and proceeding as in the proof of Theorem 6.3.3, we can see that there exists a t, > to such that
<
< <
(i) m ( t ) u(t, E ) , t o t t, ; (ii) m(tl) = ~ ( t, E, ) , t = t , . From (i) and (ii), we get the inequality D-m(t1) 3 u p , , .)
= g(t1 , U ( t ,
4 ) + 6-
(8.1.10)
Sinceg(t, u) + E is positive, the solutions u(t, E ) are monotonic increasing in t , and therefore, by (i) and (ii),
I mtl Setting that
#
= xt,(t0 ,$,)
10 = m(t1) =
4 4 4. 9
and noting that #(O)
= x(t,
,~ ) ~ ) ( t , it ) , follows
This means that q5 E L?,, and, consequently, using the Lipschitzian character of V(t,x) in x and the relation (8.1.6), we obtain, after simple computation, the inequality D-m(t1)
< g(t1
>
4tl)).
Th i s is incompatible with (8.1.10), on account of (ii). It therefore follows that (8.1.8) is true, and the proof is complete. T h e following corollary is a useful tool in itself in certain situations.
COROLLARY 8.1.1. Let V E C [ [ - T , 00) x S , , R,] and locally Lipschitzian in x. Assume that, for t > to ,q5 E .R, , D-V(t, 4(0),4)
V(t, x) be
< 0.
Let x(t, , be any solution of (7.1.1) such that % ( t o +,)(t) , E So for t E [to , t,] C J. Then,
84
CHAPTER
8
Proceeding as in Theorem 8.1.1 with g = 0, we arrive at the inequality Proof.
w,4 t o
?
G
+o)(tN
w z
9
4 t o do>(tz)>, 7
where t, E (to , t l ) . Since V ( t , , ~ ( t,&)(t,)) , > 0, the assumptions L(u) imply that
< < t, . T h e rest of the proof is
which shows that x l ( t 0 , +,) E Qo , to t similar to the proof of Theorem 8.1.1.
T h e next comparison theorem gives a better estimate.
THEOREM 8.1.2. Let the assumptions of Theorem 8.1.1 hold except that the inequality (8.1.6) is replaced by
+
+ C(ll d(0)lI) < g(t, V t ,+(O)>>,
D+V(t,d(O), 4)
(8.1.11)
for t 3 t o , E C, , where the function C E X . Assume further that g ( t , u ) is monotone nondecreasing in 2c for each t . T h e n (8.1.7) implies
+
Set == x,(t, ,+o) so that +(O) = x(to ,+,,)(t). We then obtain, using the condition (8.1.1 I), the inequality D+m(t1)
7
4tl)).
Here, we have used the monotonicity of g(t, u ) in u and the fact that
q t ,4 t o
9
+o)(t))
G4th
while applying the assumption (8.1.11). With these changes, it is easy to prove the stated result, following the arguments in the proof of Theorem 8.1.1. T h e following variant of Theorem 8.1.1 is more useful in certain situations.
8.1.
BASIC COMPARISON THEOREMS
85
THEOREM 8.1.3. Assume that the hypotheses of Theorem 8.1.1 hold except that the inequality (8.1.6) is replaced by
for t > t o ,4 E SZ, Then,
, where A(t) > 0 is continuous on [-T, a).
and therefore, in view of the assumption (8.1.13), it follows that D-L(t, +(O),C)
e g(t, L(t, C(O))),
for t > t o , 4 E Q, , where Q, , in this case, is to be defined with L(t, x) replacing Y(t,x) in (8.1.2). It is clear that L(t, x) is locally Lipschitzian in x, and, thus, all the assumptions of Theorem 8.1.1 are satisfied, with L(t, x) in place of V ( t ,x). T h e conclusion is now immediate from Theorem 8.1.1. On the basis of the comparison theorem for functional differential inequalities developed in Sect. 6.10, we are now in a position to prove the following result, which plays an equally vital role in studying the behavior of solutions of functional differential systems.
THEOREM 8.1.4. Y E C[-T, a)x S , , R,], and V(t,x) is locally Lipschitzian in x. Assume that, for t E 1,4 E C, , D+V(t,(b(O),
where
Y t = Y(t
+ s, 4(s)),
4)< g(4 w,+(ON, -T
< s < 0,
g
E
Vt),
(8.1.15)
C [ J x R, x % + , R],
86
CHAPTER
8
g(t, u , a) is nondecreasing in u for each ( t , u ) , and r ( t o ,o0)is the maximal
solution of the functional differential equation
(8.1.16)
u‘ = g(t, u, U t )
with an initial function uo E %?+, at t = t o , existing for t >, t o . If x(t0 ,40) is any solution of (7.1.1) defined in the future such that Vt0 = J”t0
+ s, 4o(s)) <
(70
(8. I .17)
>
then we have
Proof.
Let x(t0 ,q50) be any solution of (7.1.1) such that
Set 4 = xt(t,, $,), which implies that +(O) m ( t ) = C'(t, X ( t ,
9
=
~ ( t,+,)(t). , Define
+o)(t)),
so that
v(t + s, d(s)). Since (8.1.17) holds, we have mto < uo . Moreover, for small h > 0, "1
=
because of the fact that V ( t ,x) satisfies a Lipschitz condition in x. This, together with (8.1.15), yields the inequality (8.1.19)
<
Then, it follows that D+v(t) 0, in view of (8.1.19). By Lemma 1.2.1, v(t) is nondecreasing, and therefore D_v(t) 0, which implies that D-4t)
< g(t, m ( t ) ,mt),
<
t >to.
T h e desired result (8.1.18) now follows from Theorem 6.10.4.
8.2.
87
STABILITY CRITERIA
8.2. Stability criteria We shall, in what follows, give sufficient conditions for various stability notions in terms of Lyapunov functions. This will be accomplished in two different ways. In one approach, the theory of ordinary differential inequalities will be used, as before, whereas in the other, the theory of functional differential inequalities play a major role. Since, for the purposes of this chapter, it is convenient to interpret the solutions of (7.1.1) as elements of euclidean space, the definitions of stability and boundedness have to be modified accordingly. For example, the definition 7.1.1 would appear in the following form.
DEFINITION 8.2.1. The trivial solution of (7.1.1) is said to be (S,) equistable if, for each E > 0, to E J , there exists a positive function
6
=
8 ( t o , c) that is continuous in to for each
E,
such that, whenever
I14 0 /lo < 6,
we have
I/ 4 t O do)(t)ll < 1
€7
t
2to.
With this understanding, we can prove the following results.
THEOREM 8.2.1. Let there exist functions V ( t ,x) and g ( t , u) enjoying the following properties : (i) V E C [ - T , GO) x S, , R,], V ( t ,x) is positive Lipschitzian in x, and V ( t ,).
< 4, II x ll),
definite,
( 4 4 E J x S" ,
where a E C[J x [0, p), R,], and a E 3'for each t E J ; (ii) g E C [ J x R, , R,],g ( t , 0) = 0, and, for t > to , D-Vt,
d(O>,4
locally (8.2.1)
E
SZ, ,
< g ( 4 q t , d(0"
Then the trivial solution of (7.1.1) is (1") equistable if the trivial solution of (6.1.2) is equistable; (2") uniform stable if the trivial solution of (6.1.2) is uniform stable and, in addition, V(t,x) is decrescent.
Proof.
Let x(t, , &) be any solution of (7.1.1). Choose
88
CHAPTER
< u,, , by
so that V(to,+,,) yields the estimate
8
(8.2.1). An application of Theorem 8.1.1
< r(t, t u , 4,
L'(t, 4 t o + o ) ( t ) ) 7
t 3 tu ,
(8.2.2)
where r ( t , t , , uo) is the maximal solution of (6.1.2). Also, because of the positive definiteness of V ( t ,x), we have
411x 11)
< t'(4
( 4 4E
x),
J x
s,,
bEx-.
(8.2.3)
Let 0 < E < p and to E ] be given. Assume that the null solution of (6.1.2) is equistable. Then, given b ( ~ > ) 0, to E J , there exists a 8 = 8(t,, E ) > 0 satisfying 4 4 t o , uo) < &),
<
provided u,, 8. Moreover, there exists a 6,
It +u
110
< 8,
(8.2.4)
t 2 to,
and
4tO
>
=
a1(t,,,e) such that
<
II d o ),1
(8.2.5)
hold together, because of the assumption on a(t, u).I t now follows, from
It d o I / d 81 that 9
b(ll4,
9
do)(t)ll)
< v, 4,+u)(t)) < r(t, to , a,,) I
<
and, consequently,
<
/I X ( t , ,do)(t)il <
w,
t 3 to,
t >, t o f
€7
whenever 11 +u I/, 6, . Thus, (1") is proved. If V ( t , x ) is decrescent, there exists a function a E X satisfying
w,4 < .(I1
x 1%
(4 4 E J x
s,
*
Hence, if we assume that the trivial solution of (6.1.2) is uniform stable, it is easy to see, from the foregoing proof, that 6, does not depend o n t o , proving (2"). T h e proof is complete.
COROLLARY 8.2.1. 8.2.1.
T h e function g ( t , u )
= 0 is admissible in Theorem
THEOREM 8.2.2. Assume that there exist functions V(t,x), g ( t , u), and A ( t )satisfying the following conditions: (i) A(t) > 0 is continuous on
[-T,
a),and A(t)-+ a as t -+a;
8.2.
89
STABILITY CRITERIA
(ii) V E C [ [ - T , co) x S o , R,], V(t,x) is positive definite, locally Lipschitzian in x, and verifies (8.2.1); (iii) g E C [ J x R, , R + ] , g ( t ,0) = 0, and, for t > t o , E SZ, ,
4)D - W
4)+ v(4 $(ON D - W
+(O),
+
< g(4 U t ,+(ON W).
Then, the trivial solution of (7.1.1) is equi-asymptotically stable if the trivial solution of (6.1.2) is equistable. Proof. If x(to,+o) is any solution of (7.1.1) such that 4 t o )4
0
9
I1$0
110)
=
uo 9
we have, by Theorem 8.1.3,
4 4 q t , 4 t O ,+o>(t)) < y ( t , t o
7
uoh
t
2 to
(8.2.6)
Let 0 < E < p and to E J be given. Let 01 = min-7Gi<mA(t).By assumption on A(t),it is clear that 01 > 0. Set q = 01b(~).Then, proceeding as in the proof of Theorem 8.2.1 with this q instead of b ( ~ )it, is easy to prove that the trivial solution of (7.1.1) is equistable. To prove equi-asymptotic stability, let q* = ab(p). Let S l ( t o , p ) be such that 11 #Io [lo 6, implies 11 %(to,#Io)(t)ll p, t 3 to . This is possible by equistability. Designate 8,(t0) = 8,(t, ,p), and suppose that (1 +o 8, . Since A(t)-+ co as t -+ 00, there exists a positive number T = T ( t o ,E ) such that
5
<
<
A(t)4)> ab(p),
t >, to
+ T.
(8.2.7)
We then have, using (8.2.3), (8.2.6), and the fact that u ( t , t o , uo) < q* if uo
<
S(t0
9
PI,
4 4 b(llX(t0
>
+o)(t)ll)
w,
%(to > +o)(t)> G G r(t, to uo) < T * = ab(p), t 3 t o . 9
If t >, to + T , it follows, from the foregoing inequality and (8.2.7), that provided 11 +o
[lo
< 8,.
This concludes the proof of the theorem.
COROLLARY 8.2.2. The functions g ( t , u ) are admissible in Theorem 8.2.2.
= 0 and A(t) = eat,
01
> 0,
90
8
CHAPTER
THEOREM 8.2.3. Assume that there exists a function V(t,x) satisfying the following conditions: (i) V t C[[-T, CO) x S, , R,], V ( t ,x) is positive definite, decrescent, and locally Lipschitzian in x; (ii) for t > t o , E Q, ,
+
u-If, m,4)< -C(ll d@)ll),
c: E .x.
Then, the trivial solution of (7.1 . l ) is uniformly asymptotically stable.
Proof. Since V is positive definite and decrescent, there exist functions a, b E .f satisfying
4 11) < V ( t ,4 < 4 x ll), Let 0
s :’
E
< p, t,
E
(4 ).
J be given. Choose 6
=
E
J x
s, .
(8.2.8)
> 0 such that
6(e)
a(S) < b(€).
(8.2.9)
<
6, then (1 x(t, ,&,)(t)/l< E , t 3 to . Suppose We claim that, if /I 4,,1 that this is not true. Then, there exists a solution ~ ( t,c, $ ~ of ) (7.1.1) with I/ 4o 1, 6 such that
<
II “ ( t o > 4 o ) ( b l l and so that
/I X ( t ,
?
4o)(t)ll
I,’(tz
?
<
t
€9
4 t , 4o)(tz>> J
=
6
E
[to >
4, (8.2.10)
3
because of (8.2.8). Furthermore, this means that x(t, , +,,)(t)E S, t E [t,,, t.1, Hence, the choice u,,= a(ll4, ),1 and the condition
,
give the estimate
If,44l
9
$o)(tN
< 41
$0
lid,
t
E
[to 9
f21,
(8.2.11)
because of Corollary 8.1.1. Now the relations (8.2.10), (8.2.1 l), and (8.2.9) lead t o the contradiction
This proves that the trivial solution of (7.1 . I ) is uniformly stable.
8.2.
91
STABILITY CRITERIA
T o prove uniform asymptotic stability, we have yet to show that the null solution of (7.1.1) is quasi-uniform asymptotically stable. For this , be any solution of (7.1.1) such that 11 +o (lo a,, purpose, let ~ ( t,+o) where 6, = 6(p). It then follows from uniform stability that
<
X(t0
Let now 0
< E < 6,
,do)(t) E so, t 2 t o *
be given. Clearly, we have b(€) < 4 8 , ) .
I n view of the assumptions on L(u), which occurs in the definition of Qo , it is possible to find a ,8 = P(E) > 0 such that
+ /3
L(u) > u
if
b(6)
< u < a(6,).
Moreover, there exists a positive integer N inequality b(€)
If, for some t
+ N/3 >
=
(8.2.12)
N ( E ) satisfying the (8.2.13)
480).
t o , we have
w4to
do>(t))2 b ( 4 ,
it follows that there exists a 6, = a,(€) > 0 such that 11 ~ ( t, cjo)(t)lI , >, 6, because of (8.2.8). This, in turn, implies that C(lI X ( t 0 do)(t)ll) ?
2 C@,) = 8 2 -
(8.2.I 4)
Obviously, 6, depends on E. With the positive integer N chosen previously, let us construct N numbers t,i = t k ( t o ,E), k = 0, 1, 2,..., N , such that 9
€)
=
tk+l(tO
7
9
t k ( t O , €)
€)
+ @is,) +
,
+1
T*
It then turns out that
+ and, consequently, letting T ( E )= N[(,8/8,)+ tk(tO
3
+
€1=
'[(/3/'2)
T],
tN(t0,
4 = t o + T(+
we have
Now, to prove quasi-uniform asymptotic stability, we have to show that
II X ( t ,
P
90>(t)II
<
€7
t
2 to
+T(4
92
CHAPTER
whenever 11 $,
8
< 6, . It is therefore sufficient to show that
/lo
q t , ,'(to 4")(4) < h ( E ) 9
i -( N - k ) 8,
2 t,
t
(8.2.15)
1
for h 0, 1, 2,..., N . For k 0, (8.2.15) follows from the first part of the proof and (8.2.13). We wish to prove the desired inequality (8.2.15) -
by induction. Suppose that, for some k
and, for some t , >, t,.,.,
-
< N , we have
(pis,),
It then follows that
and, consequently, we derive from (8.2.12) the inequality L(V(t* "( to 4o)(t*))) 9
for t,
7
< [ < t , . This implies that
> h(6) + ( N 4 B > "(to $ o ) ( E ) ) ,
w,
-
t
L ( V ( t , ,4(0)))'> V(t* t- s, $(S))>
-7
< s < 0,
where 4 = xl,(t,, , +J, so that +(O) = %(to, $J(t,). Hence, 4 E Qo . It therefore follows from condition (ii) and the relation (8.2.14) that
n- V ( t ,4(0),4)<
< 0.
-82
This shows that, if ever LXt, "( to
,4o)(t)) < b(6)
+(N
-
1)
-
B 7
for some t >, t x t l (pis,), then the same inequality will hold henceforth. Now, supposing that ~
q t , "(to, 40)(t)) 2 b ( E ) for tl,+, - (pis,) 1 'ft*. Xfto
t
+ (N
-
k
-
1) B
< t < t*, we deduce, by arguing as before, do)(t*)) < b(6) -t ( N - R, P
< q t * , 4 t O ,$o)(t*))
- 82(t* -
82(t*
- tk+l
+
- GC+l),
@/'2))
8.2.
93
STABILITY CRITERIA
<
which proves that t* t,,,, . I n other words, we have verified the truth of (8.2.15) for all k. T h e proof of the theorem is therefore complete. T h e next theorem may be useful in some situations.
THEOREM 8.2.4. Assume that there exist functions V ( t ,x) and g(t, u) satisfying the following conditions:
(i) V E C[[-T, GO) x S o ,R,], V(t,x) is positive definite, decrescent, and locally Lipschitzian in x; (ii) g E C [ J x R, , R,], g(t, 0 ) = 0, g(t, u) is monotone nondecreasing in u for each t , and, for t t o ,(b E C , ,
> D+ V(4 4(0)7 a, + C(Il 4(0)il) < g(4
d(O))),
where C E Z. Then, the trivial solution of (7.1 .l) is uniformly asymptotically stable if the trivial solution of (6.1.2) is uniform stable.
Proof.
By Theorem 8.1.2, we have
v4 4 t o 4o)(t)) 9
<
As this implies,
~
st
C(llX(t0
+ r(4 t o , to
UO),
< r(4 t o , U o ) ,
V t , 4 t o +o)(t)) 9
7
4o(s)ll) ds
t
2 to *
t 3 to ,
on the basis of Theorem 8.2.1, it is clear that the trivial solution is uniform-stable. Let 8, = S,(p), where S,(p) is the number obtained for uniform stability. Similarly, let S,(E) be the number corresponding to E, and suppose that 11 x(t0 ,(bo)(t)J( >, S,(E) for t E [to, to -t TI, where T is chosen to satisfy the inequality (8.2.16) Since r(t, to , uo) follows that
< b(p), t
b(llX(t0
9
2 to ,
4o)(to
whenever uo = u(ll $o
+ T)I/ < - C ( U E ) )
Ilo)
< Q),
it
T -1- b(P).
<
This implies, because of (8.2.8), that 0 < b(S,(c)) 0. This contradiction proves that there exists a t , E [to , to T ] such that
+
II 44l do)(t1)ll < S1(+ 7
94
CHAPTER
8
I t therefore follows, from the decrescent nature of V ( t ,x), that, in any case,
II 4 t " , (bo)(t)ll < whenever
< 8,.
I/ +o 11"
t
€7
3 to
+ 1'9
T h e proof is therefore complete.
COROLLARY 8.2.3. T h e function g(t, u) = 0 is admissible in Theorem 8.2.4. We shall now consider Eq. (7.1.8) and illustrate the practicality of using Lyapunov functions instead of functionals. Take L(t, x) = A(t) V ( t ,x) = eatx2, a: > 0. Then, the set QA is defined by
which implies#Jz(s) (7.1.9), we see D - L ( ~+(o>, , +)
< +2(0)e-as,
= eat [ -
q
- T
2(" ~
~
< s < 0. Hence, using the relation
+ b) ~ ( 0 )
+(s) ds
2ab$(0)
-
2b2$(0)
-7
When
#J
E QA
/"
I$(S
~
T) ds]
.
-7
, this reduces to, assuming b 3 0,
=
[.2(a + b ) + (4ub + 4b2 eaT)(@+
N
-
"1
q t , $(O)).
We wish to apply Theorem 8.2.2 with g ( t , u ) e 0. This means that a, 6 , OL, and T must satisfy the condition
Clearly, by Theorem 8.2.2, the trivial solution of (7.1.8) is uniformly asymptotically (exponentially) stable, provided (8.2.17) is verified. Observing that e" - 1 3 x,x > 0, and ex >, 1 , it follows, from (8.2.17), that 01
+ 2b(U + b ) < 2 ( U f b). 7
8.2.
Now, choosing condition
01
=
+ b), 0 < y < 1,
2y(a
< br < 1
0
95
STABILITY CRITERIA
we readily deduce the
< 1,
-y
which is the known condition for asymptotic stability. Since OL is arbitrary, letting 01 -+0 and noting that lim
m
earl2 -
4
1
__ -
01
7
-
2’
the condition (8.2.17) yields 0
< 1,
which is the condition obtained before for uniform stability. If we wish to obtain the stability region in a different way, we have to compute D L ( t , 4(0),4) using (7.1.8) directly, instead of (7.1.9) as before. We then get the inequality D-L(t,
W), 4) < [a
-
2(a
-
I b I ear)]L(t, d(0)).
T o apply Theorem 8.2.2 with g = 0, we must have 0
< 0112 < a - I b I eur,
which implies that the region for uniform asymptotic stability is 1 b 1 < uerar. If 01 -+ 0, this condition reduces to 1 b I a, which is the region for uniform stability for Eq. (7.1.8). Notice that the latter condition I b I < a may also be obtained by taking V ( t ,x) = x2, computing D-V(t, +(O), 4) for 4 E Q, , and using Theorem 8.1.1 with g = 0. Finally, take V ( t ,x) = x2,L(u) = u/q, 0 < q < 1, and
<
C(U) =
241
- q 1 / 2 ) 242.
Then, one can verify the conditions of Theorem 8.2.3 by a calculation similar to the foregoing. The trivial solution of (7.1.8) is uniformly asymptotically stable if I b I < The following theorem shows that the study of the stability properties of the functional differential system can be reduced to that of a single functional differential equation. Naturally, this technique depends heavily on Theorem 8.1.4, as anticipated. Analogous to the stability definitions with respect to the scalar differential equation (6.1.2), if we say that the trivial solution of (8.1.16) is equistable, we mean the following.
96
CHAPTER
8
DEFINITION 8.2.2. T h e trivial solution of (8.1.16) is said to be equistable if, for each E .: 0, to E J , there exists a 6 = 6 ( t o , E), which is continuous in t, for each E , such that
to , o o ) ( t ) <
<
t >, t o
€9
provided 1 uo lo 6, where u ( t o , uo) is any solution of (8.1.16) with an initial function uo E 9,at t = t o . Other definitions may be understood similarly.
THEOREM 8.2.5. Assume that there exist a function V ( t ,x) and a functional g(t, u, u) satisfying the following conditions: (i) Y E C[[-T,
CO)
4 x 11)
x S, , R,], V(t,x) is locally Lipschitzian in x,and
< Ut, 4 < 4 x I\),
(4 2) E J x
s,,
where a, b E .X; (ii) g E C[J x R , x %+ , R],g(t, 0 , O ) E 0, g ( t , u, a) is nondecreasing in o for each ( t , u), and, for t E J , 4 E C, ,
W),$1 < g(4 v(4 W)), Vt).
D'CIt,
Then, the trivial solution of (7.1.1) satisfies one of the stability notions, if the trivial solution of (8.1.16) obeys the corresponding one of the stability definitions. Suppose that the trivial solution of (8.1.16) is equistable. Let ) 0, to E J , there exists a p and to E J. Then, given b ( ~ > 6 ( t o , e ) > 0 such that 1 uo lo 6 implies
Proof.
0 <E 6
=
< :
<
4 t O , oo)(t) < b(c),
t 3 to *
Let ~ ( t,+o) , be any solution of (7.1.1) existing for t 3 to such that V(to+ s, do(s)) u0(s).We then have, by Theorem 8.1.4, the inequality
<
L7(t,
to
9
+o)(t))
G
r(tn
, oo)(t),
t
2 to ,
where r ( t o , a,) is the maximal solution of the functional differential equation (8.1.16). Choose 6, > 0 such that 6, = a-'(6), and let jl +o 1, 6, . I t turns out that
<
41 to ,+n)(t)ll) G
V t , 4 t n +o>(t>)
< r(tn
on)(t)
< b(€),
t
>, to
9
8.3.
97
PERTURBED SYSTEMS
which implies that
I1 4 t o > +o)(t)ll < E ,
<
t
2 to >
provided that 11 +o 1l0 6 , , showing that the trivial solution of (7.1.1) is equistable. T h e proof corresponding to other notions may be given by modifying the arguments suitably. T h e theorem is therefore proved. COROLLARY 8.2.4. T h e functional g ( t , u, U) = 0 is admissible in Theorem 8.2.4 to yield uniform stability of the trivial solution of (7.1.1).
8.3. Perturbed systems Let us consider the perturbed system (7.4.1) corresponding to the unperturbed system (7.1.1). We then have
THEOREM 8.3.1. Suppose that (i) V E C[[-T, co) x S o , R,], V ( t ,x) is positive definite and satisfies > 0; (ii) for t > to ,4 E 9, ,
a Lipschitz condition in x for a constant L = L ( p )
D - V , 4(0),+)(,.IA
e 0;
(iii) w E C[J x R, , R,],w ( t , 0) = 0, and, for t
> t o ,4 E 9, ,
I/ R(t,4111 e w(t9 V(4 d(0))). Then, the trivial solution of (7.4.1) is equistable (uniform stable) if the trivial solution of (6.1.2) with g(t, a) = Lw(t, a) is equistable (uniform stable). Proof. Let t > to and and (iii), we have D-V(t, 9(0),4)(,.4.1)
4 E Q, . Then,
in view of assumptions (i), (ii),
< limk-0-inf lz-l[V(t + h, 4(0) + h { f ( t , 4 ) + q t , 4)) -
v(t + d(0) + h f ( 4 4111
+ lim inf W V ( t + h, +(O) + V ( t ,4)) < L II R(t,9)ll + D-C'(t, +(O), h9
-
h 4 -
4h7.1.1)
< Lw(t, q t , W)))= g(tt F ( t , #(ON.
T h e desired result is now a consequence of Theorem 8.2.1.
I.-(t, d(0))l
98
CHAPTER
8
THEOREM 8.3.2. Suppose that (i) V E C[[-T, 00) x S, , R,], V(t,x) is positive definite and satisfies a Lipschitz condition in x for a constant L = L ( p ) > 0; A(t)400 as t 00, and, (ii) A ( t ) > 0 is continuous on [ - T , a), fort>t,,+EQ,, ---f
+ V t ,4(0))D - 4 t ) < 0; C [ J x R, , R,],w(t, 0) 0, and, for t > t o ,+ A f t )D-V(t, 4(0),6)
(iii) w
E
E
A ( 4 I/ R(t,6111
< 4 4 JV,
E
Q, ,
A@)).
Then, the trivial solution of (7.4.1) is equi-asymptotically stable if the null solution of (6.1.2) with g ( t , u ) = Lw(t, u ) is equistable.
Proof.
If t
> to , + E QA , it follows that
which implies the inequality a t ) D-V(4 +(Oh
d4(7.4.1)
+ v(t,(b(0))D - 4 ) < Lw(4 w, $((I)) A@)).
Consequently, the conclusion follows by Theorem 8.2.2.
THEOREM 8.3.3.
Suppose that
(i) V E C[[-T, 00) x S, , R,], V ( t ,x) is positive definite and satisfies a Lipschitz condition in x for a constant L = L(p) > 0; (ii) for t >, to , E C, , and C E X ,
+
D'V(t3
4h.l.l)
< -C(ll4(0)11);
(iii) w E C [ J x R, , R,.], w(t, 0 ) = 0, w(t, u) is nondecreasing in u for each t E J , and, for t 3 t o , E C,, ,
+
/I R(t, 4111 < w(t, V ( t ,4(0))). Then, the uniform stability of the trivial solution of (6.1.2) with g ( t , u ) = Lw(t, u ) assures the uniform asymptotic stability of the trivial solution of (7.4.1).
Proof,
Let t
3 to and + E C, . Then, as previously,
D+V(t,+('),
$)(7.4.1)
d(o)li) + Lw(t, V(t,+(O))),
and, therefore, the uniform asymptotic stability of the trivial solution of (7.4.1) follows by Theorem 8.2.4.
8.3.
THEOREM 8.3.4.
99
PERTURBED SYSTEMS
Assume that
(i) V EC[--7, CO) x S o ,R,], V(t,x) satisfies a Lipschitz condition in x for a constant L = L ( p ) > 0, and b(ll x II)
in
< V t , 4,
(4 4 E J x
s,,
b E .x;
(ii) g, E C [ J xR, x V, , R],gl(t, 0, 0) 3 0, gl(t, u, u) is nondecreasing u for each ( t , u), and, for (t,4) E J x C, , D+V(t,#('),
gl(t, v(t,#(O)), ;)6'
#)(7.1.1)
(iii) g, E C[J x V+ , R,],gz(t,0) = 0, gz(t,u) is nondecreasing in u for each t E J , and II R(4 #)I1 < gz(4 II # 11)Then the stability properties of the trivial solution of (8.1.16) with g(4 u, 4 = gl(4 u, 4
+ Lg&, b - W
imply the corresponding stability properties of the trivial solution of (7.4.1). Proof.
Let t E J and #J D'v(t, #(O),
E
#)(7.4.1)
Co . Then, D'v(t, #(O), #)(7.1.1) < gl(C v(4#(O)), V t ) < gl(4 v(4#(O)), V t ) = g(4 W)), Vt),
w,
+ 11 R ( f ,#>I1 + Lgz(4 II 4II) + Lgz(4 b-YVt))
because of assumptions (i), (ii), and (iii). Now, Theorem 8.2.5 can be applied to yield the stated results.
COROLLARY 8.3.1.
T h e functions gl(4 u, 0) =
--01u
+
PT
sup -T<S
44,
> 0, p > 0, and g2(t, 0) = y sup-,GsGo u(s), y being sufficiently small, are admissible in Theorem 8.3.4, provided b(u) = u and 0 < T < (01 - y ) / P , to guarantee that the trivial solution of (7.4.1) is exponentially asymptotically stable. 01
Proof. Under the assumptions, it is easy to see, as in Theorem 6.10.7, that u' = -01u (PT y ) sup-,GsGo ut(s) admits a solution r ( t o ,uo) that tends to zero exponentially as t + CO, and, therefore, the conclusion follows from Theorem 8.3.4.
+
+
100
CHAPTER
8
8.4. An estimate of time lag We wish to estimate the time lag r in order that the solutions of an ordinary differential system (8.4.1)
x’ = f ( t , x)
and a functional differential system Y’
=
(8.4.2)
F(4 Y t )
may have the same behavior, namely, exponential decay. Since Eq. (8.4.2) may also be written as x’
where
=f
( 4 x)
R(4 x, 4
+ R ( t , x, 4,
= F(4
(8.4.3)
4 - f ( k 4,
it is sufficient to consider the perturbed system (8.4.3).
THEOREM 8.4.1. Suppose that (i) I/ E C [ S , , R,], V ( x ) is positive definite and satisfies a Lipschitz condition in x for a constant L = L(p) > 0 ; (ii) f E C[/ x S , , R”], and, for (t, x) E J x S,, ,
< -aL’(x),
D+V(X)(&4.1)
(iii) R
E
01
> 0;
C [ / x S, x C, , R”],and for 4 E C, ,
/I R ( f >+(Oh +)I1
< NT
SUP
--7<s
%w).
Then, if 0 < r < a/LN, every solution ~ ( t, +o) , of (8.4.3) with +o E C, at t = t o , which is defined in the future, tends to zero exponentially as t 4 a.
Proof. Let t E J and 4 E C, . Then, using assumptions (i), (ii), and (iii), we have
8.5.
101
EVENTUAL STABILITY
It is clear that g(t, u, a) satisfies the conditions of Theorem 6.10.4, and, consequently, J+(t,
7
<4 t O
4o)(t))
9
t
oo)(t),
2 to
Y
is any solution of (8.4.3) with the initial function E C,, where x(t0 , at t = to , defined in the future, and r ( t , , uo) is the maximal solution of (8.1.16) with uo E V+ at t = to such that V(+,(s))< uo . All that remains to be shown is that r(to , uo) exists in the future and tends to zero exponentially as t -+ CO, if 0 < T < a/LN. I n fact, as we have seen in the proof of Theorem 6.10.7, the equation u'
-mu + L N T
= g(t, u , u , ) =
sup
u,(s)
-T<S
admits solutions ~ ( t,,uo) that tend to zero as t + co exponentially, if 0 < T < a/LN. T h e theorem is therefore established. COROLLARY 8.4.1.
Assume that
(i) f E C [ J x S,,, P
II x (ii) R
E
I , and, for all sufficiently small h > 0,
+ hf(4 .)I1 < II x ll(1
-
ah),
01
> 0,
( 4 x) E J x so;
C [ J x S,, x C, , R"], and, for ( t , + )E J x C, ,
II R(4 d(O), 4111 G NT --r<s
8.5. Eventual stability We shall extend the results on eventual stability discussed in Sect. 3.14 to the functional differential system (7.1.1). Following the definition 3.14.1, we can formulate eventual stability concepts in the present case. For example, the set = 0 is said to be [with respect to the system (7.1.1)] ( E l ) eventually uniform@ stable if, for every E > 0, there exists a 6 = S ( E ) > 0 and T~ = T ~ ( E )> 0 such that
+
+, <
I/ X ( t 0 ,4o)(t)ll < E ,
t
2 t o 2 70
f
provided 11 6. Analogously, we shall denote by ( E l * ) to (E4*)the corresponding notions of the set u = 0 with respect to the functional differential equation (8.1.16).
102
CHAPTER
THEOREM 8.5.1.
8
Let the following conditions hold:
(i) V E C[[--7,
GO)
x S, , R,], V ( t ,X) is locally Lipschitzian in X, and
411x 11)
< V t , "> < 4x 11)
for 0 < OL < 11 x 11 < p and t >, O(ct), where a, b E X and O(u) is continuous and monotonic decreasing in u for 0 < u < p ; (ii) g E C [ J x R+ x V+ , R ] , and the set a = 0 is eventually uniformly stable with respect to (8.1.16); (iii) f E C [ J x C,, R"], and
for every
+
D+V(t,d@), 4) E
C, such that 0 < a
Then, the set (7.1.1).
+
=
e g(t, L T 4 %(0)),Vd
< I/ +(O)lI < p and t >, O(a).
0 is eventually uniformly stable relative to the system
Proof. Let 0 < E < p. By assumption, given b ( ~ > ) 0, there exists a T ~ ( E> ) 0 such that 6, = a,(<) > 0 and a T ,
<
1 q,lo
4 t o , uo)(t)< 44,
t
2 t o 2 71
(8.5.1)
9
8,. Let us set 6 = ~ ~ ( 6 ,and ) T , ( E ) = 0[8(~)]. Let max[T1(E), T ~ ( E ) ] . We claim that (El)is satisfied with this choice of S ( E ) and T ~ ( E ) .Suppose that this is not true. Then, there would exist a solution x(t, , +,) of (7.1 . I ) and two numbers t, , t, such that t, > t, > to 2 T o ( € ) , whenever =-
T"(E)
=
II "(to and 8
=
> sbO)(tl)ll
< ll "(to
7
II 443
8,
4o(t)ll
< El
3
t
%o)(tz>lI E (tl
7
=
tz).
<
E,
+
Moreover, 11 xt(t0 , $o) l o E , t E [to , t l ] . Setting = x,(t, , +o) for t t (tl , t,), we see that S < jl +(O)lI < E . We choose I u0 10 = a(ll $1Il,), where +1 = xl,(t,, +"). Then, condition (iii) and Theorem 8.1.4 yield the inequality I '(t>Y ( t l +l)(t))
<4tl
?
t
oo)(t),
E [tl
> t21,
(8.5.2)
where y(tl , is any solution of (7.1 . I ) through (tl , +1), and Y ( t , , a,) is the maximal solution of (8.1.16) through ( t l , a,,). I t turns out that (8.5.2) is also true for ~ ( t, +,J , on the interval t, t t, . Hence, we get
< <
b(E)
-< r,-(t, , "(to ,% " ) ( t Z ) )
Y(t,
3
uo)(tz)
< b(E),
8.5.
103
EVENTUAL STABILITY
on account of the fact that t, > t, > to >, T ~ ( E )and the uniformity of the relation (8.5.1) with respect to t o . This contradiction shows that ( E l ) is valid, and the theorem is proved. COROLLARY 8.5.1. The uniform stability of the trivial solution u = 0 of (8.1.16) is admissible in Theorem 8.5.1 in place of the eventual uniform stability of the set u = 0. I n particular, g ( t , u, a) = 0 is admissible.
THEOREM 8.5.2. Assume that (i) V EC[[-T, 00) x S, , R,], V(t,x) is Lipschitzian in x for a constant L = L(p) > 0, and
&ll x II)
e v(44 < 4 x ll),
for 0 < 01 < 11 x 11 < p and t >, O(a), where a,b E X and O(u) is continuous and monotonic decreasing in u for 0 < u < p ; (ii) f E C [ J x C, , R"],and Df
w,W),4)e 0,
for every 4 E C, such that 0 < 01 < 11 +(O)lI < p and t 2 O(cx); (iii) R E C[J x C,, R"], and, for every 4 E C,*, p* < p and t >, 0,
Then the set 4 = 0 is eventually uniformly stable with respect to the perturbed system (8.5.3) x' = f ( 4 X t ) R(t, 4.
+
Proof. Let 0 T ~ ( E )such that
< E < p* 248)
be given. Choose the numbers 6
< b(e)
11 R(t, $)I[. Define h(t) = max,,+,,oGDt to find a T ~ ( E ) > 0 such that rrca
and
T~(E= )
0(8(e)),
=
a(€) and (8.5.4)
Since h(t) is integrable, it is possible (8.5.5)
provided to >, T ~ ( c ) where , L is the Lipschitz constant for V ( t , x). Let = m a x [ ~ ~ (T~(E)I. ~),
TO(€)
104
CHAPTER
8
Suppose that there exists a solution x(t0 , of the perturbed system (8.5.3) and two numbers t , , t, such that t , > t , >, to >, T,,(E),
At t where = xl,(to, (8.5.5) and the fact that I/
< V(tz ,
X(t,
7
t,
=
< 6,
, we therefore obtain, in view of
Co)(tz)>
< 4 8 ) + 4) = 24%
which is incompatible with (8.5.4). This shows that ( E l ) holds, and the theorem is established.
THEOREM 8.5.3. Let assumption (i) of Theorem 8.5.1 hold. Suppose further that f E C [ J x C, , R"] and
+
D+
$(O),
4 ) < -C(ll4(0)ll),
for every E C, such that 0 < 01 < I/ +(O)lI < p and t >, e(u) and C E %. Then, the set = 0 is eventually uniformly asymptotically stable.
+
Proof. T h e eventual uniform stability of the set C$ = 0 follows by Corollary 8.5.1. Let 0 < E < p be given. Choose 6, = 6(p), T~ = ~ ( p ) , and T ( E )= T ( E ) {a(p)/C[G(e)]). Assume that to >, T~ and 11 I(o 6,. It is sufficient to show that there is a t , E [to T ( E ) , to T ( E )such ] that
+
+
+
II X ( t 0 4 O ) ( ~ l ) I l < Y E ) , Y
in order to complete the proof. Suppose, if possible, that
< /I "(tn
8(~)
4n)(t)ll <
PF
t E [to
+
4e)t
to
+ T(E)I.
<
8.6,
Letting #I
=
105
ASYMPTOTIC BEHAVIOR
xl(t0 ,#Io), we see that
G )G 114(0)11 < P and 1) $ )lo
Hence, it follows that
It then turns out that
which is an absurdity. This means that there exists a ti E
[to
+
~ ( c ) , tn
+ T(6)I
such that 11 ~ ( t,+,,)(t,)ll , < a(€), and therefore, in any case,
/I 4%do)(t)ll < 3
provided 11 +o (I,
€9
t 2 to
+T(4
to
2 44,
< 6, . T h e proof is complete.
8.6. Asymptotic behavior I n this section, some of the results of the Sect. 3.15 will be extended to the functional differential systems.
THEOREM 8.6.1. Assume that (i) V E C[[-T, a)x R", R,], V(t,0) = 0, V ( t ,x) satisfies a Lipschitz condition in x for a continuous function K ( t ) 0, and
>
411x 11) G
w 4,
where b E X is such that b(u) -+ co as u + 00;
106
8
CHAPTER
(ii) g E C [ J >: R, x %+' , R ] ,g ( t , u, a) is nondecreasing in u for each ( t , u),and, for t E J , 4,4 E Vn,
n+ If,+(O) =
--
4 $1
$(Oh
-
lirn sup h-YC'(t h-Of
+ h, 4(0)
Then, every solution ~ ( t,, t + co.
Proof.
Let t
E
-
$(O)
+ h { f ( t ,4)- f ( 4 $)I
of the system (7.1.1) tends to zero as
en.Then,
J and y5 E
+ h, 4(0) + W ( t ,4)) - v(t,d(0))l < K(t)llf(t, 0)ll + lim sup h-l[V(t + A, 4(0)
I)+ I -(t,4(0), 4 ) -= lim sup h-l[V(t h-04
+ h { f ( t ,+) - f ( C
h 4 f
0)))
~
V t , 4(0))l
< K(t)llf(t, 0)Ii + g(t, U t , W)),l~'t>,
in view of assumption (ii). Let ~ ( t, $ , ,) be any solution of (7.1.1) such that V(t, S , $ ~ ( S ) ) u,(s). It then follows, by Theorem 6.9.4, that
+
<
v(f,"(to
9
+o)(t>)
< r(to
3
oo)(t),
t b to >
(8.6.2)
where r ( t , , u0) is the maximal solution of the functional differential equation (8.6.1). T h e stated result is now a direct consequence of the hypotheses (i) and (iii). T h e next theorem is very useful in applications, since it does not demand V ( t ,x) to be positive definite.
THEOREM 8.6.2. Assume that (i) f E C[J x C,, R"], and
llf(f, 4111 < M ,
tE
1,
I1 4 I10
< P*
8.6.
107
ASYMPTOTIC BEHAVIOR
(ii) V E C [ [ - T , 03) x S, , R,], V ( t ,x) is locally Lipschitzian in x,and
o+w,W),4)< -C[d(O)l, for t E J and 4E 52, where C(x) is positive definite with respect to a closed set 52 in S, ; (iii) all the solutions x(to ,$,J of (7.1.1) are bounded for t >, to . Then, every solution of (7.1,l) approaches the set 52 as t + CO.
Proof. Let x ( t 0 , 40) be any solution of (7.1.1). By assumption (iii), it is bounded, and, hence, there exists a compact set Q in S, , such that t 2 to to ,do)(t) E Q , Moreover, it also follows that 11 x , ( t o , +o)llo < p* < p , t 3 t o , and
therefore, by assumption (i), we have lI.f(t9
%(to
9
d0))ll
< M-
Suppose that this solution does not approach 52 as t + CO. Then, for some E > 0, there exists a sequence {trc},t, + co as k + CO, such that 4 t n > ddtd E
S(Q3
€1" n 8,
where S(Q, E)" is the complement of the set S(52, e ) = [x : d(x,Q) < €1. We may assume that t, is sufficiently large so that, on the intervals t, t t, ( E / ~ Mwe ) , have
< < +
4 t o 2 do)(t) E S(Q,
4v nQ-
(8.6.3)
These intervals may be supposed to be disjoint, by taking a subsequence of (t,), if necessary. By Theorem 8.1.2 and assumption (ii), we get V
9
4 t O do)(t))
<
SUP --7<s
Uto>do(4
-
st
to
C[x(to9 CO)(S)I ds,
(8.6.4)
for t >, to . Since C(x) is positive definite with respect to 52, the relation (8.6.3) shows that there exists a 6 = 6 ( ~ / 2 )> 0 such that C[x(tO
~ d O ) ( ~ ) l2
It therefore turns out that
tk
< <
t7c
f (E/2M).
(8.6.5)
I08
CHAPTER
8
on account of (8.6.4) and (8.6.5). T h e foregoing inequality leads to an absurdity as tZ 03, since, by assumption, V ( t ,x) >, 0. As a con! as t -+ 00, and the sequence, any solution x(t, , $o) tends to the set 2 theorem is proved. Making use of two Lyapunov functions, we can extend Theorem 4.2.1 to functional differential system (7.1.1). ---f
THEOREM 8.6.3.
Let the following assumptions hold:
(i) f~ C [ J x C,, , R n ] , f ( t ,0) = 0, andf(t, 4)is bounded on J x C,, ; (ii) V , E C[[-T, CO) x S o ,R,], Vl(t,x) is positive definite, decrescent, locally Lipschitzian in x, and, for t E J , (b E C, ,
< 44(0)) < 0,
D+Li(t,4(0i 4)
where ~ ( x is) continuous for x E So ; (iii) V , E C[[-T, co) x S o ,R,], Vz(t,x) is bounded on J x S, and is locally Lipschitzian in x. Furthermore, given any number a, 0 < 01 < p, there exist positive numbers [ = ,$(a)> 0, 7 = ~ ( a> ) 0, 7 < 01 such that
d(O), 4)> 4 that < II $(O)ll < p
D’L72(t,
for every t where
3 0, $ E C,, such E
OL
= [x €
so: a(.)
=
and d(+(O),E )
< 7,
01,
and d(x,E ) is the distance between the point x and the set E. Then, the trivial solution of (7.1.1) is uniformly asymptotically stable.
Pmof. As the proof requires appropriate changes in the proof of Theorem 4.2.1, we shall indicate only the modifications. Let 0 < E < p and to E J. Since Vl(t,x) is positive definite and decrescent, there exist functions a, b E X , satisfying b(lI x 11)
Let u s choose S
< l’d4 x) < 4
ll),
( 4 ). fzJ x
so.
6 ( ~> ) 0 such that
:
b ( € ) > a(6).
Then, by Corollary 8.2.4, the uniform stability of the trivial solution of (7.1.1) results. Let us designate 6, = S(p). Assume that ]I +o ]lo < S o . T o prove the theorem, it is sufficient to show that there exists a T = T ( E )such that, , < S ( E ) . As in the proof of for some t , E [t, , to ?’],jl ~ ( t,+o)(t,)jl Theorem 4.2.1, this will be achieved in a number of steps:
+
8.6.
(1") If d[x(t,), x(t,)]
109
ASYMPTOTIC BEHAVIOR
> r > 0, t, > t, , then Y
< Mnl/2(t,
t,),
~
<
where x(t) = x(t, ,+,)(t) and Ilf(t, +)\I M , ( t ,+) E J x C, . There is no change in the proof of this statement. (2") By assumption (iii), given 6 = a(€), 0 < 6(c) < p, there exist two positive numbers ( = ((E), 7 = q ( ~ )7, < 6, such that
for every t E J , sider the set U
+
=
D+V,(t, 4(0), 4 ) > 5 E
< 11 +(O)lI < p,
C, such that 6
[x E S, : 6
< 11 x 11 < p
and
d(+(O),E ) < 7. Con-
d(x, E ) < q],
and suppose that S U ~ V,(t, ~ ~x) ~= L. ~ Assume ~ < ~ that, at t = t, , we have x(tl) = x(t, , +,)(t,) E U. Then, for t > t, , it follows, by letting m ( t ) = V2(t,x(t)),that D+m(t) 2 D+V,(t,
4)> E,
because of condition (iii) and the fact that V2(t,x) satisfies a Lipschitz condition locally in x. I n obtaining the foregoing inequality, we have set 4 = xt(t,, 4,) so that +(O) = x(t) and used the inequality 11 4 1, = 11 xt(t, ,$,)1 0 < p, which is a consequence of uniform stability. Since m(t)
~
m(tl) =
It
D+m(s) ds,
tl
as long as x(t) remains in U , we should have
> E(t
-
tl).
Th i s inequality can be realized simultaneously with m ( t ) t
< t,
+( W E ) .
< + ( 2 L l f )with the
Hence, it follows that there exists a t, , t , < t, t, property that x(t2)is on the boundary of the set U. (3") Consider the sequence {tic} such that t , = to
+ k(2L/5)
< L only if
(k = 0, 1, 2 ,..,,).
I LO
CHAPTER
8
Defining n(t) = Vl(t,x(t)) and using assumption (ii), we obtain D+n(t)
+
< D+li(t, W),4)< 0,
where = xt(to , do), as before. Let A, = h7/2Mr~l/~,where h = inf[l(w(x)i, 6 < I/ x /I < p, and d ( x , E ) >, 71/21. Suppose that x(t) satisfies, for t, t t,,, , the inequality 6 < 11 x(t)ll < p. Then, arguing as in the proof of Theorem 4.2.1, with obvious changes, we can show that
< <
f71(Gc+2 > 4 t k f 2 ) )
< Vdtk
> 4tk)) -
A,
*
We now choose an integer K* such that h,K* > a(S,) and let T = T ( E )= 4K*L/l. Assuming that, for to t to T , we have
< < +
/I 4t)lI 2
w
7
we arrive at the inequality, as in Theorem 4.2.1, I yi(to
+
7’9
X(tn
+ T ) ) < Vi(to, 40) ~
< a@,)
-
K*h,
K*hi
< 0,
which is absurd, since Vl(t, x) is positive definite. It therefore turns out TI such that (1 x(t*)ll < 6, and this proves that there exists a t* E [to , t, the uniform asymptotic stability of the trivial solution of (7.1 .I).
+
8.7. Notes T h e comparison theorems 8.1.1 and 8.1.3 are due to Lakshmikantham [ I , 61. See also Driver [3]. Theorem 8.1.4 is new. Theorems 8.2.1 and 8.2.2 are adapted from the work of Lakshmikantham [l, 61, whereas Theorem 8.2.3 is based on the result of Driver [3]. See also Krasovskii [2, 51. Theorems 8.2.4 and 8.2.5 are new. T h e examples in Sect. 8.2 are taken from Lakshmikantham [6] and Driver [3]. All the results of Sect. 8.3 are based on the work of Lakshmikantham [6], whereas Theorem 8.3.4 is new. Section 8.4 contains new results. See also Halanay [22] for particular cases. T h e results of Sects. 8.5 and 8.6 are new. For many similar results for delay-differential equations, see Oguztiireli [I]. For related work, see Driver [3], El’sgol’ts [4], Krasovskii [l-51, Lakshmikantham [I], Oguztoreli [l], and Razumikhin [2, 61. For the use of vector Lyapunov functions in studying the conditional stability criteria of invariant sets, see Lakshmikantham and Leela [2].
PARTIAL DIFFERENTIAL EQUATIONS
This page intentionally left blank
Chapter 9
9 .O. Introduction T hi s chapter is devoted to the study of partial differential inequalities of first order. We consider some basic theorems on partial differential inequalities, discuss a variety of comparison results, and obtain a priori bounds of solutions of partial differential equations of first order in terms of solutions of ordinary differential equations as well as solutions of auxiliary partial differential equations. We also treat the uniqueness problem, error estimation of approximate solutions, and simple stability criteria. We make use of Lyapunov-like functions to derive sufficient conditions for stability behavior. For systems of partial differential inequalities of more general type, we merely indicate certain analogous results.
9.1. Partial differential inequalities of first order We shall use the well-known notation am at
mi=-,
m, =
am
ax
-
wp a2m
mxg=
whenever convenient. It is, however, necessary to caution the reader not to confuse the symbol m l with the one used while considering functional differential systems. Let 01, ,l3 E C [ J ,R ] , and suppose that a ( t ) < P(t), t E J. Assume that a’(t), P ’ ( t ) exist and are continuous on J . For to E J , we define the following sets: 113
114
CHAPTER
9
E
=
[ ( t ,x): to
< t < co,.(t) < x < P ( t ) ] ,
E, aE,
=
[ ( t ,2 ) : t ,
< t < co,.(t) < x < P ( t ) ] ,
=
[ ( t ,x): t
aE,
== [ ( t ,x):
aE,
=
t o , &(to) < x < &to)], < t < 03,x = &(t)],
=
to
[ ( t , %): t ,
Let us first prove a lemma that is very useful in the subsequent discussion.
LEMMA 9.1.1.
Assume that
(i) m E C [ E ,R ] , and m(t, x) possesses partial derivatives (not necessarily continuous) on E and total derivative on aE, u aE, ; (ii) if ( t l , xl)E E, , " ( t 1 , xl) = m,(t, , xl) = 0, then m,(tl , xl) < 0; (iii) if ( t , , xl)E a E 2 , m(t, , xl) = 0, and m,(t, , x,) 0, then
<
mdt, , XI)
(iv) if ( t , , xl)
E
9
< 0;
Xl)
aE, , m(t, , xl) = 0, and m , ( t l , x,) 3 0, then m d t , x1) 7
(v) m(t, x)
+ 4 t l ) %(t,
+
P'(t1)
mdt, , x1)
< 0;
< 0 on aE, .
Under these assumptions, m(t,x)
<0
on
(9.1.1)
E.
Proof. Suppose that the inequality (9.1.1) is false. Then, by assumption (v) and the continuity of m(t, x) on E, it follows that there exists a point ( t l , x,) E E , t , > t, , such that m ( t , , x,) = 0
Evidently, m(t, , x) situations arises:
and
m(t, x)
< 0,
t
< t, ,
( t , x) E E .
< 0 for (tl , x) E E , and, hence, one of the following <
, XI) E aJ% and m,(t,, XI) 0; (b) ( t l , x1) E and m,(t,, XI) 3 0; (c) ( t l , xl) 6 aE, u aE, and m,(t, , x,) = 0. (a) (tl
(9.1.2)
Consider the case (a). Because of condition (iii), we have
9.1.
115
PARTIAL DIFFERENTIAL INEQUALITIES OF FIRST ORDER
On the other hand, it turns out, by the relation (9.1.2), that
which is absurd in view of (9.1.3). Similarly, in the case (b), we obtain
because of condition (iv). Again, we get a contradiction to (9.1.4), since the relation (9.1.2) also yields
I n the last case (c), the contradiction results from assumption (ii). This proves the truth of the inequality (9.1.1). REMARK9.1.1. Notice that it is necessary for the proof only that the derivatives mt , m, or the total derivative exist at such points (t, x) for which m(t, x) = 0. We shall now prove some basic results in partial differential inequalities of the first order.
THEOREM 9.1.1. L e t f E C [ E x R x R, R ] , u, ‘u E C [ E ,R ] , and suppose that the following conditions hold: (ao) u, ‘u posses derivatives u L,‘ut ,u, ,u‘ , on E (not necessarily continuous), and the total derivative on aE2 u aE, ;
(ad
<
f(t, x, u, %), v t > f(t, x, v, v,) Ut
on
E;
(a2)f ( t , x,z , p ) - f ( t , x,z , q) 3 - a ’ ( t ) ( p - q), p (a31 f ( t , x,.,p) - f ( t , x,z , 4 ) -P’(t)(p - q), P (a4) u(t, x) < v(t, x) on aE, .
<
Then, u(t,x)
on
3 q, on aE2 ; 3 9, on aE3 ;
E.
Proof. It will be shown that the function m(t, x) defined by m(t, x) = u(t, x)
-
n(t, x)
(9.1.5)
116
9
CHAPTER
verifies the hypotheses of Lemma 9.1.1. Clearly, assumptions (i) and (v) are satisfied in view of (ao)and ( a 4 )of the theorem. Furthermore, m(t, x) fulfills condition (iv) of Lemma 9.1 . l . For, let ( t l , xl)E aE, ,m(t, ,xl)= 0, and m,(t, , xl) 3 0. Then, we have
4f, 4
“(t, , x1)
?
%(tl , x1)
and
2 %(tl , 4.
Consequently, by the condition (a,), we derive f(tl x1 1 4 t l
<
>
-f(h
uz(t1 , x1))
.%)Y
-P’(fJu,(t,
x1) -
9
%it, 7
>
x1 9 4 t l > 1 XI), t i%
%)I.
>
4)
I t now results, using the inequalities (a,), that 4 f l
?
x1) -
vdt,
9
XI)
<
-P’(t1>[uz(t,
, x1)
- %(tl
,%)I,
which implies that %(tl
?
x1)
i
P’(t1) %(tl
> Xl)
< 0.
This verifies condition (iv) of Lemma 9.1.1. T h e remaining assumptions may be similarly checked, and the conclusion (9.1.5) is immediate by the application of Lemma 9.1.1. T h e proof is complete. I t is clear, from Theorem 9.1.1, that one of the inequalities (al)must be strict for the validity of the claim (9.1.5). T h i s can, however, be dispensed with, if the function f satisfies a uniqueness condition. T o that effect, we present the following.
THEOREM 9.1.2. Let the assumptions of Theorem 9.1.1 hold, except that the differential inequalities (al) are replaced by
Suppose further that f(t, S , ui
>
P) p f ( t , X,
uz
>
P)
< g ( t , ui 4,
~1
> ~2
9
(9.1.6)
where g E C [ J x R, ;R ] , and the maximal solution r(t, to , 0 ) of y’
= g(t,y),
At”) = 0
(9. I .7)
is identically zero. Then, u(t, x)
< v(t,x)
on
i3E,
(9.1.8)
9.1.
117
PARTIAL DIFFERENTIAL INEQUALITIES OF FIRST ORDER
implies u(t, x)
< v(t, x)
Proof. Consider the solutions y(t, differential equation Y’
for all sufficiently small
+
= &,Y) E
on
=
€)
Y(Q
E,
> 0, which
u(t, x,
= y ( t , t,
E)
=
exist for t
u(t, x)
-
(9.1.9)
E.
, 0,
E)
of the ordinary (9.1.10)
E,
2 t, . Define (9.1.1 1)
y ( t , €).
Then, because of the relation
Y V ?4 > g(t, Y ( 4 E ) ) ,
we obtain x,
€1 = u,(4 x)
> 0, it follows,
Y’(t, .)
44 4,U , ( t , 4)- g(t, Y ( 4
< f(t,
Since y(t, e )
-
x>
€))-
(9.1.12)
from (9.1.1l), that
u(t, x,€1 < 44 x),
and hence, by (9.1.6), there results the inequality f ( t , x, u ( t , X ) > % ( t ,
4)-f ( 4 X? 44 x, .),
in view of the fact that u,(t, x, E) yields a&, x,
€1
= u,(t,
%44 x,
6))
d g(4 r(t,€>>,
x). This, together with (9.1.12),
x, u(t, x,
€1, %(C
x,
€1).
Moreover, we have also the inequality u(t, x,
6)
< v(t, x)
on
aE,.
Hence, an application of Theorem 9.1.1 shows that u(t, x,
E)
< v(t, x)
on
E.
Since, by assumption, lim,,,y(t, E) = 0, the stated result (9.1.9) is an immediate consequence of (9.1.11). Th is completes the proof. Consider the s’pecial case f ( t , x, u, u,)
= F ( t , x,
u ) - A(t, x) u,
.
Let a ( t ) , P ( t ) be any functions satisfying the ordinary differential inequalities
118
CHAPTER
9
such that a ( t ) < P(t). Suppose that we define the set E by means of these functions a ( t ) , /3(t). Then, we have the following corollaries. COROLLARY 9.1.1. Let F E C [ E x R, R ] ,X E C [ J x R, R ] ,u, E C[E,R], and the assumptions (ao), ( a l ) , and ( a 4 ) of Theorem 9.1.1 hold with f = F ( t , x, u) - X ( t , x) u, . Then, (9.1.5) is valid. COROLLARY 9.1.2.
Suppose that
F ( t , -r, %) - F ( 4 x, u2)
< g(4 u1
- uz),
u1
> u2,
where g(t, u ) satisfies the same assumptions of Theorem 9.1.2. Assume further that the hypotheses of Corollary 9.1.1 hold except that the inequalities (al) are replaced by (-al*) of Theorem 9.1.2. Then, (9.1.8) implies (9.1.9). For the proof of the foregoing corollaries, it is sufficient to show that the conditions (a,), ( a 3 ) of Theorem 9.1.1 are fulfilled. Verification of this fact is trivial in view of the choice of functions a(t) and /3(t). Suppose, on the other hand, we assume that If(4
-r,
u, P) - f ( 4 x, u, 411
-
q I.
(9.1.13)
+
Choose n(t) a - L(t - to), P ( t ) = --a L(t - to), and y < aL. Let to t < to y . Then, we may define the set E, as before, by means of these functions and the time interval. When (9.1.13) is verified, it can be easily checked that f satisfies the conditions (a2),(a3)of Theorem 9.1.1. From this observation stem the following corollaries.
<
+
:
COROLLARY 9.1.3. Let the hypotheses of Theorem 9.1.1 hold except the conditions ( a 2 ) and (a3). If f verifies the inequality (9.1.13), the conclusion (9.1.5) is valid. COROLLARY 9.1.4. Let the hypotheses of Theorem 9.1.2 hold except the conditions ( a 2 ) and (a3). I f f verifies the condition (9.1.13), then (9.1 .S) implies (9.1.9). 9.2. Comparison theorems
I n dealing with the applications of ordinary differential inequalities to partial differential equations, we have to estimate the solutions of such equations, which are functions of several variables, by functions of one variable. While doing so, the following lemmas prove to be useful.
9.2.
119
COMPARISON THEOREMS
LEMMA 9.2.1. Let G be a bounded open set G C R". Let u E C [ f x G, R], and let M ( t ) = max,.c u(t, x). If M(t,) = u(t, , x,) and au(t, , xo)/at exists, for (to, x,,)E (0, co) x G, then
< au(t0, x,>/at.
D-m(t,)
Proof.
(9.2.1)
Choose a sequence {t,}, so that tk < to , t, + to as k + co, and (9.2.2)
For k sufficiently large, we have (9.2.3)
On the other hand, by the definition of M ( t ) and the fact that M(t,,) = u(t, , x,,), for k sufficiently large, it follows that
which, on account of the relations (9.2.2) and (9.2.3), yields (9.2.1).
LEMMA 9.2.2. Let G be a bounded open set such that G C R n and C[J x G, R].Let
uE
w ( t ) = max I u(t, x)I, XtG
M ( t ) = max u(t, x), XEG
and N ( t ) = maZ[-u(t, x)]. X€G
Let ( t o ,xo)E (0, m) x G. Then,
(1") w(t,) > 0 implies either w(t,) = M(to)or w(to) = N(t,,); (2") eu(to) > 0 and w(t,) = M(t,) implies D-w(t,) < D-M(t,,); (3") w(to) 0 and w(t,,) = N(t,,) implies D-w(t,) D-N(t,,).
<
>0
Proof. Suppose that w(t,) the inequality U(to
, x)
< I u(to
9
.)I
and w(t,,)
=
u(t, , x,,). It follows from
< w(to) == ~ ( t,,xo),
x E G,
I20
9
CHAPTER
, If, on the other hand, we suppose that that M(to)= ~ ( t , xo). w(to) = -u(t0, xo) > 0, the inequality -u(t,
7
x)
< I u(t0
3
)I.
<
to)
=
to, xg),
x
E G,
shows that N ( t o )= - u ( t 0 , x,,). Assume now that w(to)= M(to),and choose a sequence {tk} such that t, < to , t, -+to as K + co, and
Since we have M(t,)
< +k)
and
M(t0) = w(Q,
there results the inequality
<W
-4 t O )
4 t k )
t,
-
k ) - M(t0)
1 ,
I
tk -
and, consequently, it follows that D-w(t,)
< D-M(t,).
T h e proof for the case (3") is analogous. We shall prove the following simple comparison theorem.
THEOREM 9.2.1. Suppose that continuous on Q, where
uE
C[Q,R],&/at, auldx exist and are
a = [ ( t , x): to < t < to + a, I x I < /3 and /3
> Ma.
< M I W a x I + g(t, I u I)
+
where g E C [ [ t o to , a ) x R, , R,] r ( t , t, ,yo)is the maximal solution of Y'
Y(t0) = Yo
=dt,Y),
9
)I.
for
I 4 4 41 < r(t, t o Yo) 3
on
Q,
(9.2.4)
and g ( t , 0) = 0. Assume that
< t < to + a. Then, I+,
implies that
M ( t - to)],
Let
1Wat I
existing on to
-
>0
Ix I on
(9.2.5)
(9.2.6)
Q-
(9.2.7)
9.2. Proof.
COMPARISON THEOREMS
121
Consider the function m(t) =
max
1x1<&M(t-t,)
1 u(t, x)I,
and assume that r ( 4 to ,Yo)
<4 t ) .
This means that t > to and m(t) > 0. Moreover, m ( t ) = u(t, xo). T h e point ( t , xo) may be an interior point of 9 or x,, = & [ p - M ( t - to)]. Suppose that (t,xo)E int Q. Then au(t, x,,)/ax = 0, and, as a consequence, by (9.2.4) and Lemmas 9.2. and 9.2.2, we get
Suppose now that xo = fi - M ( t - to).Then, there are two possibilities: (a) 4 t ) = u(t, B (b) m(t) = -u(t,
-
M ( t - to)), M ( t - to)).
-
Consider the case (a). If x is sufficiently close to xo = p - M ( t - to), u(t, x) ,< u ( t o ,x,,), and, therefore, we get
B
-
w
ax
-
to))
>, 0.
As a result, we obtain the inequality (9.2.8)
at ( t , x,,). Furthermore,
U(S,
fi - M ( s - to))< m(s),s < t, and
Hence, using (9.2.8), we deduce that
< I-Mat I ax aU
aU
122
9
CHAPTER
Suppose that the possibility (b) holds. Notice that -u(t, /3
Hence, - u ( t , x)
< -u(t,
-
M(t
p
-
-
to)) = max 1 u(t, .)I.
M ( t - to)),and, thus,
Since
we get (9.2.9)
at
(2,
xo). On the other hand, --u(s,
p
-
M(s - to)) < m(s),
s
< t,
and -u(t, /3
-
M(t
-
to))= m(t),
which implies that D-m(t)
<
au
--
at
+ M ax -
(9.2.10)
at ( t , xo). By the relations (9.2.9) and (9.2.10) and the fact au
au
-+M-< at
ax
I at I au
-
au ax
+M-
results the inequality D - 4
< At, m(t)).
T h e proof for the case xo = - [ p - M ( t - to)] is similar. Thus, we have shown that the inequality r ( t , to , y o ) < m(t) implies that D-m(t)
< g(t, m(t>).
<
This yields, by Theorem 1.4.1, that m ( t ) r(t, to ,yo), contradicting our supposition. Hence (9.2.7) is true, and the theorem is proved.
We shall next prove a more general result whose proof is similar to the one just presented.
9.2.
123
COMPARISON THEOREMS
THEOREM 9.2.2. Letf E C [ E x R, x R, , R,], u E C [ E ,R],u possess partial derivatives on E and total derivative on aE2 u aE, and on E, (9.2.11)
Assume further that aE, , (9.2.12) aE, , (9.2.13) (9.2.14)
(9.2.15)
where r(t, t o , yo) is the maximal solution of (9.2.5), existing for t
> to .
Proof. Let us define m(t) = maxm(t)GzGo(t) I u(t, .)I and suppose that r(t, t o ,y o ) < m(t). This implies, as before, t > t o , m ( t ) > 0 for t > t o , and there is an xo for each t such that m(t) = I u(t, xo)l. Suppose that the point ( t , xo) E E, . Then, au(t, xo)/ax = 0, and, as a result, D-m(t)
<
1 g 1 Gf(4
xo
9
I u(t, X0)L I 4, X0)l)
>
X0)L
0)
XoN
&l
= At, m ( t ) ) ,
by (9.2.11), (9.2.14), and Lemmas 9.2.1 and 9.2.2. Suppose that ( t , xo) E aE, , t > t o . Then, there are two possibilities: (a) (b)
4 4 = u(t, B(t)), Nt)= - 4 t , &I).
Let us consider the case (a). We have u(t, 4
d 4 4 xo),
124
CHAPTER
if x is sufficiently close to x,
= P(t),
9
and, therefore,
As a result, it follows that
Also, for
s
< t, u(s, P(s))
< m(s)
I u(t, B(t))l
and
=
m(t).
Hence, we deduce, using (9.2.12), (9.2.13), and (9.2.14), that D_m(t)
< D-
I u(t, B ( W
If (b) holds, we notice that -4t,
and hence au(t, “,,)/ax
4
< -44 P(%
< 0. Consequently,
On the other hand, -u(s, P(s))
< wz(s),
s
< t,
and
-u(t, ~ ( i ) =.m(t). )
This implies, as before, using (9.2.1 1), (9.2.12), and (9.2.14), that D-m(l)
<
au
- - - f3’(t) -
at
au ax
:1 I
<
aU
- - P‘(t) ax
9.2.
125
COMPARISON THEOREMS
It therefore follows that, in any case, the assumption r(t, to , y o ) < m ( t ) implies the differential inequality D-m(t) ,< g(t, 49).
T h e proof for the case ( t ,x,,) E aE, , t > to , is similar. By Theorem 1.4.1, we have m ( t ) r(t, t,, ,yo), which contradicts the assumption m(t) > r(t, to , y o ) , Hence, the conclusion (9.2.15) is valid, and the proof is complete.
<
Theorem 9.2.2 may be proved by reducing it to Lemma 9.1.1 as follows.
Secondproof of Theorem 9.2.2. Let y ( t , c) = y(t, to ,yo , c) be a solution of (9.2.16) Y' = g(t,y) c, y(t0) = yo E,
+
for sufficiently small
E
+
> 0. It is then
enough to show that
< Y ( t ,4
1 4 4 .)I
on
E,
since lime+,, y(t, e ) = r(t, to ,yo). Let us first show that
u(t, x)
on
c)
E.
Set m(4 4 = 44 -4 - Y ( t , 4
so that
m(t,x)
<0
on
Suppose that (tl , xl)E aE, , m(tl , xl) implies 4 t l , .1) = Y(t1 , 4
aE,.
= 0,
and
>0
and m,(tl , xl) >, 0, which %(tl,
x1)
3 0.
Hence, using (9.2.11), (9.2.13), and (9.2.14), we arrive at the inequality mdt1
?
Xl) = U d t l p.1)
G f(tl > x1
YYtl
~
9
4tl
,<
-S'(t1)
=
-S'(t1) %(tl
mdt1 9 x1)
> 21)) -
.1>?
, x1)
%,ti
which leads to
9
> c)
, x1)
g(t1 > YO1 >
c
~
~
+ S'(t1)
€7
>
x1)
< 0.
6))
~
6
126
CHAPTER
9
It turns out that we have so far verified conditions (i), (v), and (iv) of Lemma 9.1.1. T h e remaining assumptions of Lemma 9.1.1 may easily be checked. This implies, by Lemma 9.1.1, that u(t, x)
< y(t,E )
on
(9.2.17)
E.
+
Defining n(t, x) = - [ y ( t , E) u(t, x)] and proceeding in a similar way, we can show that n(t, x) satisfies the hypotheses of Lemma 9.1.1. Consequently, we obtain - u ( f , x)
< y(t,E )
on
E.
(9.2.18)
T h e relations (9.2.17) and (9.2.18) together imply that
I 44 .)I < Y ( t , .)
on
E,
and this proves the theorem, as observed earlier. Finally, we state the following useful comparison theorem.
THEOREM 9.2.3. Let m E C[E,R + ] , ~CE[ E x R, x R, R ] , satisfying the following assumptions: (i) m(t, x) possesses continuous partial derivatives ut , u, on E and total derivative on aE, u aE, ; (ii) for ( t ,x) E E, (9.2.19)
(iii) the relations (9.2.12), (9.2.13), and (9.2.14) hold. Then, the inequality m(t, x) 4 44
< y o on aEl implies
< r(4 t o , yo)
on
E,
where r ( t , to ,y o ) is the maximal solution of (9.2.5). Proof.
Consider the function v(4 x)
=
m ( 4 x)
-
Y ( 4 4,
where y(t, E) is any solution of (9.2.16) for small E > 0. We wish to show that v ( t , x) satisfies the conditions of Lemma 9.1.1. Clearly, assumptions (i) and (v) of Lemma 9.1.1 hold. Suppose that ( t l , xl>E aE, , n(tl , xl)= 0, and nz(t, , xl) 0, which assures that
9.3.
127
UPPER BOUNDS
m,(t, , xl) >, 0. Hence it follows, in view of the assumptions (9.2.13), (9.2.14), and (9.2.19), that
744 , x1)
Y’(4 4 x1 > 4 4 > X 1 h %it1 < -B’(Q %it, ,x1) - 6 = -P’(h> 4 1 , x1) - E,
=4
1
9
Xl) -
7
>
- g(t1 >
x*))
Y(tl
%
€1) -
from which we infer the inequality 4 t l 9 x1>
+ B’(tl>%(tl
9
x1)
< 0.
This verifies condition (iv) of Lemma 9.1.1. Since it is easy to verify the other conditions of Lemma 9.1.1,the proof of the theorem is complete by the application of Lemma 9.1.1.
COROLLARY 9.2.1. Under the assumptions of Theorem 9.2.3, if m(t, x) 0 on aE, and if the maximal solution r(t, t o , 0) is assumed to 0 on E. be the trivial solution of (9.2.5), then m(t, x)
<
<
9.3. Upper bounds
I n this section, we shall obtain upper bounds of solutions of partial differential equations of first order: Ut =
(9.3.1)
f(4 x, U, %).
DEFINITION9.3.1. Any function u(t, x) is said to be a solution of (9.3.1) if the following conditions hold: (i) u E C[E,R], u(t, x) possesses continuous partial derivatives on E, and total derivative on aE, u aE, ; (ii) u(t, , x) = +(x), where $(x) is continuous on ..(to) x P(t,,); (iii) u(t, x) satisfies the equation (9.3.1) for ( t , x) E Eo .
< <
The following theorem gives an estimate on the growth of solutions of Eq. (9.3.1).
128
CHAPTER
Assume that f(4
9
< g(t, u),
x, u, 0 )
(9.3.4)
2 0,
u
where g E C[J x R, , R] and g(t, 0) = 0. Suppose that r(t) = r(t, to , y o ) is the maximal solution of (9.2.5) existing for t >, t o . If u(t, x) is any solution of (9.3.1) such that
< Yo
l4(x)I we have
on
< r(t)
I u(t, x)i
>
(9.3.5)
E.
(9.3.6)
aEi
on
Proof. Let u(t, x) be any solution of (9.3.1) satisfying (9.3.5), and let y ( t , e ) = y ( t , t o ,y o , e ) be a solution of (9.2.16) for small E > 0. Since lim,,,y(t, c) = r ( t ) , it turns out that it is enough to show that 1 4 4 41 < Y ( 4 4
on
(9.3.7)
E,
to prove (9.3.6). In fact, we prove -u(t, x)
< y(t, 6)
and
u(t, x)
< y ( t , c)
on
E,
so that (9.3.7) holds. First of all, consider the function m ( t , x)
=
u(t, x) - y ( t ,
€).
It is clear that conditions (i) and (v) of Lemma 9.1.1 are satisfied. Let us suppose that (tl , xl) E
aE, ,
m(tl , xl)
=
0,
and
m,(tl
, xl)
0.
(9.3.8)
Hence, it follows, on account of (9.3.3) and (9.3.4), that m*(t,, x1)
x1)
= Udtl =f(t1
Y
x1
-
Y’(h > .)
44
!
x1>, %(tl ? x1))
<
pP‘(t1)
%it, x1) - €
=
-P’(Q
m,(t,
- g(t1 >
Y(tl
>
4-E
9
9
.1>
- 6.
Consequently, we have
4 4
> XI)
+ P’(t1)
%(tl
> XI)
<
0 7
verifying condition (iv) of Lemma 9.1.1. T h e verification of the remaining assumptions is similar. Thus, by Lemma 9.1.1, u(t, x)
< y ( t , C)
on
E.
9.3.
129
UPPER BOUNDS
We consider next the function
4,4 = -[Y(4 4
+
UP, XI19
and, proceeding as before, it can be shown that m(t, x) satisfies the hypotheses of Lemma 9.1.1, so that
< y(t, E )
-u(t, x)
on
E.
T h e proof is therefore complete, as observed earlier.
COROLLARY 9.3.1. T h e conclusion of Theorem 9.3.1 remains true even if we replace (9.3.4) by
If@,
< g@,
0)l
x,
I I).
THEOREM 9.3.2. Suppose that, instead of (9.3.4), we have the weaker assumption
f(4
3,
u, 0 )
< g(t7
4
(9.3.9)
7
for u E Q, where SZ is defined by Q
= [u: r ( t )
< u < r ( t ) + E o , t >, to]
for some e0 > 0. Then, the assertion of Theorem 9.3.1 is valid.
Proof. Let [ t o , TI be any given compact interval. Then, by Lemma 1.3.1, the maximal solution r(t, e) = r(t, t o ,y o , E ) of (9.2.16) exists on [ t o , TI, for all sufficiently small E > 0 and lim r(t, E ) = r ( t ) , €4
uniformly on [ t o , TI. As a result, there exists an E* > 0 such that r(t, E ) < r(t) c0 for t E [ t o , TI. Furthermore, we have, by Theorem 1.2.1,
+
which implies the inequality r ( t ) < r ( t , .)
< r(t)
+
t
€0,
E
(9.3.10)
[ t o , TI.
T o prove (9.3.6), it is sufficient to show that
I aft, )I.
< r(t, c )
on
E.
We proceed to show, as in Theorem 9.3.1, that u(t, x)
< r(t, c )
and
-u(t,
x)
< r(t, c )
on
E.
130
CHAPTER
9
Following the proof of Theorem 9.3.1, we wish to verify the assumptions of Lemma 9.1.1. I n verifying assumption (iv) for m(t, x) = u(t, x)
-
r(t, 6 ) ,
we arrive at the step u(tl
,
=
~ ( t,,c)
uz(t, , xl)
and
2 0,
analogous to (9.3.8). This, together with (9.3.10), leads to the fact u = ~ ( t, ,xl) E Q, and, hence, we use the weaker assumption (9.3.9) to show that mdt,
4 -tP'(t1) %At, , x1) < 0,
7
as before. With this change in the argument, the rest of the proof is similar. Instead of the maximal solution of (9.2.5), if we know a positive solution of the differential inequality Y'
R(CY),
Y(t0) = Yo
> 0,
(9.3.1 1)
it is easy to prove the following result.
THEOREM 9.3.3. Let f~ C [ E x R x R, R], and let the boundary conditions (9.3.2) and (9.3.3) hold. Assume that y ( t ) > 0 is a solution of (9.3.1 1). Suppose that f ( t 7
x, u, 0)
< g(t, u),
(9.3.12)
for u E Q", where L?*
2 to].
[u: u = y ( t ) ,t
Then, (9.3.5) implies
I u(t, x)l < y ( t ) Proof.
on
E.
All that is necessary to show is that u(t, x) iy ( t )
and
-u(t, x)
on
While verifying the assumptions of Lemma 9.1.1, with
E.
9.3.
131
UPPER BOUNDS
This implies that u = u(t, , xl)E Q*, and, hence, we can use (9.3.12) to verify the hypotheses of Lemma 9.1.1. With this observation, the rest of the proof runs similar to the proof of Theorem 9.3.1. T h e next theorem allows us to estimate the difference of any solution
of (9.3.1) and any solution of
ZIt
THEOREM 9.3.4. Letf, F
E
(9.3.13)
= F ( t , x, ZI, ZIJ.
C [ E x R x R, R], and
whereg E C [E x R, , R,] andg(t, 0) = 0. Assume that r(t) = r ( t , to , y o ) is the maximal solution of (9.2.5) existing for t 3 t o . Suppose that
and
Let u(t, x), v(t, x) be any two solutions of (9.3.1) and (9.3.13), respectively, such that u ( t o ,x) = +(x), v ( t o ,x) = $(x), on aE, , and
Under these assumptions,
I u ( f ,x) - v(t, x)I
< r(t)
on
E.
T h e proof is very much the same as the proof of Theorem 9.3.1.
COROLLARY 9.3.2. T h e function F ( t , x, u, u,) = f ( t , x,u, u,) is admissible in Theorem 9.3.4 to yield an estimate for the difference of any two solutions of (9.3.1). I n fact, instead of (9.3.14),
is sufficient. We shall next consider a priori bounds for solutions of (9.3.1) in terms of solutions of another partial differential equation. Sometimes, this has an advantage of obtaining sharper bounds.
132
CIIAPTER
THEOREM 9.3.5.
Let f E C [ B x R
9
x R, R], and
I f(t>2 , 1 4 %)I cz G(t,x, I I, I u, I), where G E C [ B x R, x R+, R , ] .Suppose that G(t, v, z,P)
G(t, v,
3 -u’(t)(p
2, q)
-
p 3 q,
q),
on
aE,,
(9.3.15)
and
where g E C [ J x R , , R ] , and the maximal solution r(t, t o , 0) of (9.1.7) is identically zero. Let u ( t , x) be a solution of (9.3.1) and x ( t , x) 3 0 be a solution of Zt =
(9.3.18)
G(t,x,Z, %),
such that zs(t, x) >, 0 and
< z(t, x)
1 u(t, x)l
on
aE,
Under these assumptions, we have j u ( t , )I.
Pmof.
<<
x ( t , x)
on
E.
Let u(t, x) be any solution of (9.3.1). We shall show that
I u ( t , x)I
--I z ( t , x)
+ y(t,e)
on
E,
where y ( t , E ) y ( t , to , 0, c) is any solution of (9.1.10), for sufficiently small E > 0. First of all, define the function ~
nz(t, x) = u(t, x)
-
z(t, x) ~- y ( t , c ) .
Evidently, m(t, x) 0 on aE, . Suppose that ( t l , xl) and m,(tl , xl) 3 0. We then derive that and
“(4 , %)
~
%(tl
Z(t,
,4
E
, x1) = Y(tl , 4 > 0
> z,(h , 4.
aE, , m ( t , , xl)
= 0,
9.3.
I33
UPPER BOUNDS
It therefore follows that
verifying condition (iv) of Lemma 9.1.1. T h e other assumptions of Lemma 9.1.1 may be checked similarly. Thus, by Lemma 9.1.1, we have
O n the other hand, defining
it is easy to show that n(t, x) satisfies the hypotheses of Lemma 9.1.1, and, consequently, -u(t, x)
< z(t,x) + y(t,
on
C)
E.
It therefore turns out that Iu(4 I).
since lim,,,y(t,
e)
<
4
on
E,
= 0. T h e proof is complete.
T h e following theorem may be proved, using the arguments similar to the preceding one with necessary changes.
THEOREM 9.3.6. Let f,F
E
C [ E x R x R, R ] , and
lf(4 x,u, us) - q t , x,a,%)I
< G(t,x, I u
a
I, I us
- us
I),
where G E C [ E x R, x R, , R,] satisfies the conditions (9.3.15), (9.3.16), and (9.3.17) of Theorem 9.3.5. Suppose that the maximal solution r ( t , t o , 0) of (9.1.7) is identically zero. Let u(t, x), v(t, x) be any
134
CHAPTER
9
solutions of (9.3.1) and (9.3.13), respectively. Then, if z(t, x) >, 0 is the solution of (9.3.18) such that
1 u(t, .Y) we have
-
1 u(t, x)
v(t, x)I
-
on
S z(t, X)
< z(t,x)
v(t, x)i
aE, ,
on
E.
9.4. Approximate solutions and uniqueness We shall consider the partial differential inequality
< 6.
I ut -f(4 x,u, u,)l
(9.4.1)
DEFINITION 9.4.1. A function u(t, x,6) is aid to be a &approximate solution of (9.3.1) if (i) u E C [ E ,K],and u(t, x, 6) possesses continuous partial derivatives LJ aE, ; /3(to); (ii) u(to, x,6) = (b(x), where +(x) is continuous on .i(to) x (iii) u ( t , x,6) satisfies the inequality (9.4.1) on E, .
on E, and total derivative on aE,
< <
T h e following theorem estimates the error between a solution and a &approximate solution of (9.3.1).
THEOREM 9.4.1.
LetfE C [ E
f ( 4 x,u , P ) where g E C[J
-f(t,
x R x R, R ] , and satisfy
x, 'u, P )
< g(4 u
-
(9.4.2)
> u,
n),
x R, , R] and g(t, 0) = 0. Assume that r(t, 6 )
=
r ( t , t o , 0,s)
is the maximal solution of Y'
existing for t
-I- 6 ,
Y(b)
=
0,
3 to , and the inequalities
f(f,x,z , P) - f ( f , f ( f , .y,
= R(4Y)
x, z , 9 )
z , P ) -f(t, x,2, 4)
2 -.'(t>(P
-< -P'(t)(P
-
-
91,
P >, 9,
on
aE,, (9.4.3)
4),
P 2 4,
on
aE,,
(9.4.4)
hold. I,et u(t, x), u(t, x,6) be a solution and a &approximate solution of (9.3.1). Then,
I u ( t , x)
-
u(t, x, S)l
< r(t, S)
on
B.
(9.4.5)
9.4. Pyoof.
APPROXIMATE SOLUTIONS AND UNIQUENESS
135
Consider the function m(t, x)
where y ( t , e )
=
4 4).
6,
c)
= y ( t , to , 0,
-
44 x, 8) - y ( 4 e ) ,
is a solution of
Y’ = g(4y)
+8 +
Y(td
6,
= e,
for small E > 0. Suppose that, for (tl ,xl) E aE, , m ( t , , xl) = 0 and , xl) 3 0. This shows that
m,(t,
> 4 t l , x1 , 8 )
4 t l 7 x1)
and
%it,
I
2 %it1
4
9
x1 , 8).
Thus, using (9.4.2) and (9.4.4), we have mt(t1
9
44
x1) = f(tl x1 7
-f(tl
f
7
+f@l 1
x1 9
P
Xl),
udt1 9 x1>>
U ( t l7 ’1
4% x1 9
UZ(tl
J
7
?
- g(t1
< -B’(t1)
,A t 1
9
%(tl
so that %(f,
I
-f@1 x1 4)- E - 6 8)
x1
Xl>
9
9
x1)
-
.1>)
61, %it, , .1>>
- f ( G , x1 > 4 t , > x1 , s>,%(tl -
9
, x1 , 8)) 9
4 t l I x1 7 8 ) , 4 1 , x1 9 8))l
€9
+ B’(tl>
,x1)
< 0.
This proves condition (iv) of Lemma 9.1.1. It is easy to show that the other assumptions of Lemma 9.1.1 also hold. Hence, by Lemma 9.1.1, ~ ( tx),
-
u(t, x,
6)
< y (t , e )
on
E.
on
E.
Proceeding similarly, we can show that ~ ( tX,, 8 )
-
u(t, x)
< y ( t ,E )
T h e estimate (9.4.5) results immediately, noting that limy(t, 6) €4
T h e proof is complete.
=
~ ( tto, 0,8).
136
9
CHAPTER
COROLLARY 9.4.1. If the function g ( t , u ) = Ku, K takes the form
1 u(t, x)
-
u(t, x,
s)i < (S/K)[exp K ( t - to) - 11
> 0,
then (9.4.5)
on
E.
We next state a uniqueness theorem of Perron type whose proof is an immediate consequence of Theorem 9.4.1 or Corollary 9.3.2.
THEOREM 9.4.2. Let f E C [ E x R x R,R] and the condition (9.4.2) hold. Assume further that the boundary conditions (9.4.3) and (9.4.4) are satisfied. If y ( t ) = 0 is the maximal solution of Y'
= g(4
Y),
Y(to) = 0,
for t >, t o , then the partial differential equation (9.3.1) admits atmost one solution.
THEOREM 9.4.3. Under the assumptions of Theorem 9.4.2, given E > 0, there exists a 8 ( ~ > ) 0 such that I d(x)
~~
$I).(
<6
on
aE,
I u(t, x)
implies
-
v(t, x)l
<E
on
E,
where u(t, x), v(t, x) are any two solutions of (9.3.1) satisfying u(t, , x) = 4 ( x ) , " ( t o ,x) = #(x), respectively, provided the solutions y ( t , t o ,yo) of (9.2.5) exist for t 3 to and are continuous with respect to the initial values (to,yo).
P ~ o o f . By Corollary 9.3.2, we have, choosing Yo
= max
%
l4(4
~
Icr(x)I,
the inequality
I 44 4 - v ( t ,41 < r ( t , t o , Yo)
on
E.
By assumption, given E > 0, there exists 8 = 8 ( ~ > ) 0 such that, if y o S, then ~ ( tt,, , y o ) < E , t 3 to , where r ( t , to , y o ) is the maximal solution of (9.2.5). T h e stated result follows at once.
9.5. Systems of partial differential inequalities of first order Many of the results obtained in the earlier sections of this chapter can be extended to more general systems. I n this section, we shall indicate some basic theorems only.
9.5.
137
SYSTEMS OF INEQUALITIES
Let 01, /3 E C [ J ,R n ] such that a ( t ) < /3(t),t E J. Assume that a’(t), ,G’(t) exist and are continuous on J . For t o e J , we may now define the sets E, E,, , aEl , aE, , and aE,, following Sect. 9.1. First of all, we shall extend the Lemma 9.1.1.
LEMMA 9.5.1.
Assume that
(i) m E C[E,RN] and m(t, x) possesses partial derivatives on E and total derivative on aE, u aE, ; (ii) for some index j , 1 \ ( j N , if (tl , xl)E E,, , mj(tl , xl) = 0, mi(t, , xl) 0, i # j , and m,J(tl , xl) = 0, then m,i(tl , xl) < 0; (iii) for some index j , 1 < j N , if (tl , xl) E aE, , d ( t , , xl) = 0, mi(tl , x,) 0, i # j , and m,j(t, , x,) 0, then
<
<
<
<
<
< < N , if
(iv) for some index j , 1 j mi(tl , xl) 0, i # j , and m,j(tl
<
( t , , xl)
, xl) 3 0, then
c n
mAt,
(v) m(t, x)
> 21)
i-
k=l
BkYtl) mk,(t,
E
aE, , mj(tl , xl) = 0,
, XI) < 0 ;
< 0 on aE, .
Then, we have m(t, x)
<0
on
E.
(9.5.1)
Proof. Suppose that the inequality (9.5.1) is false. Then, the set 2 given by Z
u N
=
[ ( t ,X)
E
E : mi(t, X)
i=l
01
is nonempty. Take the projection of Z on t-axis and denote by t, its greatest lower bound. It follows from assumption (v) that t, > t o . T h e set 2 is closed, and, hence, we conclude that there is a point ( t , , xl)E E and an index j , 1 j N , such that
< <
m y t , , xl) = 0, d ( t , x)
< 0,
t
< t, ,
and, for i # j , m y t , , x)
< 0.
( t ,x) E E,
(9.5.2)
138
CHAPTER
9
Now, one of the following situations arises:
in view of condition (iii). O n the other hand, it follows by (9.5.2) that d
m3(t1 , 4 t l ) )
2 0,
which is incompatible with (9.5.3). Following now the proof of Lemma 9.1.1 and that just given, it is easy to obtain a contradiction for the cases (b) and (c) proving that the inequality (9.5.1) is true. An extension of Theorem 9.1.1 to systems of inequalities is the following.
THEOREM 9.5.1.
the fol'
(i) z1 on aE, (ii)
Let f E C [ E x RN x Rn, R N ] ,u, v ing assumptions: ,assess
+E,;
E
C[E,R N ]satisfy
continuous partial derivatives on E and total derivatives
o n E , i = 1 , 2,..., N ; (iii) for vach i,
UtZ
G f " 4 x,U , U5z),
a?
>f " t ,
x,TI, U,Z)
9.5.
139
SYSTEMS OF INEQUALITIES
(v) u(t, x) < v(t, x) on aEl ; (vi) f ( t , x, u, p ) is quasi-monotone in u for each fixed ( t , x, p ) . These conditions imply that u(t, x)
Proof.
< v(t, x)
on
(9.5.4)
E.
Consider the function m(t, x)
=
u(t, x)
-
v(t,x).
Evidently, m ( t , x) verifies assumptions (i) and (v) of Lemma 9.5.1, on account of assumptions (i) and (v). Suppose that, for some index j , 1 <j N , ( t l , xl)E aE, , we have mj(t, , xl) = 0,mi(tl , xl) 0, i # j , and m,J(tl , xl) 3 0. This implies
5
<
(9.5.5)
Since f ( t , x,u, p ) is quasi-monotone in u, we have, because of relations (9.5.5). the ineaualitv
which implies that
140
CHAPTER
9
This verifies assumption (iv) of Lemma 9.5.1. T h e checking of the remaining conditions is similar, and the stated result follows, by Lemma 9.5.1. Making use of a similar argument, we can prove the following theorem, which generalizes Theorem 9.1.2.
TIIEOREM 9.5.2. Let the assumptions of Theorem 9.5.1 hold, except that condition (ii) is replaced by ulz
vtz
f"t,
x,u, uz",
:f z ( ( t ,x , v , azz)
on
(i L- I , 2 ,..., N ) .
E
Assume further that, for each i, f"t,
x , a, p )
x,u, p ) - f " 4
< g(4
- 4,
uz > az,
where g E C [ J x R, , R ] , and the maximal solution r(t, t o , 0) of (9.1.7) is identically zero. Then, whenever
u(t, .x)
< v(t,x)
on
E
u(t, x)
< v ( t ,x)
on
aE,.
We shall next prove a comparison theorem analogous to Theorem 9.2.3.
THEOREM 9.5.3. Let m E C [ E ,R+N], f that the following conditiQns hold:
E
C [ E x RN x Rn, RN] such
(i) m ( t , x) possesses continuous partial derivatives on E and total derivative on aE, LJ aE, ; I , 2,..., N , (ii) for ( t , x) E B and i (iii)
m t ( t , x )
f ( 4 -t", u, 0)
< !At, 4,
(9.5.6) (9.5.7)
where g E C[J x R+N,R N ] ,g ( t , 0 ) = 0, and g ( t , u ) is quasi-monotone in u for each t E (iv) for each i,
1;
9.5. Then, m(t, x)
141
SYSTEMS OF INEQUALITIES
< y o , on aEl , implies m(t, 4
< r ( t , t o , Yo)
on
B,
where r ( t , to ,yo)is the maximal solution of the differential system existing for t
Y’
3 to .
Proof. Let y(t, E ) differential system
= s(t,Y),
Y(to)
= y ( t , t o ,yo , E )
Y’
= g(t,y)
for sufficiently small vector
E
+
= Yo
Y
be any solution of the ordinary Y(td = yo
6,
+
6,
> 0. Since it is known
lim Y(4 .) t-0
=
that
r(t, t o 7 Yo),
it is sufficient to show that m ( t , x)
< y(t,
on
E)
(9.5.10)
B.
It is easy to verify that v(t,x) = m(t, x) - y ( t , E ) satisfies the hypotheses of Lemma 9.5.1. Clearly, assumptions (i) and (v) hold. Moreover, for an index j , 1 < j N , (tl , xl) E aE, , suppose that d ( t , , xl) = 0, v i ( t 1 ,xl) 0, i f j , and vE:3(tl, xl) 3 0. Then,
<
<
and
, 4 > 0,
mj(t1 , X l ) = Yqtl
mE(t, , x1)
m,J(t1 , x1)
< y“t1 ,
i
E),
#i
t 0.
Thus, using the relations (9.5.6), (9.5.7), (9.5.8), and the quasi-monotone character of g ( t , u), we get m,3(t,, x1)
-
Y
dY’P(t1
, €1
x1 m(t1 Y x1>, mz’(t1 , x1)) 9
-P(t1
, Y(tl 4) - 8 7
, x1 , m(t1 , Xl), mzj(t1 , x1)) - g3(tl , mYt1 , x1), ..., y3(tl , 6),..., mN(tl , xl)) c3 , 0) +f’(tl , x1 , m(t1 , Xl>,O)
-
7
7
-
i?(tl
c
7
9
m(t1 x1)) 9
-
8
n
<
-
=
-
k=l
Pkl(t1)
?
21)
- E3
n
1
k=l
Pk’(t1) 4 # 1
, x1)
-
2.
142
CHAPTER
9
proving that hypothesis (iv) of Lemma 9.5.1 is valid. Arguing similarly, it is easy to verify the assumptions of Lemma 9.5.1, and consequently (9.5.10) is true. This proves the theorem. Following the proofs just presented, the growth estimates of solutions and approximate solutions and a uniqueness result for the systems of the type (i = 1, 2 )...,N ) , u: f y t , x,u, u,i) 1
where f E C [ E x RN x R”, RN],may be proved. We leave it as an exercise to the reader. Finally, we shall give below a theorem on differential inequalities for overdetermined systems of inequalities. For such systems, the time variable t also may be a vector, such that t,O t, < GO, y = 1, 2,..., k. T h e sets E , E, , aE, , aE, , and aE, may be understood with this change. Then, we have
<
THEOREM 9.5.4. Let f E C [ E x RN x Rn, RN], u, v E C[E,RN] satisfy the following conditions: (i) u(t, x), v(t, x) possess continuous partial derivatives on E and total derivative on aE, v aE, ; (ii)
4,< fy((4x, u, u;),
’fy2(4 x,‘u, ‘uri), on E , i = 1, 2,..., N , y (iii) for each i and y,
1, 2,..., k ;
(iv) the functionf(t, x, u, p ) satisfies the quasi-monotone property in
u for each ( t , x,
(v) ~ ( tx),
p);
v(t, x) on aE, .
9.5.
143
SYSTEMS OF INEQUALITIES
Under these assumptions, u(t,x) < v(t,x)
Proof.
on
E.
Introduce Mayer’s transformation t,
where A, >, 0 ( y
=
=
t,O
iA,$,
1, 2,..., k). Define the functions
c(s, x,A ) E(S,
+ As, x), v(to+ As, x).
= .(to
x,A)
=
By assumption (v), it follows that Zi(0, x,A)
< qo, x, A),
and the functions u“, v” are defined on
where s is a single variable. Moreover,
c c
~,%,(tO
+
A,v,,(to
+
k
cs =
ES =
v=l
k
,=l
AS,
x),
AS,
x).
We define also the functions for each i,
so that the inequalities (ii) reduce to
cs < P ( S , x,c, C2), fjs
> Fi(s, x,fi,
on
E.
Evidently, F(s, x, z , p ) satisfies quasi-monotone property in z for each ( t , x, p ) . Furthermore, the inequalities (iii) reduce to
-w,x, z , 4 ) 3 --k 1 n
Fi(S, x,x,P)
and
.j’(t)[Pj
-
%I,
p 3 q,
on
aE,
Pj’(t)[Pj
-
4j1,
P 2 4,
on
aE,.
j=1
c n
Fi(S, x,.,P) -FFi(S,x,z, 4 ) $ --R
j=1
144
9
CHAPTER
I t then follows that the functions u", 5, and F satisfy the assumptions of Theorem 9.5.1, and, consequently,
< 5(s, x, A)
C(s, x, A)
on
E,
and, in particular, for s = 1 , 1(1,x, A)
< $(I,
x, A).
Let, now, ( t , x) be an arbitrary point, and let A
=
(t
-
t,)
=
(tl
-
tl,,
..., t ,
- t,O)
be scuh that A, >, 0. Then, on E, we have u(t, x ) = C(1, x , t
-
to) < 5(l, x,t
~
to) =
v(t, x),
which is exactly the relation we have to prove. This completes the proof. One could formulate and prove analogous results for systems of the type u;, = Fji(t, x,u, u;),
on the strength of Theorem 9.5.4. We shall not attempt such a formulation of the results.
9.6. Lyapunov-like function Let us consider a first-order partial differential system of the form U t = f(t,
x, u, uzi),
(9.6.1)
where f E C [ E x RN x R", R N ] .We wish to estimate the growth of solutions of (9.6.1) by means of a Lyapunov-like function. T o this end, we have
THEOREM 9.6. I .
Assume that
(i) V c C [ J x RN,R , ] , V(t,a ) possesses continuous partial derivatives with respect to t and the components of u, and at
zL)
i ~.f(t, x , u, uZi) a2l
< G(t,x , l,'(t, u), V,(t, u)),
where G E C [ E x R-, x R, R] and V J t , u) = (aV/aa). (&/ax);
9.6.
145
LYAPUNOV-LIKE FUNCTION
(ii) G(t,x,x , p ) - G(t,x,z , 0) 3 -a’(t)p, p 3 0, on aE, ; (iii) G(t, x,x , p ) - G(t, x,z, 0) -,B’(t)p, p 3 0, on aE, ; (iv) G(t,x,x, 0 ) d g(t, x), z > 0, where g E C [ J x R, , R]; (v) the maximal solution r ( t , to , yo) of (9.2.5) exists for t 3 to .
<
Then, any solution u(t, x) of (9.6.1) satisfying
wo
?
4 ( 4 < yo
on
(9.6.2)
aE,
allows the estimate V ( t , 4 4 4)d
t o , Yo)
on
E.
(9.6.3)
Proof. Let u(t, x) be any solution of (9.6.1) such that (9.6.2) holds. Consider the function m(t, x)
=
V(t,u(t, x).
By assump.tion (i), we have
and
< G(4 x,m(t, 4,m,(t, XI), on
m ( t 0 , x)
aE, *
I t is evident that the hypotheses of Theorem 9.2.3 are fulfilled, and, as a result, on E, V ( t >44 XI) = m ( t , 4 < r ( t , to ,yo) where r(t, to ,yo) is the maximal solution of (9.2.5). T h e proof is therefore complete.
THEOREM 9.6.2.
Assume that
(i) the assumption (i) of Theorem 9.6.1 is satisfied; (ii) G(t,x,z, p ) - G(t,x,x, Q) 3 - ~ ’ ( t ) ( p- Q), p 3 q, on aE2 ; (iii) G(t,x,z , p ) - G(t,x,z , p) -P’(t)(p - Q), p 3 4, on a - 4 ; (iv) G(t, x, z1,p ) - G(t, x,z2,P) g(t, z1 - 4,XI 3 Z Z ; (v) the maximal solution r ( t , t o ,0) of (9.1.7) is identically zero.
<
<
Then, if z(t, x) >, 0 is the solution of (9.3.18) such that x(t,, x) = #(x) 3 0 on aE, and V t , > $(XI) < ?4) on >
146
CHAPTER
we have V ( t ,u(t, x))
9
< x ( t , x)
on
E.
Proof. If u(t, x) is any solution of (9.6.1), we obtain, as in Theorem 9.6.1, the inequality am(t’ at
< G(t, X, m(t, x), m,(t, x)).
Since z(t, x) is the solution of (9.3.18), we have
As the hypotheses of Theorem 9.1.2 are satisfied, the conclusion follows immediately, and the theorem is proved. Let us now assume the existence of solutions of (9.6.1). Suppose also that the system (9.6.1) has the trivial solution u = 0. We may then formulate the definition of stability of the trivial solution of (9.6.1). DEFINITION 9.6.1. T h e trivial solution u = 0 of (9.6.1) is said to be stable if, for every E > 0 and to E J , there exists a S > 0 such that 11 +(x)ii < 6 on aE, implies
/I u (4 .)I1 < 6
on
E,
where u(t, x) is any solution of (9.6.1) with u(t, , x) = +(x) on aE, . T h e trivial solution u = 0 of (9.6.1) is said to be asymptotically stable if it is stable and, for every E > 0, to E J , there exist positive numbers 6, and T such that 11 $(x)li < 6, on 8E, implies
On the strength of Theorem 9.6.1, it is easy to state the sufficient conditions for the stability behavior of the trivial solution of (9.6.1). THEOREM 9.6.3. further that
Let the assumptions of Theorem 9.6.1 hold. Suppose b(l/ 2.1 11)
<:
q t ,4
< 41u Il),
(9.6.4)
where a, b E X . Then, the stability or asymptotic stability of the trivial solution of the ordinary differential equation (9.2.5) implies the stability or asymptotic stability of the trivial solution of the system (9.6.1).
9.6.
147
LYAPUNOV-LIKE FUNCTION
Proof. Suppose that the trivial solution of (9.2.5) is stable. Let E > 0 and to E J . Then, given b ( ~> ) 0 and to E J , there exists a 6 > 0 such that y o 6 implies
<
Y(t3 to ,Yo)
< 6,
t
2 to
(9.6.5)
7
where y(t, to ,yo ) is any solution of (9.2.5). By Theorem 9.6.1,
w,44 4)< r(t, to
I
Yo)
on
(9.6.6)
E,
for any solution u(t, x) of (9.6.1), r(t, to ,y o ) being the maximal solution of (9.2.5). Choose a positive number 6, such that a(6,) = 6, and assume that (1 d(x)ll 6, . This implies that
<
V t o>4(4
< 4 $(.)I
< 4%)= 8.
Choose yo = supzsaEIV(to,C#J(x)). It then follows, by the relations (9.6.4), (9.6.5), and (9.6.6), that
4114 4 a)<
w ,4 4 4) < r(4 to ,Yo) < &)
on
E,
which leads to a further inequality
I1 44 )1.I
<E
on
E,
<
provided II+(x)ll 6,. This proves the stability of the trivial solution of (9.6.1). Now, suppose the trivial solution of (9.2.5) is asymptotically stable. Let E > 0 and to E J- Then, given b ( ~ > ) 0 and toE J , there exist two positive numbers 6, and T such that yo 6, implies
<
to ,Yo)
< b(E),
t
+ T.
2 to
As before, we choose y o = supzsaE, V(to, +(x)). Furthermore, let ~ ( 8 , )= 8, and assume that 11 C#J(x)lI . These considerations show that, as previously,
< so
M u ( 4 41) < V(t>44 4)
< r(4 t o
9
Yo) < 44,
+ T and m ( t ) < x < ,B(t).From this follows the inequality t >, t o + T , II u(t, 1)I. < m ( t ) < x < P(t), provided 11 +(x)II < so. It is easy to see that this assures the asymptotic for t 3 to
E,
stability of the trivial solution of (9.6.11, in view of the foregoing proof. T h e theorem is completely proved.
148
CHAPTER
9
Theorem 9.6.2 may also be used to discuss stability properties of the trivial solution of (9.6.1). For this purpose, let us assume that Eq. (9.3.18) possesses the trivial solution and that all the solutions x(t, x) with ~ ( t, ,x) = $(x) 3 0 are nonnegative on E. Then, we can define stability notions with respect to the trivial solution of (9.3.18), noting that all the solutions are nonnegative. DEFINITION 9.6.2. The trivial solution of (9.3.18) is stable if, for every c > 0 and to E J , there exists a 6 > 0 such that x(t, x) < t on E, provided $(x) 6 on aE, . The definition for asymptotic stability may be similarly formed.
<
THEOREM 9.6.4. Let the assumptions of Theorem 9.6.2 hold, and let V ( t ,ZL) satisfy the inequality (9.6.4). Then, the stability or asymptotic stability of the partial differential equation (9.3.18) implies the stability or asymptotic stability of the trivial solution of the partial differential system (9.6.1). 9.7. Notes The results of Sect. 9.1 are due to Plis [6]. Lemmas 9.2.1 and 9.2.2 and Theorem 9.2.1 are adapted from Szarski [8]. Theorems 9.2.2 and 9.2.3 are new. T h e contents of Sects. 9.3 and 9.4 are modeled on the basis of the work of Plis [6] and are new. For the result of the type given in Theorem 9.5.4, see Szarski [S]. T h e other results of Sect. 9.5 are new. Section 9.6 contains new results. For further related work, see Plis [l-51 and Szarski [I-3, 6, 81.
Chapter 10
10.0. Introduction I n this chapter, we shall investigate partial differential equations of parabolic type. First of all, we shall concentrate in obtaining certain results concerning parabolic differential inequalities in bounded domains and comparison theorems connected with such inequalities. We consider different kinds of initial boundary-value problems, obtain bounds and error estimates, and prove uniqueness of solutions. Stability criteria of the steady-state solutions is discussed. Many of the results have been extended to systems of parabolic differential equations and inequalities in bounded domains. Introducing the concept of Lyapunov functions, we give sufficient conditions for stability and boundedness of various types. Criteria for conditional stability and boundedness are discussed in terms of several Lyapunov functions. Regarding the parabolic differential equations in unbounded domains, we have basic results concerning parabolic differential inequalities and uniqueness of solutions. Finally, we treat the exterior boundary-value problem. We have given uniqueness criteria only.
10.1. Parabolic differential inequalities in bounded domains Let H be a region of ( t ,x) space in Rn+l satisfying the following conditions: (i) H is open, contained in the zone to < t < co, to 3 0, and the intersection of H , the closure of H , with any zone to t T is bounded; (ii) for any t , E [ t o ,co), the projection Ptl on Rn of the intersection of R with the plane t = t , is nonempty; and for every sequence {tic} such that (iii) for every ( t l , xl) E
< <
149
150
CHAPTER
tk t [ t o ,GO), t,<- f t , as k xk t P t kand xli ---t x1 as k
-+ co,
---f
10
there is a sequence {xk} satisfying
60.
We shall denote by aH that part of the boundary of H which is contained in the zone to < t < co. It is easy to see that the topological product H = ( t o ,a)x D, where D C Rn is an open, bounded region, satisfies the preceding conditions. Also, the pyramid defined by the inequalities to
< t < t" + T ,
1 xi
-:x
1
< a, -L(t
- to),
i
=
1 , 2,..., n,
where L >, 0, ai > 0, and T = min(ui/L), verifies the requirements of H . On the other hand, H = H , u H 2 , where
< t < 1 , 0 < x < 21, 1 ,< t < 2 , 0 < x < I],
Hl
=
[(t,x): 0
Hz
=
[ ( t ,x):
does not satisfy condition (iii) at the point (1,
g).
Let us prove some basic lemmas, which we shall use frequently.
LEMMA 10.1.1. Assume that (i) m E C [ H , R ] , the partial derivatives m , , m, , mzz exist and are continuous in H ; (ii) m(t, x) 0 on P f 0u a H ; (iii) if ( t l , xl) E Ptl , m ( t , , xl) = 0, m,(t, , xl) = 0, and the quadratic form C,
for arbitrary vector A, then m,(t, , xl)
< 0.
These assumptions imply that m ( t , x) ,0
Bmf.
on
(10.1.1)
H.
Suppose that the inequality (10.1.1) is not true. Then, the set
z = [(t, ).
€
H : m(t, ). >, 01
is nonempty. Let 2, be the projection of Z on t axis and t , We thcn have m(t, x) 0 on IZ n [ t o , t l ] .
=
inf Z , .
10.1.
PARABOLIC DIFFERENTIAL INEQUALITIES
151
We assert that m(tl , x) has a maximum equal to zero for some x1 E Ptl . If this is false, on account of assumption (ii), we must have m(t, x) < 0 for all-x E Ptl . This contradicts the definition of t, . Hence, there is an x1 E int Ptl such that m(t, , xl) = 0. It therefore follows that %(tl
3
4 2 0.
(10.1.2)
Since m(t, x) attains its interior maximum at ( t l , xl), m,(t, , xl) = 0 and the quadratic form
we obtain
for an arbitrary vector A. We then deduce by condition (iii) that mt(t, , x,) < 0, which is incompatible with (10.1.2). Thus, the set 2 is empty, and the inequality (10.1.1) is true. Let m ( t , x) be continuous on aH. We shall denote by aH, that part of the boundary aH on which a(t, x) > 0, that is, aH,
=
[ ( t , X)
E
aH: a(t, X) > 01.
(10.1.3)
Let T be a direction defined at each point of aH, in a continuous manner. We shall say that a direction T at ( t l , x,) E aH, points into the interior of H , if there exists a finite ray with (tl , xl) as its origin, all of whose is interior points lie in H . For any continuous function u(t, x), ~ a directional defined as follows. If T points into the interior of H , a u / a is derivative, that is, (10.1.4)
where (tl , x,) lies on the ray issuing from ( t , x) in the direction of T, the distance from (t, , xl) to ( t , x) being 47. We shall always assume that the direction T points into the interior of H , and we shall take aul3-r to be defined in the sense of (10.1.4). If, in addition, the direction T is orthogonal to the t axis, then it is denoted by T ~ and , the corresponding directional derivative will be given by
3 4 4 X) a70
=
lim u(t, 4
Ar,,+O
-
u(t, x1)
AT^
,
where (t,xl)lies on the ray issuing from ( t , x) in the direction of T ~ the , We . shall use both types of distance between ( t , xl) and ( t , x) being h O directional derivative in the sequel.
152
CHAPTER
10
We are now in a position to prove a variant of Lemma 10.1.1.
LEMMA 10.1.2. Let the hypotheses of Lemma 10.1.1 remain the same except that condition (ii) is replaced by (iia) m ( t , x) < 0 on P l 0 and aH - aHa ; (iib) if ( t l , x,) E aH, and m ( t , , x,) = 0, then a m ( t , , xl)/a7 where, for each ( t , x) E aH, , am/aT is assumed to exist.
< 0,
Then, the conclusion (10.1.1) is valid. Proof. T h e proof is similar to the proof of Lemma 10.1.1, the only difference being that (tl , xl) cannot lie on aH, , in view of condition (iib). For, if ( t , , x,) E aH, , since m ( t , , xl) = 0, we have, by condition (iib), (10.1.5)
On the other hand, from the fact that m(t, x) deduce that
< 0 on A n [ t o ,t l ] , we
contradicting (10.1.5). Let us now consider the partial differential operator given by qu1
= Ut
-f(C x,u, u, , u,,),
where f E C [ H x R x Rn x Rn2,R ] , u, u,, = (a2u/ax12, azu/ax, ax, ,..., a2u/axn2).
=
( 10.1.6)
(au/ax,,..., au/axn), and
DEFINITION 10.1 .l. T h e function f ( t , x, u, P, R ) is said to be eZZiptic at a point (tl , xl) if, for any u,P i , Qik , Ri, (i, k = 1, 2,..., n), the quadratic form
for arbitrary vector A, implies
If the foregoing property holds for every ( t , x) E H , thenf(t, x, u, P, R ) is said to be elliptic in H. Moreover, we shall say that the differential operator T is parabolic iff ( t , x,u, P, R ) is elliptic.
10.1.
PARABOLIC DIFFERENTIAL INEQUALITIES
153
A fundamental result in parabolic differential inequalities may now be proved.
THEOREM 10.1.1. Assume that (i) u, z, E C [ H ,R],the partial derivatives u l , u, , u,, , v 1 , v, , vz, exist and are continuous in H ; (ii) f~ C[H x R x Rn x Rnz,R ] ,the differential operator T is parabolic, and T [ u ]< T[v] on H; (iii) u(t, x)
< v(t,x)
on Pt0u aH.
Under these assumptions, we have u(t, x)
< v(t, x)
on
(10.1.7)
H.
Proof. We consider the function m(t, x)
=
u(t, x)
-
v(t, x).
Clearly, m(t, x) satisfies assumptions (i), (ii) of Lemma 10.1.1. We shall show that condition (iii) of Lemma 10.1.1 is also true. Suppose that (tl , xl) E Ptl , m ( t , , xl) = 0, m,(t, , xl) = 0, and the quadratic form
5
mzizk(tl >
hXk
i,k=l
<
for arbitrary vector A. This implies that (10.1.8)
and
Consequently, using the ellipticity of the function f,we obtain At,
> x1 >
4tI
7
7
4, %(tl
x1 > 4 t l
7
x1>, U,,(tl
,Xl),
>
U d t l > Xl),
x1))
%dtl
, x1)).
This, together with the inequality T[u]< T [ v ] ,yields
154
CHAPTER
10
T h e next theorem allows nonlinear boundary conditions and is, as such, more general than Theorem 10.1.1.
THEOREM 10.1.2. Let the assumptions of Theorem 10.1.1 remain the same except that the boundary condition is replaced by (iiia) u(t, x) < v(t, x) on Ptoand aH - aH, ; (iiib) for each ( t , x) E aH, , a u / a ~av/ar , exist, and .(t, x)
where Q E C[aH, x
y+
Q(t, x, u(t, x))
< 0,
R,R].
Then the inequality (10.1.7) is true.
Proof. T h e proof is similar to the proof of Theorem 10.1.1 except that we have to verify conditions (iia), (iib) of Lemma 10.1.2. T h e definition of m(t, x) and condition (iiia) imply condition (iia). If ( t l , xl) E aH, and m ( t 1 ,xl) = 0, then it is easily checked that a m ( t l , xl)/& < 0 because of condition (iiib) and the fact that a(t, , xl) > 0 whenever ( t l , xl) E aH, . This verifies condition (iib), and therefore Lemma 10.1.2 assures the desired result. COROLLARY 10.1.1.
Assume that
(i) u E C[H,R], the partial derivatives u t , u, , u, exist and are continuous in H ; (ii) f E C[H x R x R" x R"', R],f ( t , x,0, 0, 0) 3 0, the differential operator is parabolic, and T [ u ] < 0 on H ; (iii) u(t, x) < 0 on Pt0u aH. Then, u(t, x)
<<0 on H.
COROLLARY 10.1.2. T h e Corollary 10.1.1 remains true even if condition (iii) is replaced by (iii*) u(t, x) 0 on Pt0and aH - aH, ; (iv) for each ( t , x) E aH, , au/ar exists and .(t, x)
aT
+ Q(t, x, u(t, x)) < 0
on aH, , where Q E C[aH, x R,R] and Q(t, x, 0) = 0.
10.2. COMPARISON
155
THEOREMS
10.2. Comparison theorems
A comparison theorem that plays a prominent part in applications is the following result.
THEOREM 10.2.1. Suppose that (i) m E C [ g ,R,], m(t, x) possesses continuous partial derivatives m, , m, , and m,, in H ; (ii) f E C[H x R x R" x Rnz,R ] ,the differential operator T is parabolic, and T[m] 0 on H ;
<
(iii) 6 E C [ J x R, , R ] , and r ( t ) = r ( t , t o ,yo) 3 0 is the maximal solution of Y' = A t l Y ) , Y ( t d = yo 2 0, (10.2.1) existing on [ t o ,0 0 ) ; (iv) m(t, x) r(t, to ,yo) on Pt0 u aH.
<
Then, one of the assumptions
<
( 4 f (4 x,u, 090) g(t, u), > 0, (b) f ( t , x, u, 0 , O ) < g ( t , u), r(t) < u implies that m(t, x)
Proof.
< r ( t ) + e 0 , for some c0 > 0,
< r(t, t o , Yo)
on
H.
(10.2.2)
We consider the function v(t, x) = m(t, x)
where r(t, c)
=
-
r(t, E),
r(t, to , y o , c) is the maximal solution of Y'
=g(4y)
+
c,
At,)
= yo
+
6,
for sufficiently small E > 0, which exists on any compact interval [to , to 81, 8 > 0, by Theorem 1.3.1. Also,
+
lirn r(t, c) = r(t, to ,yo). OE'
Furthermore, this implies that there exists an c0 r ( t ) < r(4 6)
< r(t>
+
€0
, [ t o , to
> 0 such that
+ PI.
(10.2.3)
We shall show that, under either one of the assumptions (a) or (b), v(t, x) satisfies the hypotheses of Lemma 10.1.1. Clearly, the conditions
I56
CHAPTER
10
of Lemma 10.1.I hold. T o verify condition (iii), let (tl , xl)E Ptl , a ( t , , xl) = 0, u,(t, , xl) = 0, and the quadratic form
for arbitrary vector A. I t therefore follows that m(t1 , x1)
= = r(t1
, €1,
mx(t, , x1) = 0,
(10.2.4)
and
Since the function f is elliptic, we obtain (10.2.5) Notice that the relations (10.2.3) and (10.2.4) yield r ( h ) < m(t1 >I.,
< dtl)
+
€0
-
Also, m(t, , x,) > 0, since ~ ( t, ,6) > 0. Thus, in any case, the assumption (a) or (b) may be used. Hence,
on account of (10.2.5) and the assumption (a) or (b). By Lemma 10.1.1, we then have v(t, .x) < O on IT, which implies that m(t, ,Y)
< r ( t , to , y o )
on
H,
since lim,,,, r ( t , 6 ) .= r ( t , to , yo).T h e proof is complete. T h e boundary condition (iv) of Theorem 10.2.1 may be replaced by nonlinear boundary conditions to achieve the same conclusion. This we state as a corollary, the proof of which may be deduced by reducing to Lemma 10.1.2.
10.2.
157
COMPARISON THEOREMS
COROLLARY 10.2.1. T h e conclusion of Theorem 10.2.1 remains valid, if, in place of boundary condition (iv), we have
<
(a) m(t, x) r ( t , t o ,yo) on P i o and aH - aH, ; (b) for each ( t ,x) E aH, , amja-,,, exists and
whereQ E C[aH, x R, ,R,], Q(t,x,u ) > 0 if u
> 0, for each ( t ,x) E aH, .
COROLLARY 10.2.2. Let assumptions (i) and (ii) of Theorem 10.2.1 hold. Suppose that g E C [ J x R, , R] and that y ( t ) > 0 is a differentiable function satisfying the differential inequality Y'(9
>g(t,y(t)),
Y(t0) = Yo
< g(4 u),
u =Y(t),
> 0.
Then, the assumption f(t, x, u, 090)
implies that m(t, x)
on
t
3 to
9
H,
provided either (a) m(t, x) < y ( t ) on Pt0u aH; or (b) m(t, x) < y ( t ) on Pt0and aH - aH, and
where Q E C[aH, x R,, R] and Q(t, x,u ) ( t , x) E aH, .
>0
if u
> 0,
for each
I n certain situations, the next theorem is more suitable in applications, since it offers a better estimate. Moreover, it shows that the strict inequality T[u]< T[v]in Theorem 10.1.1 may be relaxed i f f satisfies certain additional restrictions.
THEOREM 10.2.2. Assume that (i) m, ZI E C [ n ,R,], the partial derivatives m , , m, , m,, , u i, u, , u,, exist and are continuous in H ; (ii) f E C [ H x R, x Rn x Rnz,R ] , the differential operator T is parabolic, and T [ m ] < T[v] on H;
158
10
CHAPTER
(iii) g E C [ J x R, , R ] ,g(t, 0) = 0, the maximal solution of Y’
= g(t,y),
Y(tJ
=
(10.2.6)
0
is identically zero, and f ( t , -x,2 1 P , R ) - f ( t , X, 2 2 , P, R ) 9
(iv) m(t, x)
< g(t,
21
-
zZ),
21
> ZZ
;
(10.2.7)
< v(t, x) on PLOu aH.
Under these assumptions, m(t, x )
< v(t,x)
on
Proof. Consider the solutions y ( t , C ) differential equation Y’
for sufficiently small
E
== g
> 0.
+
(t,~)
= y ( t , to , 0, E )
=
€9
(10.2.8)
H.
of the ordinary
€9
(10.2.9)
Define the function
z(t,x)
:-=
v(t, x)
+y(t,
€).
Clearly, m ( t , x) < z(t, x) on PLO u a p . Furthermore, observing that y ( t , e ) > 0 and using the relation (10.2.7), we have
qz) = % ( t ,.x) - f ( f , x , z(t, x ) , % ( t ,x), %.dt, x ) ) = v,(t,). -f(t,
x, v(t,).
+ f ( t , .2^,
+ Y ( t ?€1, % ( t ,4,%(t, x))
v(t,x ) , v r ( t , XI1
%Z(t,
- K t , x , n(t, x), %(tr x ) , v&,
3 T[vI
+
+ Y’(L
6)
x))
4)
E.
This implies, in view of the fact that T [ v ]3 T [ m ] , the inequality T[m] T [ x ] . Hence, applying Theorem 10.1. I to functions m(t, x), z ( t , x), we obtain the relation I
,
m(t, x )
< z(t, x )
on
H.
Since, by assumption, limG+,,y ( t , c) = 0, the desired result (10.2.8) follows immediately. T h e proof is complete. 10.2.3. Let the assumptions of Theorem 10.2.2 remain the same except that boundary condition (iv) is replaced by rhEOREM
(a) m(t, x)
< v(t, x) on PtOand aH
-
aH, ;
-
10.2.
I59
COMPARISON THEOREMS
(b) for each ( t , x) E aH, , am/aTO,a v / a ~ ,exist, and
.(t, x)
+ Q(t, x, v(t, x)) > 0
870
where Q E C[aH, x R, , R],and Q is increasing in u for each ( t , x).
Proof. T h e proof is similar to the proof of Theorem 10.2.2. I n the present case, we verify the assumptions of Theorem 10.1.2. Evidently, m(t,x)
< z(t,x)
on
Pi0
and
aH
-a€€, .
T h e monotonicity of the function Q in u, together with the fact that z(t, x) > v(t, x), shows that, on aH,,
> -Q@,
x,
4 4 4).
T h e application of Theorem 10.1.2 now yields the stated result. As an immediate consequence of Theorems 10.2.2 and 10.2.3, we derive weak maximum and minimum principles.
COROLLARY 10.2.3. Assume that (i) u E C [ a ,R],u(t, x) possesses continuous partial derivatives u t , u, , and u, in El; (ii) f E C[H x R x Rn x Rn2,R ] , and the differential operator T is parabolic; (iii) f ( t , x, z, 0, 0) 0 if z > 0 and T[u] 0; (iv) either
<
<
< <
(a) u(t, x) 0 on Ptou aH; or (b) u(t, x) 0 on Pto and aH - aH, au/aTo exists and a(t, xJ
370
, and for each ( t , x) E aH, ,
+ Q(t, x, u(t, x)) < 0
on
where Q E C[aH, x R , R ] and Q(t, x, z ) is increasing in z for each ( t , x). Then we have u(t, x)
on
B.
I60
CHAPTER
10
COROLLARY 10.2.4. Let assumptions (i) and (ii) of Corollary 10.2.3 hold. Assume further that (iii*) f ( t , x,z , 0, 0) 3 0 if z (iv*) either
&/&,
< 0 and
T ( u ) 3 0;
(a) u(t, x) >, 0 on Pt0u aH; or (b) u(t, x) 3 0 on Pt0 and 8H - aH, , and for each ( t , x) E aH, , exists and on
where Q
E
C[aH, x R, R] and Q(t, x,z ) is increasing in z for each ( t , x).
Then, we have u ( t , x)
O
on
H.
Finally, we shall prove a comparison theorem that will be useful in considering the stability of steady-state solutions of nonlinear diffusion equations. Let G be an open bounded region in Rn and aG be its boundary. Denote by H the topological product [0, a)x G.
THEOREM 10.2.4. Suppose that (i) u, v E C[H, R ] ,the partial derivatives u i, v t , u, , a,, u, , v,, exist and are continuous in I€; R ] , the differential operator T is para(ii) f~ C [ H x R x Rn x R7&', bolic, q / a u exists and is continuous, and T[v] T[u]on H ; , the direction of the (iii) the derivatives h ( t , x)/~T,,, &(t, x ) / ~ T ,in outward normal to the hypersurface (0, a)x aG exist, and
<
where Fl, F, are continuous functions with closed domain and bounded derivatives, such that Fl(4 < F2(u),
if u belongs to the common domain of definition of Fland F, ;
Then, we have v(t, x)
< u(t, x)
on
H.
(10.2.10)
10.2.
161
COMPARISON THEOREMS
Proof. We divide the proof into two parts. T h e first is a proof of the theorem if the condition T [ v ] T[u] is replaced by strict inequality, that is, T[v]< T[u].We consider the function
<
m(t, x) = v(t,x)
-
u(t, x),
and proceed as in the proof of Lemma 10.1.I. T h e only difference in the proof is in showing that m ( t l , xl) has a maximum equal to zero for some x1 E G. Suppose that x1 E aG. Then, by assumption, Wtl
9
31) -
av(t1 9 x1)
-
870
a70
= FMtl >
Wl
9
x1)
370
4 - FMtl
9
XI))
> 0, since u(t, , xl) = v(t,, xl). Let x* E G be a point on the normal to the hypersurface (0, co) x aG at xl, sufficiently close to x l . Then, we obtain, from the fact that m ( t , x) is continuously differentiable and by application of mean value theorem, that m ( t l , x*) > 0. Since m(t, x) attains its maximum equal to zero at (tl , xl), this is an absurdity. Hence, x1 E G. T h e rest of the proof is standard, and (10.2.10) is therefore true. We now prove the second part, that is, we shall not demand T[v]< T[u].Let us deny the conclusion (10.2.10). Then, there exists a t, > 0 and an x1 E G such that 4 t l 1 x1)
34tl
2
4-
(10.2.11)
We define a function w(t, x) by
where both
E
> 0 and n >, 2 will be specified later.
and w(0, x)
=
v(0,x)
+E n-1
We have
162
CHAPTER
10
Since af/au exists, it follows that
vt
+ ( t + I)” E
~
> f ( t , x , v, v, ,az,)
-
vt
=
T[v], (10.2.12)
provided
and
E
> 0 is sufficiently small. Let
We now choose n so large that
With this value of n, the inequality (10.2.13) holds for x E G and t E [0, t J . There is a number p > 0 such that F 2 M 4 x1))
-
FMt, x1))
>P
9
for t E (0, t,] and x1 E aG, since the left side of the foregoing inequality is positive on a closed set. Hence, for all w ( t , xl) belonging to the domain of F, , there results W,(t,
4 = %(4 x1) = F2(44 x1)) > Fl(44 4 ) > Fl(44 x1)) > F1
+P
+ (n
(ax1) + .(
EM
€
+
1y-1
l)(t + 1)”-1
1
= Fl(W(4 xJ),
if E is sufficiently small. Here M is an upper bound of 1 F,’ now choose E > 0 so that (10.2.12), (10.2.14), and min[u(O, x ) X€G
-
v(0,x)] > E
(10.2.14)
1. Let
us
(10.2.15)
10.3.
163
BOUNDS. UNDER AND OVER FUNCTIONS
hold. Note that there is a positive value of E satisfying (10.2.15), since the left side is the minimum of a continuous positive function on a closed set. Thus, it follows, from (10.2.12) and T [ v ] T[u],that
<
T[w]< T[v]< T[u] Also, w(0, x)
=
v(0,x)
on
xE
G,
t
+2 n - 1 < u(0, x),
E
[O, tl].
x E G.
Applying the first part of the proof, we obtain w ( t , x)
< u(t, x)
[O, tl] x G.
on
However, this, together with (10.2.1 l), leads to W(t1
?
x1)
< u(t, > XI)
< 4 4 , x1) E
(n - I)(tl
+
1),-1
This contradiction shows that there does not exist a t, (10.2.1 l), and, hence, the proof is complete.
> 0 satisfying
10.3. Bounds, under and over functions Consider the partial differential equation where f E C[H x R x Rn x Rn2,R]. DEFINITION 10.3.1. Given an initial $(t,x), which is defined and continuous on Pt0U aH, a solution of (10.3.1) is any function u(t, x) satisfying the following properties:
(i) u ( t , x) is defined and continuous for ( t , x) E H ; (ii) u(t, x) possesses continuous partial derivatives u l , u, , u,, in H and satisfies (10.3.1) for ( t , x) E H ; (iii) u(t, x) = +(t, x) for ( t , x) E Pl0u aH. T h e problem of finding a solution to the partial differential equation (10.3.1) is called a first initial-boundary-vale problem. Let Q E C[aH, x R, R ] ,and suppose that (a) u(t, x)
= $(t, x)
on Pi0 and aH
-
aH, ,
164
CHAPTER
10
(b) a ( t , x)[au(t,x ) / a ~ ]$- Q(t, x,u(t, x)) = +(t,x) on aH,, where it is assumed that a ~ / &exists for ( t , x) E aH, .
If, instead of boundary condition (iii) in the Definition 10.3.1, we ask (a) and (b) to be satisfied, we have a mixed initial-boundary-value problem.
REMARK 10.3.1. If Q(t, x,u ) = P ( t , x)u, where P ( t , x) > 0 on aH,, the problem is said to be a first mixed problem; and if is not restricted to be positive on aH, , it is called a second mixed problem. If a ( t , x) = 0, the boundary condition (b) is of Dirichlet type, and the first mixed problem coincides with the first Fourier problem. O n the otherhand, if ~ ( tx), = 1 and P ( t , x) .= 0, the boundary condition (b) is of Newmann type, in which case the mixed problem reduces to second Fourier problem. If a(t, x) = 1, so that aH, = aH and P(t, x) is continuous on aH, the problem is called a third initial-boundary-value problem. If, in addition, H is a cylinder and the directions T are inward conormals, it is said to be a second initial-boundary-value problem. I n what follows, we shall assume the existence of solutions for the two boundary-value problems just stated, that is, first and mixed initialboundary-value problems.
THEOREM 10.3.1. Assume that (i) f E C [ H x R x R” x Rnz,I?], f ( t , x, u, P, R ) is elliptic, and one of the assumptions
<
g(t, I U I), (a) l f ( 4 x,U , 0,O)I (b) l f ( t , x, u, 0, 0)l < g ( t , I I), r ( t ) < < r ( t ) € 0 , for some co > 0, holds; (ii) g E C[/ x R, , R,], and the maximal solution r ( t ) = r(t, to , y o ) of (10.2.1) exists for t 2 t , ; (iii) u(t, x) is a solution of the first initial-boundary-value problem, such that on Pt0 u aH. I u ( t , .x)1 < ~ ( tto, , y o )
Then,
1 Pyoof. of
u ( t , %)I
+
< r ( t , to , y o )
on
H.
By Theorem 1.3.1, the maximal solution r ( t , c ) Y’
for sufficiently small
= g(t,y) E
+
€7
y(to>= yo
+
=
r ( t , to , y o ,c )
E,
> 0, exists on any compact interval
[ t o ,to
+ y],
10.3.
165
BOUNDS, UNDER AND OVER FUNCTIONS
> 0, and limc+or ( t , c) = r ( t , to , y o ), uniformly on [ t o , to Furthermore, we have
y
r ( t ) < r(t, .)
for some eo
< r ( t ) + €0 ,
> 0. Let us first
t
E
[ t o , to
+ rl,
+ y].
(10.3.2)
consider the function
m(t, x) = u(t, x)
-
r(t, €).
It is easy to see that m(t, x) satisfies conditions (i), (ii) of Lemma 10.1.1. T h e condition (iii) is also verified. For, let ( t l , xl) E Ptl , m(tl , xl) = 0, m2(tl , xl) = 0, and, for an arbitrary vector A,
Since this implies that u(t1 , xl) = r ( t l ,
c)
> 0, uz(tl , xl)= 0, and
we have, on account of the ellipticity off,
f(h
1
x1
9
U ( t l 3 xl>,U Z ( t l
7
'117
UZZ(tl
> .1>>
,U(t1 , Xl), 0,O).
(10.3.3)
Moreover, because of (10.3.2), 4tl>
< I 4 t l ,%>I < r(t1)
+
€0.
(10.3.4)
Thus, either of the assumptions (a) or (b) of the theorem shows that
using (10.3.3) and the fact that u(t1 , xl) by Lemma 10.1.1, that u(t, x)
< r(t, E )
Let us next consider the function
4 4 2)
=
-[r(t,
=
on
r(t, , c). We therefore obtain,
8.
€1 + 44 4 1
(10.3.5)
166
CHAPTER
10
and show that it also fulfills the conditions of Lemma 10.1.1. It is only required to check condition (iii) of Lemma 10.1.1. Suppose that (tl , .I> E ptl n(t1 7 . 1 ) = 0, n,(t, , x1) = 0, and 7
for some arbitrary vector X f 0. Then, we have -4t1,
€1 > 0,
x1) = r(t1 9
-4, , .1) = 0,
and
By the ellipticity of the functionf, we get
Also, (10.3.4) holds. Hence, as before, we obtain
on account of (10.3.6) and either one of the conditions (a) or (b), noting that ~ ( t,I),., = -u(t, , x,) = r(t, , e ) . Hence, by Lemma 10.1.1, -u(t, x )
< r(t, 6)
a.
on
(10.3.7)
T h e two inequalities (10.3.5) and (10.3.7) now yield the desired estimate
I 44 .)I
e r ( 4 t o ,YO)
on
H,
and the proof is complete.
COROLLARY 10.3.1. T h e assertion of Theorem 10.3.1 remains true if u(t, x) is a solution of the mixed initial-boundary-value problem such that
(ad !
u(t,
41 < r ( 4 t o ,Yo) on P," and
-
aH, ;
10.3.
(a2) for some p
I 4 4 ).
167
BOUNDS, UNDER AND OVER FUNCTIONS
> 0,
=+ 87,
e Br(4 1, ,Yo>
Q(4 x, u(t, .))I
on
aH, ,
where(? E C[aH, x R,R],Q(t, x, 2) is increasing in x for each ( t ,x) E aH,, Q(t, x, -4 = -P(t, x,4, and P(t,t o 3 Yo) Q(4 x,44 t o ,Yo))-
<
T h e next theorem offers a better bound and is a variant of Theorem 10.2.2.
THEOREM 10.3.2. Assume that (i) f~ C [ a x R x Rn x Rnz,R ] , the differential operator T is parabolic, and (10.3.8) f(t, x, -u, -P, -R) = -f(t, x, u, P, R); (ii) g E C [ J x R, , R ] ,g(t, 0) = 0, the maximal solution r ( t , to , 0) of (10.2.6) is identically zero, and f(t, x, zi 3 P, R ) - f ( t , x,z2 9 P, R )
< g(t, z1 - 4,
21
> z2 ; (10.3-9)
(iii) w E C [ H ,R,], v(t, x) possesses continuous partial derivatives 2 0, w t uzz, and T[v] >, 0; (iv) u(t, x) is a solution of the first initial-boundary-value problem satisfying 1 u(t, x)I < v(t, x) on Pi,, u 8H. u,
9
Then, we have
< v(t, x)
1 u(t, x)l Prooj.
on
H.
(10.3.10)
Let us consider the function
44 4 = -[u(t, 4
+ v(t, x) + Y ( t , 41,
where y(t, E) = y ( t , t o , 0, E ) is any solution of (10.2.9), for sufficiently small E > 0. Let (tl , xl)E Pt1, n(tl , xl) = 0, n,(tl , xl) = 0, and the quadratic form
for some vector A. This means, noting y(t, E ) -4t1 -U&1
> x1) f
> 0, that
> "(tl , 4,
4 = "&,
t
XI)
b 0,
168
CHAPTER
10
and
Because of the ellipticity off, it results that f(t1
, x1 , - 4 f 1 > X l ) , --%(f1 4,-u,,(t1, x1)) 5f ( t l , x 1 , - 4 t 1 , 4,-%c(t, , 4,%it1 , XI)). 9
Furthermore, using (10.3.8), (10.3.9), and the preceding inequality, we obtain "Ltl
7
x1)
= -
G
4 t l
-,f(t1
> x1)
s < -e,
I
J
>
x1)
, x1
,r ( t 1
?
?
XI),
1
I.'@,
c)) -
- Y'(t1
9
%(tl
4
9
4,%til1
-%(t,
, XI), %&I
'UAtl 9
4,%At,
XI>,
At1
7
> Xl), U,,(tl
Xl), %1 ti
1
x1 > v(t1 9 x1),
*2^1 1 - 4 t 1
Kfl
"dtl
4tl
x1
1
-f(t1
-
7
Y(tl
7
> x1))
, x1)) - Y'(t1 €1 , XI)) , x1)) YYtl 4 9
-
9
4) - e
which implies that n,(t, , xl) < 0. Clearly, n(t, 2) satisfies all the assumptions of Lemma 10.1.1, and hence
< v(t, x)
-u(t, x)
+y(t,
c)
on
H.
Proceeding similarly, we can show, on the basis of Lemma 10.1.1, that u ( t , x)
< v(t, x) + y ( t , c)
on
H.
T h e preceding two inequalities, together with the fact that lim,,,y(t, c) 0, yield the estimate (10.3.10). T h e theorem is proved. ~
COROLLARY 10.3.2. Let the hypotheses of Theorem 10.3.2 remain the same except that condition (iv) is replaced by (iv*) u ( t , x) is a solution of the mixed initial-boundary-value problem satisfying (a) 1 u(t, .)I v(t, x) on P t 0 and i3H - aH, ;
<
(b)
I a ( t , x)
au(t, x) ~
87"
+ a t , x, 44 .))I
av(t x ) < " ( t , x) -A 370
10.3.
I69
BOUNDS, UNDER AND OVER FUNCTIONS
where Q E C[BH, x R, R ] , Q(t, x, x) is increasing in z for each ( t , x ) aH,, ~ andQ(t, x, -z) = -Q(t, x, 2). Then, (10.3.10) is valid. We shall now introduce the notion of under and over functions with respect to the parabolic equation (10.3.1).
DEFINITION 10.3.2. Let u E C [ H ,R ] , and let u(t, x) possess continuous partial derivatives u t , u,, u,, in H. If u(t, x) satisfies the parabolic differential inequality T[u] < 0
on
H,
together with u(t, x) = + ( t , x) for ( t , x) E Pt0u aH, we shall say that an under function with respect to the first initial-boundary-value problem. On the other hand, if
u(t, x) is
T[u] > 0
on
u(t, x) is said to be an over function.
satisfies
on
u(t, x) = +(t, x)
a(t, x)
If
H,
&/a7 and
Pt0
9 +
Q(t, x, u(t, x))
exists on aH, and u(t, x)
= +(t,x)
aH
-
aH,,
on
we shall say that u(t, x) is an under or over function with respect to the mixed initial-boundary-value problem according as T[u]< 0 or T[u]> 0, on H.
As a direct consequence of Theorems 10.1.1 and 10.1.2, we have the following.
THEOREM 10.3.3. Let f~ C [ H x R x Rn x Rnz,R] and the differential operator T be parabolic. Suppose that u(t, x) and v(t, x) are under and over functions with respect to the first initial-boundary-value problem. If x(t, x) is any solution of the same problem such that u(t, x)
< x ( t , x) < v(t, x)
then u(t, x)
< z(t, x) < v(t, x)
on
Pt0u a H , on
H.
(10.3.1 1)
T h e inequality (10.3.11) remains true, even when u(t, x) and v(t, x)
170
CHAPTER
10
are under and over functions with respect to the mixed initial-boundaryvalue problem, provided z(t, x) is any solution of the same problem and u(t, x)
< z ( t , x) < v(t,x)
on
and
Pt0
i3H - aH, ,
+ Q(t,x, u(t, x)) < a(t, x) !%a7?? + Q(t, x, z(t, x)) av(t,x) < "(4 x) ____ a7 + Q(4 x, 4 4 4).
a(t, x)
10.4. Approximate solutions and uniqueness We shall begin with the theorems that estimate the difference between a solution and an approximate solution of (10.3.1).
THEOREM 10.4.1. Assume that (i) f E C[H x R x Rn x Rnz7R],the operator T is parabolic, and .f(t, X, 2 1
,P, R ) - f ( t ,
X, ~
2
P,%R )
< g(t, z1 - zz),
~1
> zz ; (10-4-1)
(ii) n E C [ p ,R],v(t, x) possesses continuous partial derivatives u t , a,, v,, such that I 7-[74 G s(t>, (10.4.2) where 6 E C [ J ,R,]; (iii) g E C [ J x R, , R],g(t, 0)= 0, and r ( t , t o ,yo ) is the maximal solution of
y'
= g(t,y)
+ S(t),
(10.4.3)
Y ( t 3 = yo 2 0,
existing for t 2 to ; (iv) u(t, x) is any solution of the first initial-boundary-value problem such that on Pto u a f f . I u ( t , ). v(t, X)l r(4 t o ,Yo)
e
~
Then, the estimate
I u ( t , ).
-
v(t, .>I
< r ( t , to ,yo)
on
is valid.
Proof.
Define N t , '4
= u ( 4 x)
-
v(t, x)
-
y(t, E),
R
(10.4.4)
10.4. where y ( t , c )
= y ( t , to
Y’
,y o , E) is any solution of
= g(t,y)
+ S(t) +
+
YPO) = yo
6,
for sufficiently small E > 0. Suppose that (tl , xl) m,(t, , xl) = 0, and, for some nonzero vector A,
Since y(t, c )
171
APPROXIMATE SOLUTIONS AND UNIQUENESS
E
(10.4.5)
E,
Ptl , m(tl , xl)
= 0,
> 0, this implies that
and
T h e last inequality yields, because of the ellipticity off, At1
Y
x1
7
aft1 7
x1>7 %(t, >
> x1
?
4tl
>
x1),
Uzz(f1
4,%it,
9
4)
7
X l ) , VU,U,(tl
, El)).
It follows, in view of the preceding inequality, that %(tl ? x1> -
vUt(t1 9 x1)
G f(tl x1 9
-f(t1 -
7
4tl
?
7
x1)7
x1 7 v(t1
9
uz(t1
>
Xl),
vJt,
Xl),
9
Trz(t1
Xl),
7
4)
vzz(t1 > x1))
T[v].
Hence, the relations (10.4.1) and (10.4.2) show that %(tl 7 5 1 )
-
4 t l
x1)
< g(t1
9
Y(tl
.>I
9
+ S(t).
We thus have mdt1 ? x1) =
<
%(tl 7 x1) --E
-
< 0.
vdt,
9
x1)
- y’(t1 9
.)
It is easy to see that m(t, x) satisfies the hypotheses of Lemma 10.1.1, and hence
u (t, x)
--
v(t, x)
< y(t, 6 )
on
H.
I n a similar way, considering the function n(t, x)
=
v(t, x )
-
u(t, x) - y(t, €)
I72
CHAPTER
10
and verifying the assumptions of Lemma 10.1.1, we can obtain v(t, x)
u(t, x)
on
ET.
T h e last two inequalities, together with the fact that lim Y ( t , €1 E
-0
r(t, to ,Yo),
assure the stated estimate (10.4.4).
THEOREM 10.4.2.
Let u(t, x) be any solution of the mixed initialboundary-value problem such that
<
(a) I u(t, x) - v(t, I). r ( t , t, , y o ) on Pt0 and aH - aH, ; (b) for each ( t ,x) E 8 H a , a(/ u - ZI I)/&, exists, and, for some j3
> 0,
where Q E C[aH, x R , , R ] , Q(t, x,z ) is increasing in x for each ( t , x), and p r ( t , t, , y o ) Q(t, x,r ( t , to , yo)),other hypotheses (i), (ii), and (iii) of Theorem 10.4.1 being the same.
<
Then (10.4.4) is true. REMARK 10.4.1. If u(t, x), v(t, x) are any two solutions of the boundaryvalue problem, we can deduce the estimate of the difference between them, as a consequence of Theorems 10.4.1 and 10.4.2. Similar remark holds for the theorem given below.
'Tmomni 10.4.3. Assume that (i) f , G E C [ H x R x RrLx Rnz,R ] , G(t,x, z , P, R) is elliptic, and, if x1 > z 2 , PI P, , f ( t 9
(ii)
F
x,2 1 1'1 > R , ) -At, ,x, 2 2 p2 > 4) 2 ,G(t,2, z1 - 22 , Pl - P, , R, - RZ); 7
9
(10.4.6)
C [ f I ,R ] , v(t, x) possesses continuous partial derivatives
ZI~,
(10.4.7)
10.4.
173
APPROXIMATE SOLUTIONS AND UNIQUENESS
whereg E C [ J x R, , R],g(t, 0) = 0, and the maximal solution r ( t , to , 0) of (10.2.6) is identically zero; (iv) z E C [ R ,R+], z, 2 0, z t , z,, exist and are continuous in H , and
2 G(t, x, z,
zt
Then,
I 44).
- v(l,
+ s(t, 4.
zzz)
< z(t, x)
on
Pi0 u aH
41 e 4 4 4
on
H,
1 u(t, x) - v(t, x)I implies
232,
(10.4.9)
where u(t, x) is any solution of the first initial-boundary-value problem.
Proof. As usual, we shall reduce the theorem to Lemma 10.1.1. We shall first consider the function
where y ( t , E ) = y ( t , to , 0, e) is any solution of (10.2.9) for small E > 0. Suppose that (tl , xl) E Ptl , m ( t , , xl) = 0, m,(t, , xl) = 0, and, for some nonzero vector A,
Since z(t, x) >, 0, y ( t , E )
> 0, the preceding 4 t l , x1)
%(tl x1) 9
-
supposition implies that
> 4 t l 4,
%it, 1x1)
9
= zz(t1,
x1)
2 0,
T h e ellipticity of G shows that
In view of this and the relations (10.4.6), (10.4.7), we have
174
10
CHAPTER
Hence, using (10.4.8) and (10.4.9), we derive 4 t l
,
< G(t, ~
<
~
?
x1 , 4 t l
G(t1 x1 7
R(tl > Y(tl
+
I
6))
< 0.
-€
+
, XI) Y(tl , E ) , %it, xl), &(tl , x1)) so1 ,)I. , X l ) , %it1 X l ) , Z , X ( t l ,x1)) - S(t1 , 4 9
Z(t1
9
7
--
This shows that m(t, x) verifies the assumptions of Lemma 10.1.1, and, therefore, on H. u(t, x) v(t, x) < z(t, x) + y ( t , E) -
Arguing similarly, we can prove w ( t , x)
-
u(t, x)
< z ( t , x) + y(t, 6)
on
H.
Since lim,,,y(t, e ) = 0, by assumption, it follows from the preceding two inequalities that
I u(t, .x)
-
v(t, .x)I
< z(t, x)
on
H,
proving the theorem. We shall next consider the uniqueness problem.
THEOREM 10.4.4. Suppose that (i) f~ C [ H x R x Rn x R’”, R ] , the operator T is parabolic, and
f(4 x, z1 , p,R ) -f(4
x,
2.21
(ii) g E C[/ x R, , R ] ,g(t, 0) of (10.2.6) is identically zero.
p, R) =
< f(t, z1 - 4,
z 1
>22 ;
0, and the maximal solution r ( t , to , 0)
Under these assumptions, there is at most one solution to either one of the initial-boundary-value problems. T h e proof is a direct consequence of Theorems 10.4.1 and 10.4.2.
10.5. Stability of steady-state solutions Let problem D represent the partial differential equation of the form ut = :
for x E [a, b], t
> 0, together
%(t, a )
= f,(u(t,
a)),
f(.x, u, u,
7
(10.5.1)
ux,),
with the boundary conditions u,(t,
b)
= f,(u(t,b)),
t
> 0,
10.5.
STABILITY OF STEADY-STATE SOLUTIONS
175
where fi ,fiare continuous functions with bounded derivatives. Let us assume that af/au exists and is continuous,f (x, u, P, R ) is nondecreasing in R for each (x, u, P ) . We use the notation u(t, x, $) to denote a solution of problem D such that
4 0 , x, 4 ) = 4 w ,
6
[a,
4,
where 4 E C [ [ a ,b], R]. DEFINITION10.5.1. Let u ( t , x, $) be a solution of problem D. We shall say that u(t, x, 4)is a steady-state solution if u(t, x, $) = $(x), t > 0.
DEFINITION 10.5.2. T h e steady-state solution u(t, x, 4) of problem is said to be stable if, given E > 0, there is a 6 > 0 such that
D
implies
DEFINITION10.5.3. A
Let
= [(x, u): x E [ a , b’J
< < #z(x)17
and
where z+hl , are arbitrary functions, twice continuously differentiable on [a, 61. Let B be the set of functions on [a, b] such that z+h E B implies
T h e steady-state solution u ( t , x, $) of problem D is said to be asymptotically stable if it is stable and lim[ max j u(t, x,+) - u(t, x, #)I]
t-m
x~[a,bl
=
0,
whenever I$ E B. T h e set A is called the domain of attraction. Sufficient conditions for a steady-state solution u(t, x, $) of problem D to be stable are given by the following theorem.
THEOREM 10.5.1. Assume that there exists a one-parameter family o(x, A), A E [A, , A,], of solutions of the equation f(X,
v, v z , vm)
=
0
(10.5.2)
176
10
CHAPTER
fulfilling the following conditions: (i) there is a A * E ( A , , A), such that v,(a, A*) =f,(v(a, A*)) v,(b, A * ) = J.(v(b, A")); (ii) zq,(x, A) > 0, x E [a, 61, A E [A, , A,];
(4a,(% 4 >f,(v(a, 4)and %(b, A)
f,(v(b, A)), A
(iv) a,(%
Then, if 4(x) lem D is stable.
Pyoof.
Let
E
=
given. Choose A, max
1 v(x, A*)
max
1 v ( x , A")
xt[a,b]
[A, , A*); E (A*, A,]. E
the steady-state solution u(t, x,4 ) of prob-
v(x, A*),
> 0 be
and
~
E
[A, , A*), Ao
E
(A*, A2] such that
v(x,A,,)\ < E
(10.5.3)
< €.
(10.5.4)
and xt[a,b]
-
W(X,
A*)(
Then, define the number 6 by 6
7-=
min[ min (v(x,A*)
-
v(x,An)),
xt[a,bl
min (v(x,An)
xt[u,bl
-
Since v,(x, A) > 0 for all x E [a, b ] , it is clear that 6 the inequality
and assume that $(x) a(x,A,)
5:
$A).(
= -
(10.5.5)
v(x,A*))].
> 0. Let $(x)
satisfy
n(x,A"). We then have, by (10.5.5),
s < #(x) < +(x)
+ < v(x, A,),
x E [a, b ] .
The fact that v(x,A) is a family of solutions of (10.5.2) and u(t, x, $) is a solution to the Droblem D imolies that
By a successive application of Theorem 10.2.4, we deduce that v(x, A") < u(t, x, I))
< v(x,A",
x E [a, b ] ,
t
2 0.
10.5.
177
STABILITY OF STEADY-STATE SOLUTIONS
This, because of the relations (10.5.3) and (10.5.4), yields u ( 4 x,
4)- E = v(x, A*)
and u(4 x, 4)
+
=
€
v(x, A*)
-E
< v ( x , A,) < u(t, x, $)
+
> v(x, An) > u(t, x, #).
€
It is evident from the preceding inequalities that
< 6.
whenever max,.[,,,l I $(x) - +(.)I
T h e proof is complete.
I n the situation in which it is difficult or impossible to find a oneparameter family u(x,A), satisfying the conditions of Theorem 10.5.1, it may still be possible to find an upper bound. This we state as a corollary.
COROLLARY 10.5.1. Suppose that there exists a solution u(x)of (10.5.2) satisfying %(a) < fl(44) and vdb) > f,(v(b))* Then, if u(t, x, $) is a solution to problem D such that #).(I
we have
< v(4,
u(4 x,4) < +),
x E [a, bl,
x E [a,bl,
t b 0.
A similar corollary may be stated establishing a lower bound.
THEOREM 10.5.2. Let the hypotheses of Theorem 10.5.1 hold. Suppose further that, for x E [a, b] and A E [A, , A,], fv(x,
v(x,
3,%(X,
A),
%,(X,
Then, if +(x) = u(x,A*), the steady-state solution u(t, x, totically stable, and the set A
=
[(x, u ) : x E [a, b]
and
(10.5.6)
A)) f 0-
v(x, A,)
4) is
< u < v(x, A,)]
asymp(10.5.7)
is a region of attraction.
Proof. T h e stability of steady-state solution u(t, x, #J) follows by Theorem 10.5.1. Let A be the set defined by (10.5.7), and let B be the set of functions such that E B implies [(x, +(x)) : x E [a, b]] C A. We
178
10
CHAPTER
shall first show that, for any E > 0 and any 4 E B, there exists a T I > 0 such that max [@, x, $) - 44 x, 411 < E . (10.5.8) x~Ca,bI t>T,
Let
E
> 0 be given, and let ha E (A*, v(x, AO)
u(x,A*)
-
A,] be such that
< E,
(10.5.9)
x E [a, b].
Without loss of generality, we may assume that f, > 0, in view of ,u2, p3 by
(10.5.6). We then define three positive numbers p l ,
v(x,
A)
~
'u(x,
< p1/2,
A*)
x E [a, b].
Define a positive number p4 by
Let H(A) be a function defined for A E [Ao ,A,], such that, for h
fZ('u(b,
4 - h ) -fz('u(b, 4)< P4/2.
< H(A), (10.5.12)
Consider the function w
=
w(x, A) = v ( x , A)
-
where 6 > 0 will be specified later. Since small 6 > 0, we have f(X,
w, w, > w,,) = f(x.
'u -
< f(X, =
0.
'u, u ', 'up
u ',
f, exists, for sufficiently
8 , u' , ,' u 3
= f(X, 71%'us > u ),',
8,
9
'um)
,'um)
- VV(& ~
8p2
'us u ',
, 'uz,)
+ 0(S2)
+ 0(S2) (10.5.13)
10.5.
We let 6,
179
STABILITY OF STEADY-STATE SOLUTIONS
> 0 be such that 6,
< rmin [v(x,A,) e[a,b]
- $(x)].
Let us now choose a positive 6 so small that (10.5.13) holds, and also (10.5.14)
( 10.5.15)
I t follows, from the inequality (10.5.13), that there exists a p5 > 0 satisfying f(X,
Let m
=
w ,w, w,,) 9
< -PF,
E
9
[A0,
41,
A*)
e-pt,
xE
[a, 4.
m(t, x) = w(x, h(t)), where h(t) = A*
+ (A,
-
p > 0, being a number to be specified shortly. Then, f ( x , m,m, , m,,) - m
t =f(x,
<
w ,w,
+ wAp(A,
+
-p5
9
~
WZZ)
A*)
e-pt
(10.5.16)
P 3 P ( A , ~- A*).
Choose p = p5/p3(h2- A * ) so that the right-hand side of (10.5.16) is equal to zero. Then the inequality f(x,
?%
m,
,%z)
- mt < f ( % u, u,
>
% I . )
U t -=
0
is verified for x E [a, b ] , t E [O, T I ] , where T I is the solution of the equation h ( T J = A*. T h e inequality (10.5.1 l), together with (10.5.14) and (10.5.15), shows that
fI(+? 4 - 6 ) -f1(+4 4) >
-(P4'2)-
(10.5.17)
From the definition of w , we deduce, in view of (10.5.10) and (10.5.17), that
f M 4 A))
4 6) >fi(+, 4) - (P4/2) > v,(a, 4
= f*(fl(a,
==
w,(a, A).
~
(10.5.18)
180
CHAPTER
10
Similarly, from (10.5.10), (10.5.12), (10.5.14), and (10.5.15), we derive $dw(b,
4) = fz(v(b, A) < v,(b, A)
9+ ( d 2 )
8) < f2(7@, wc(b, A).
-
=
Clearly, (10.5.18) and (10.5.19) hold for all X E [A*, A,]. Theorem 10.2.4, it follows that m(t, ).
2
u(t, x,
x E [ a , 61,
$17
t
E
[O,
(10.5.19)
On the basis of 7-11>
since m(0, x) > $(x), x E [a, b ] . From the definition of m and Tl , we get
> u(T1 , x, $),
v(x, AO)
x E [a,4 .
Thus, by Corollary 10.5.1, v(x, Ao)
> u(t, x, $),
x E [a, 61,
t
> T, ,
which, together with (10.5.9), gives us (10.5.8). T h e next step of the proof is to show that there exists a T, > 0 such that (10.5.20) max [u(4 x, 4)- 4 4 x, $11 < xe[a.bl t>T,
T h e proof of this consists of showing that there is a lower bound for $) which can be increased with time until it is within E of u ( t , x, 4) at some time T, . T h e proof of this fact is similar to the first part and differs only in minor details. Let T = max[Tl , TJ. Then, from (10.5.8) and (10.5.20), we obtain u(t, x,
max
xtla.61
I u(t, x, 4)
~
u(t, x,
$)I < E .
t>T
This completes the proof of the theorem. COROLLARY 10.5.2. T h e conclusion of Theorem 10.5.2 remains valid if (10.5.6) is replaced by either (i) f,, = 0 andf,&, v(x, 4, %(X, 4, % r ( X , (ii) f r = 0, f U Z = 0, and fv,,(x, 4x, A), x E [a, 4 , A E [A, A,].
4) f %(X,
0 ; or A), %,(x,
A)) f 0 for
9
We now give an example to illustrate Theorems 10.5.1 and 10.5.2. Consider the partial differential equation u.
-
(1
+
242)
u,. - uu,z,
xE
[I, 21,
10.6.
181
SYSTEMS OF PARABOLIC INEQUALITIES
subject to the boundary conditions I)
T h e equation (1
+
uz(4 2) = fz(u(4 2)).
and
= f&(4 1))
242)
u,,
-
uu,2 = 0
has a one-parameter family of solutions v ( x , A ) given by v(x, A ) Notice that v,(x, A) = x-l[l + &(x, A)]1/2 sinh-l v ( x , A).
=
sinh Ax.
Suppose that
+ u2)1/2sinh-l u, fl(u) < (1 + u2)1/2sinh-l u, ti(.) < +(l + u2)1/2sinh-l u, f2(u) > t(l + u2)1/2sinh-' u,
fl(u) > (1
u u u u
> 0, < 0, > 0, < 0.
fl(0)
=
0,
f2(0)= 0,
Then, by Theorem 10.5.1, u(t, x, 4) = 0 is stable. T o apply Theorem 10.5.2, we observe that fv(x, v, 0,
,v,,)
=
2vv,, - vz2 2A2 sinh2Ax - X2 cosh2Ax
=
A2(sinh2Ax- 1).
=
Thus, fv < 0 if sinh2 Ax < 1 or u < 1. Hence, u(t, x, 4) = 0 is asymptotically stable, by Theorem 10.5.2.
10.6. Systems of parabolic differential inequalities in bounded domains Let us consider now a partial differential system of the type where
u: = f i ( t , x, u,u,i, u;,),
u,i =
, d2,-*.,
i
=
1, 2 ,...)N ,
u&),
For convenience, we shall write the preceding system in the form
182
CHAPTER
10
where f E C [ w x R N x Rn x Rn', RN]and each function f is elliptic,
so that the system is parabolic.
We shall first state the following lemmas, which are extensions of Lemmas 10.1.1 and 10.1.2.
LEMMA 10.6.1. Suppose that (i) m E C [ H ,RN],m(t, x) possesses continuous partial derivatives m ,, m, mzz in H ; (ii) m(t, x) < 0 on Ptou aH; 7
<
(iii) for any (tl , xl)E Ptl and an indexj, 1 < j N , if mj(t, , xl) mi(t, , 31) 0, i # j , mzj(tl , xl) = 0, and the quadratic form
<
=
0,
h being an arbitrary vector, then mi(t,, xl) < 0.
Under these assumptions, we have m(t, x)
Proof.
<0
on
H.
(10.6.2)
Assume, if possible, that the set 2=
N
(J [ ( t ,x) E H : myt, x) 3 01 i=l
is nonempty. Let Z , be the projection of Z on t axis and t, = inf Z. I t follows from condition (ii) that t , > t o . Since the set 2 is closed and condition (ii) holds, we conclude that mi(tl , .Y)
<0
on
Ii n [to , t l ] ,
for i = 1, 2 ,..., N , and there is an indexj, 1 such that "'(tl , xl) = 0.
<j < N , and an x1 E int Pt1
We therfore have m,'(t, , x1) >, 0.
(10.6.3)
On the other hand, since mj(t, x) attains its maximum at (tl , xl)E int P t l , we get n2,'(t1 , XI) = 0,
10.6.
183
SYSTEMS OF PARABOLIC INEQUALITIES
and, for some nonzero vector A,
By condition (iii), it follows that contradicting (10.6.3). Hence, the inequality (10.6.2) IS ' true. Let 01 E C[aH,R N ] ,and denote by aHmithat part of the boundary aH on which d ( t , x) > 0. Corresponding to this change, we may define, as before, for each i = 1, 2,..., M,and ( t , x) E aHmi,aui/a?, aui/a.roi,T ~ .r0i being the directions. Then, we have an extension of Lemma 10.1.2 which may be proved by combining the proofs of Lemmas 10.1.2 and 10.6.1.
LEMMA 10.6.2, Let condition (ii) of Lemma 10.6.1 be replaced by (iia) mi(t, x) < 0 on Pt0and aH - aHmi; (iib) for any index j , 1 j N and (tl ,xl) mi(t, , xl) = 0, i j , then
+
< <
E
aH,i, if d ( t , , xl)
=
0,
where, for each ( t , x) E aHmi, ami/& is assumed to exist. Then, the assertion (10.6.2) is true if the other assumptions remain the same. We are now in a position to extend the previous results to systems of parabolic inequalities. We shall present only some typical theorems.
THEOREM 10.6.I. Assume that (i) u, z, E C[H ,R N ] ,the partial derivatives a t , u, , u,, , ot, v, , v,, exist and are continuous in H ; (ii) f E C[R x RN x R" x Rn2,R N ] ,the differential operator T is parabolic, the function f ( t , x, u, P, R ) is quasi-monotone nondecreasing in u for each ( t , k, P , R), and T[u] < T[v]
(iii) u(t, x) < v(t, x) on Pt0u aH. Then, we have u(t, x ) < v(t, x)
on
on
H;
H.
(10.6.4)
,
184
CHAPTER
10
P Y O O ~We . shall show that the vector function m ( t , x)
=
u(t, x)
~
v(t, x)
satisfies the hypotheses of Lemma 10.6.1. I n view of conditions (i) and (ii) of Theorem 10.6.1, it is sufficient to check assumption (iii) of Lemma 10.6.1. Suppose that ( t l , xl)E Ptl , mj(t, , xl) = 0, 1 < j N, mi(tl , xl) 0, i # j , m,j(t, , xl) = 0, and
<
<
for an arbitrary vector A. This means that (10.6.5)
and
Since f j is elliptic, we get
This, together with (10.6.5) and quasi-monotonicity of fi, implies that
10.6.
185
SYSTEMS OF PARABOLIC INEQUALITIES
THEOREM 10.6.2. Let assumptions (i) and (ii) of Theorem 10.6.1 hold. Suppose further that (iii*) ui(t, x) < vi(t, x) on Pt0and aH - aHai, i = 1, 2,..., N ; (iv) for each ( t , x) E aHai, aui/aTi, avi/ad exist, and uyt, x)
t, x aTi
+ gyt, x, u(t, x))
where Q E C[HH x RN,RN], Q(t, x, u ) is quasimonotone nondecreasing in u for each ( t , x). Then, the inequality (10.6.4) remains valid.
Proof. The proof proceeds as in Theorem 10.6.1. T o verify the assumptions of Lemma 10.6.2, we have only to check condition (iib). N and (tl ,xl)E aH,j. Suppose that &(t, , xl) = 0 and Let 1 < j mi(t, , xl) 0, i # j . This implies that
<
<
Uj(t1
Ui(tl
,x1) = vj(t1 , X d , , x1) < vi(tl , Xl),
#j.
Hence, by the quasi-monotonicity character of Q, we have P(t11 x1 , 4 t l 9 x1))
< Qvl,
XI, UP1
, x1)).
Thus, there results the inequality
q,, x1) amj(t1
7.1)
aTj
< P(tl =
I
x1 ,Q(t1 ,Xl))
-! a t 1
1
x1 , v(t1 , x1))
0,
which assures that amj/aTj(t, , xl) This completes the proof.
< 0,
since d ( t l , xl)
>0
on aH,j.
It is now easy to formulate and prove comparison theorems, componentwise bounds, and error estimates for systems of parabolic differential equations. We shall not include such results. However, the following comparison theorem will be needed later, and, hence, we shall merely state it.
THEOREM 10.6.3. Suppose that (i) m E C [ g , R+N],m(t, x) possesses continuous partial derivatives m, 1 m, , mzz in H ;
186
CHAPTER
10
(ii) f E C [ H x RN x Rn x Rn2,R N ] ,the differential operator T is parabolic, and T [ m ] 0 on H ; (iii) g E C [ J x R t N , R N ]g(t, , y ) is quasi-monotone nondecreasing in y for each t E J , r ( t , to ,yo) 2 0 is the maximal solution of the differential system
<
Y'
existing for t
20
3 to , and .y',
f(f,
(iv) m(t, 4
Y(t0) = Yo
g(t,y),
=:
< r(t, t o
9
x, 0,O)
< g(4 4,
z
2 0;
yo) on Pto u aH.
Then, m(t, x)
< r ( t , to , y o )
on
H.
10.7. Lyapunov-like functions
Let us continue to consider the partial differential system (10.6.1). We shall restrict ourselves to the first initial-boundary-value problem of such systems. I n what follows, a solution will always mean according to Definition 10.3.1, unless specified otherwise. Let V E C [ H x R N ,R,] and V = V ( t ,x, u ) possess continuous partial derivatives with respect to t and the components of x,u.Let
"'1.
L..,
p, p1 ~=1, v , v l - 1 , 2,..., n
T h e following theorem offers an estimate of solutions in terms of Lyapunov-like functions and is useful subsequently.
THEOREM 10.7.1. Suppose that (i) f E C [ H x R N x Rn x Rn2,R N ] ,G E C [ H x R, x Rn x Rn2,R], G is elliptic, and
a v av at + a , f ( t , x, u, u,Z, &) < G(4 v(4 x, u), V,, X,
V,,),
(10.7.1)
10.7.
I87
LYAPUNOV-LIKE FUNCTIONS
where V E C[R x RN,R,] and V , , V,, are given in the foregoing; (ii) g E C [ J x R, , R] and G(t,x, z, 0 , O )
< g(4 4;
(10.7.2)
(iii) the maximal solution r(t, to , y o ) of the differential equation (10.2.6) exists for t 3 to ; (iv) u(t, x) is any solution of the partial differential system (10.6.1) such that
v(t,X, w,x))
G r(t, to ,yo)
on
ptou aH.
(10.7.3)
a.
(10.7.4)
These assumptions imply
v(t,x, 4 4 4)< r(t, t o ,Yo)
on
Proof. Let u(t, x) be any solution of (10.6.1) satisfying (10.7.3). Define the function m(t, x)
Then,
=
V(t,x, u(t, x)).
4, ). < r(4 to ,Yo)
on
Pto u aff.
Moreover, because of (10.7.1), we obtain
Now, a straightforward application of Theorem 10.2.1 yields the estimate (10.7.4).
THEOREM 10.7.2.
Let the assumptions of Theorem 10.7.1 remain the same except that the relations (10.7.1) and (10.7.3) are replaced by 41)
av av [ x + 3F .f ( 4 x,
u, u,i, uL,]
+ V(4 x, 4 4)
< G(t,x, 4 t ) V t ,x, 4,4)v, , 4 t ) V,,),
(10.7.5)
where A ( t ) > 0 is continuously differentiable on J and A(t) v(t,3, +(t, x))
< r(t, t o ,yo)
Ptou aH-
on
Then, the inequality (10.7.4) takes the form A(t) V(t, x, u(t, x))
< r(t, to ,yo)
on
H.
(10.7.6)
188
Proof.
CHAPTER
10
Defining lTl(t,
x,).
=
A ( t ) q t , 2,
4,
it can be easily checked that Vl(t,x,u) preserves the properties of V ( t ,x,u ) in Theorem 10.7.1, and hence (10.7.6) follows from Theorem 10.7.1.
REMARK10.7.1. Taking A(t)= 1, we see that Theorem 10.7.2 reduces to Theorem 10.7.1. Since Theorem 10.7.1 is an important result in itself in the study of various problems of partial differential equations, we have listed it separately. We note that g ( t , z ) in (10.7.2) need not be nonnegative, and, hence, this has an advantage in obtaining sharper bounds. For example, taking V = 11 u 11' andg(t, u ) = L(t)u,whereL(t) is continuous on J , one can get an upper bound, from Theorem 10.7.1, as follows:
provided
where
max II + ( t o XtPd
I
.)I1
< Yo.
If we assume that V ( t ,x,u ) = 0 if and only if u may be used to prove a uniqueness result.
=
0, Theorem 10,7.€
THEOREM 10.7.3. Assume that (i) f E C [ H x R N x Rn x Rn2,R N ] G , E C[H x R , x Rn x Rnz,R ] , G is elliptic, and
-< G(t,X,
V ( t , x, u
-
v), Vz(t,x,u
-
v), Vzs(t,x,u - v));
,.
(ii) g E C [ J x R , R ] ,g(t, 0) = 0, and G(t,5,z,0,O)
< g ( 4 4,
> 0;
(iii) the maximal solution r(t, to , 0) of (10.2.9) is identically zero. Then, the parabolic differential system (10.6.1) admits a unique solution.
10.7. LYAPUNOV-LIKE FUNCTIONS
189
Let V I EC[H x R N ,R+N], and suppose that V ( t ,x, u ) possesses continuous partial derivatives V t , V , , V,, , V u. Then, the following extension of Theorem 10.7.1 may be proved.
THEOREM 10.7.4. Let the following assumptions hold: (i) f E C[H x RN x Rn x Rn', R N ] GE , C[H x R,N x Rn x Rn2,R N ] , G is elliptic, and
(ii) g E C [ J x R+N,R N ]g(t, , y ) is quasi-monotone in y for each t E J , and G(4 X, z,o, 0) < g ( t , 4, 3 0; (10.7.8) (iii) the maximal solution r(t, to ,yo) of the ordinary differential system Y'
= g(t,y),
Y(tJ
= Yo
30
exists for t 3 to ; (iv) u(t, x) is any solution of the system (10.7.1) such that
Under these assumptions, we have
Proof.
Consider the vector function
where u(t, x) is any solution of (10.6.1) such that (10.7.9) holds. By the assumption (10.7.7), we deduce the inequality m(t, 4
e G(4
X,
dt,
4, d ( 4 4, &it,
XI).
Since m(t, x) satisfies the assumptions of Theorem 10.6.3, it follows that 4 4 4 < 44 to Yo) 9
and the proof is complete.
on
H,
190
CHAPTER
10
10.8. Stability and boundedness Let u(t, x) be any solution of the partial differential system (10.6.1). We shall assume that the system (10.6.1) possesses the identically zero solution. Denote
I1 44 .)llPt
=
max I1 4 4 411. XCP,
DEFINITION 10.8.1. T h e trivial solution of the partial differential system (10.6.1) is said to be equistable if, for each E > 0 and to E J , there exists a positive function 8(t,, E ) , which is continuous in to for each E , such that
On the basis of this definition, it is easy to formulate the various definitions of stability and boundedness analogous to those in the earlier chapters. We shall now give sufficient conditions for stability and boundedness of solutions of the parabolic difierential system (10.6.1). 'rHEOREM
10.8.1. Assume that there exist functions V ( t ,x,u ) and
g(t, y ) satisfying the following conditions:
(i) g E C [ J x R , RI, g(t, 0 ) 0; (ii) V E C[H x RN,R,],V ( t ,x,u ) possesses continuous partial derivatives V , , I/, , V,, , V,, in H , and 9
~
WI 11)
< 1 -(t, x, ). < a(t, I1
ll),
(10.8.1)
where h t N ,a E C [ J x R , , R,], a F Y- for each t E J ; (iii) f E C[H Y R N x R" x Itnn,R N ] ,G E C [ H < R , x Rn x Rn',R] G is elliptic, and
a[- a v 27 + (iv) G(t, X,
Z,
.f(t,
0, 0 )
12, 21,
usL,? 1 1 J
< G(t,x, b.(t, .w, u ) , I T z , 17rE);
< g(t, z), 2 3 0.
10.8.
191
STABILITY AND BOUNDEDNESS
Then, the equistability of the trivial solution of the ordinary scalar differential equation y'
= g(4y),
Y(t0) = Yo
(10.8.2)
30
implies the equistability of the trivial solution of the partial differential system (10.6.1).
Proof. Let E > 0 and to E J be given. Assume that the trivial solution of (10.8.2) is equistable. Then, given b ( e ) > 0 and to E J , there exists a 6 = 8 ( t o ,E) > 0 that is continuous in to for each E such that Y ( t , to ,Yo)
<
<w,
t b to
(10.8.3)
?
provided yo 6, where y(& to , y o ) is any solution of (10.8.2). Let be any solution of (10.6.1) such that
u(t, x)
Then, by Theorem 10.7.1, we obtain
q t , x,u(t, .)I < r(t, t o ,yo)
on
(10.8.5)
H,
where r(t, to ,yo)is the maximal solution of (10.8.2). Choose yo such that a ( t o ,II $ ( t o , .)IIpto) = y o . Then, there exists a 6 , = S1(to, E) such that
II +(to
9
.)lIpt0
< 81
and
a(to 9 II d(to , .)IIPJ
<8
(10-8-6)
hold simultaneously. It is clear that 6 , is continuous in to for each We claim that, if (4 (b)
then
I1 +(to >
%P,,
lld(4 .)llm <
< 8;
€9
t
2 to
/I 44 .)IIP, < E ,
t
2
If this is not true, suppose that, for some t,
II 4 t , *
.>IIP,,
1
to.
> t o , we
have
36
Then, by condition (b), there exists an xo E int P,, such that
II +I
7
x0)Il
= E.
E.
192
CHAPTER
10
It now follows from relations (10.8.1), (10.8.3), and (10.8.5) that
< i;(tl
44
, xo "(t,, %)) <
, t o Yo) < 44,
9
9
which is a contradiction. This proves the equistability of the trivial solution of (10.6.I), and the proof is complete.
THEOREM 10.8.2. Let the assumptions of Theorem 10.8.1 hold except that the function a(t, y ) occurring in (10.8.1) is independent of t , that is, a ( t ,y ) = a(y)E X . Then, the uniform stability of the trivial solution of (10.8.2) implies the uniform stability of the trivial solution of (10.6.1). Proof. I n this case, it is enough to choose 6, = ~ ~ ( 6 Since ) . 6 is independent of t o , it is clear that 6, is independent of to . T h e rest of the proof is very much the same as that of Theorem 10.8.1.
THEOREM 10.8.3. Under the assumptions of Theorem 10.8.1, the equi-asymptotic stability of the trivial solution of (10.8.2) implies the equi-asymptotic stability of the null solution of (10.6.1). Proof. Assume that the solution y = 0 of (10.8.2) is equi-asymptotically stable. This implies that (S,*) and (S,*) hold. Hence, we need to prove only the quasi-equiasymptotic stability of the trivial solution of (10.6.1), as, by Theorem 10.8.1, the equistability is guaranteed. Let E > 0 and to E J be given. It then follows, on account of (S3*), that, given b(c) > 0 and to E J , there exist positive numbers 6, = so(to) and T = T ( t o ,E) such that
+
(10.8.7) 2 to T , provided yo < 6, . Choosing yo = a(t, , 1) $(to , -)I\ plo),as before, we can Y ( 4 t o 3 Yo) < b ( 4 ,
t
show the existence of a positive number
/I C(to -)llp'o ,< &I I
and
8,
=
8,(t,) such that
4 t o 7 I1 +(to
9
.)llPJ
< so
hold at the same time Suppose now that n ; (a) I/ +(to * ) l / P t o (b) I1 +(t, *>/laIf < ' t 3 to
+ T.
+
Let there exist a sequence { t k } , t, 3 to T and t,+ co as k + co, such that / / u(t, , *)llPtk 2 E for some solution u(t, x). Then, there exist x, t int P t k satisfying (1 u(t, , x,)[[ = E in view of condition (b). Thus, using relations (10.8.1), (10.8.5), and (10.8.7), we arrive at the contradiction b(c) < V(t, , X k , U ( t , , X k ) ) < Y ( t , , to ,Yo) < 4).
10.8.
193
STABILITY AND BOUNDEDNESS
Thus, the quasi-equi-asymptotic stability holds, and, as a result, the trivial solution of ( 10.6.1) is equi-asymptotically stable.
THEOREM 10.8.4. Under the assumptions of Theorem 10.8.2, the uniform asymptotic stability of the trivial solution of (10.8.2) implies the uniform asymptotic stability of the trivial solution of (10.6.1). Proof. Assume that the trivial solution of (10.8.2) is uniformly asymptotically stable. Then, we have (S,*) and (S,*). By Theorem 10.8.2, the uniform stability of the trivial solution of (10.6.1) follows. T o prove the quasi-uniform asymptotic stability of the trivial solution, we proceed as in Theorem 10.8.3 and choose 8, = u-l(S0), observing that So and T are both independent of t o . T h e proof is complete.
THEOREM 10.8.5. Assume that there exist functions V(t,x, u), g(t, y ) , and A(t) satisfying the following conditions: (i) A(t) > 0 is continuously differentiable for t E J , and A(t)+ co as t + co; (ii) g E C[J x R , , R],and g(t, 0) = 0; (iii) V E C[H x R N ,R,], V(t,x, u ) possesses continuous partial derivatives V , , V, , V,, , V, in H, and (10.8.1) holds; (iv) f E C [ H x R N x Rn x Rn', R N ] ,G E C[H x R, x Rn x Rnz,R], G is elliptic, and
(v)
< G(t,x,4 t ) w, x,4, A ( t ) vz , A ( t ) V z z ) ; G(t,x, z, 0 , O ) < g(t, 4,27 3 0.
Then, the equistability of the trivial solution of (10.8.2) guarantees the equi-asymptotic stability of the trivial solution of (10.6.1). Proof. Let E > 0 and to E f be given, and let u = min,,, A(t). By assumption (i), G > 0. Set 7 = o ~ ( E ) . Assume that (S,*) holds. Then, given 7 > 0, to E J , there exists a 6 = 6(to , E ) such that ~ ( ttn, yo) 9
whenever y o
t
(10.8.8)
to,
< 6. Let u ( t , x) be any solution of (10.6.1) such that A(t0) V(t0 P x,+(to
and
< 7,
>
4)G Yo
7
A ( t ) v(t,x,+ ( t , x)) G r(t, t o >Yo)
x E pto
9
on
aN.
194
CHAPTER
10
Then, by Theorem 10.7.2, it follows that
<
A(t) V ( t ,x, u(t, 4) r ( t , to ,yo)
on
(10.8.9)
H,
where r ( t , to ,y o ) is the maximal solution of (10.8.2). We choose yo so that a(to , 11 +(to , -)]I pt) A(to)= yo . Then, we can assert the existence of a 8 = 8 ( t o , E ) such that the inequalities
/Id(to , .)I/Pto < 8
and
A(t0)4
0
9
/I+ ( t o
7
.)IIP,,>
<6
hold together. With this 8, the equistability of the trivial solution of (10.6.1) is verified. For, otherwise, we arrive at the contradiction, proceeding as in the proof of Theorem 10.8.1, +I =
G K E )
,< 4,) Vtl
7
xo 3 4 t l
,xo))
< Y(t1
7
to ,Yo)
< 17.
For a fixed E = , we designate by 6, = 8,(t0) the number 8 ( t o , el). Let 0 < E < el and to E J be given. Suppose that 11 +(to,-)lip 6, . Since t0 A(t) 00 as t -+ 00, there is a number T = T ( t o ,E ) satisfying
<
---f
b(6) A ( t ) > 17,
t
2 to
+ T.
(10.8.10)
It is easy to show that, with this So and T , the quasi-equi-asymptotic stability holds. Suppose that this is not true. Then, there exists a sequence (tk},t, >, to T , and t, co as k + c o such that 11 u(t,, .)I1 3 E for Ptk 6, and II+(t, -)[IaH < E , some solution u ( t , x) satisfying ll+(to, -)/Ip t 0 t >, to + T . Also, there exist xlj E int Ptk such that 11 u(t, , xk)lI = E . T h e relations (10.8.8) and (10.8.9) yield
+
---f
<
4 t k )
b ( 4 < 17,
which is an absurdity in view of (10.8.10). It therefore follows that the trivial solution of (10.6.1) is equi-asymptotically stable, and the proof is complete. Let us consider the following example. Let L(u) denote the differential form
Jw c %(t, n
=
j,k=1
+ c bk(t, n
X)%j”r
4 u Z k
9
k=l
where aj,rc(t,x) and bk(t,x) are continuous functions on quadratic form n
(10.8.1 1)
g, and the ( 10.8.12)
10.8.
STABILITY AND BOUNDEDNESS
195
on R, h being an arbitrary vector. Let F E C [ H x RN,RN].Consider the system (10.8.13) ut = L(u) F(t, x,u).
+
Assume that F(t, x, 0)
= 0 and
f
i=l
u W ( t , x,u )
< A(t)
(ui)2, <=1
where h E C [ J ,R]. Taking V ( t ,x, u) = 11 u inequality (10.8.12), we obtain
and making use of the
it follows, from Theorem 10.7.1, that
provided
where If, in addition,
the application of Theorem 10.8.1 yields equistability of the trivial solution of (10.8.13). On the other hand, the assumption
1, h(s)
ds = -0
implies the equi-asymptotic stability of the trivial solution of (10.8. I3), by Theorem 10.8.3. Regarding the boundedness of solutions of the partial differential system (10.6.1), we have the following result.
196
10
CHAPTER
THEOREM 10.8.6. Assume that there exist functions V(t,x, u ) and
g(t, y ) fulfilling the following hypotheses:
(i) V E C[H x RN,R,], V ( t ,x, u)possesses continuous partial derivatives V , , V , , V,, , V , in H , and b(ll u II)
-
< v(t,x,4 < 44 ll u II),
(I 0.8.14)
where b E .X, b(y) 4 co as y a,a E C[J x R, , R+],and a E Zfor each t E J ; (ii) f E C[H x R N x R" x R"', RN], G EC[H x R, x Rn x Rnz,R], G is elliptic, and
(iii) G(t,x,x, 0, 0 )
< g(t, x), x 3 0, where g E C [ J x
R, , R ] .
Then, the equiboundedness of Eq. (10.8.2) implies the equiboundedness of the system (10.6.1).
<
Proof. Let a > 0 and to E J be given. Suppose that 1) $ ( t o , .)lip a. t0 Define u1 = a(to ,a). Let Eq. (10.8.2) be equibounded. Then, given a1 > 0 and t, E J , there exists a P1 = &(to , a ) such that ~ ( tt o, 3 yo)
if yo < a1 . Since b(y) + co as y that
b(P)
9
( 10.8.15)
t bto,
co, we can find a P
= B(to , a ) such
2 PI.
(10.8.16)
Let u(t, x) be any solution of (10.6.1). Then, if (10.8.4) holds, we have (10.8.5). Choose yo = a(t, , 11 +(to , -)IIp, ). Suppose that there exists a solution u(t, x) of (10.6.1) such that
(a) I / +(to > . ) / / P ( o (b) It 9(t, .)llm
<
01,
P, t 2 to
9
and 11 u ( t , , .)itp f l 3 /3 for some t , > to . This implies the existence of an xo E int Ptl satisfying I/ ~ ( t, xo)li , = 8. We are now lead to an absurdity, in view of the relations (10.8.14), (10.8.5), (10.8.15), and (10.8.16): b(P)
< v(t, xo , ~ ( t ,xo)) < r ( t , , to >
9
9
yo)
< B1 < b(P).
This proves the equiboundedness of the differential system (10.6. l), and the theorem is proved.
10.8.
STABILITY AND BOUNDEDNESS
197
On the strength of Theorem 10.8.6 and the parallel boundedness and Lagrange stability results in ordinary differential equations, the following theorem on various boundedness results and Lagrange stability may be proved. We only state the theorem, leaving the construction of the proof as an exercise.
THEOREM 10.8.7. Let assumptions (i), (ii), and (iii) of Theorem 10.8.6 hold, except that the relation (10.8.14) is strengthened to b(ll u 11)
< v(t, 4 < 41u 11)) X,
where a, b E X and b(y) + co as y -+ 03. Then, if the ordinary differential equation (10.8.2) satisfies one of the notions (B,*) to (&,*), the partial differential system verifies the corresponding one of the concepts (B,) to (BJ. Corresponding to the system (10.6.1), we may consider the perturbed system U t = f(t, X, % u:, 4%)F(t, X, 4, (10.8.17)
+
where F(t, x,u ) is a perturbation term. Then, we have the following result.
THEOREM 10.8.8. Assume that (i) V E C[H x R N ,R+],V ( t ,x, u) possesses continuous partial derivK , and atives V , , V,, V Z z ,V , in H , j ! aV(t, x,u ) / & 11
<
Wll * II)
< w,4 < 4 z l II)? X,
where a, b E X ; (ii) f E C [ H x R N x Rn x Rn', R N ] , F E C[H x R N , R N ] , G E C [ H x R, x Rn x Rnz,R], G is elliptic, and
for some 01 > 0; (iii) G(t, x,x, 0, 0 ) g(t, 0 ) 3 0 ; (iv) llF(t,x,u)ll
< g(t, z),
z
>, 0, where g E C [ J x R , , R],
< vV(t,x,u), and a 3 Kv.
Then, one of the stability notions of the trivial solution of (10.8.2) implies the corresponding one of the stability results of the trivial solution of the perturbed system (10.8.17).
I98
Proof.
CHAPTER
10
Using the respective assumptions in (i), (ii), and (iv), we find
I t is evident, from this inequality, that we can directly apply Theorems 10.8.1-10.8.4 to obtain the desired result. T h e proof is complete. Although we can prove a number of results by the techniques just used, that is, by reducing the study of partial differential system to the study of ordinary differential equations, in certain situations, this method does not yield all the information about the given system. For instance, consider again the example (10.8.11). Suppose we now assume that the ui,k(t, x) hihk is positive definite instead of positive quadratic form semidefinite, as demanded in (10.8.12). This stronger hypothesis has no effect. I n other words, we do not get more information because of this assumption. T o be more specific, suppose F = 0 so that g = 0. Then, we can conclude by Theorem 10.8.1 that the trivial solution of (10.8.1 1) is stable. This conclusion remains the same even when the preceding quadratic form is assumed to be positive definite. I n such situations, the following theorem is more fruitful.
xF,k=l
THEOREM 10.8.9. Assume that (i) V E C[H x R N ,R,], V ( t ,x, u ) possesses continuous partial derivatives V , , V , , V,, , Vt,in N,and b(ll u II)
< v(t,x, u ) e 44 II u Ill,
where b E Z, a E C [ ] x R, , R,], and u E X for each t E f; (ii) f E C[H x RN x Rn x Rn2,RN], G E C[f7 x R, x Rn x Rn2,R ] , G is elliptic, G(t,x, 0, 0, 0) = 0, and
10.8.
where g E C[J x I?, r(t, t o , 0) of
199
STABILITY AND BOUNDEDNESS
,R ] , g(t, 0) E 0, and the maximal soiution Y’
= &Y),
Y(t3
0
=
is identically zero. Then, the equistability of the trivial solution of zt =
(10.8.18)
G(t,x, z, z, %), 9
implies the equistability of the trivial solution of (10.6.1). Proof. Let u(t, x) be any solution of (10.6.1) such that
and
V(to,x, +(to
,.I)
,< z(t0
9
4
v(t,x, +(4 x)) < z(t, x)
on
Pto
on
aff,
where x ( t , x) >, 0 is the solution of (10.8.18). Define m(t, x)
V ( t ,x, u(t, x)).
=
Then, we get
< G(t,x, 4 4 4, m d t , 4,mm&, 4). If we write T[v]= v t - G(t, x,v, v,, vZJ, then it is clear that T[m]< T [ z ] . Furthermore, m(t, x) < z(t, x) on PL0u aH. All the assumptions of Theorem 10.2.2 being verified, we deduce that V ( t ,x, u(t, x))
< z(t, x)
on
R.
Let E > 0 and to E J be given. Assume that the trivial solution(l0.8.18) is equistable. Then, given b ( ~ > ) 0 and to E J , there exists a 6 = S(to , E) such that
<
(9 max,,pt, X ( t 0 > x) 8, (ii) maxZEaHx(t, x) < E , t b to , implies
max z(t, x) XSP,
< E,
t 3 to
Let maxZSp ~ ( t ,, x) = .(to, 11 # ( t o , -)[!pto),and let 6, = &(to, E) be the 50 same number chosen according to the inequalities (10.8.6) in the proof of Theorem 10.8.1. Suppose that I I + ( t o , ,< 6, and II+(t, - ) / I aH < E , to
200
CHAPTER
10
t 3 t o . Assume that there exists a solution u(t, x) of (10.6.1) such that, for some t , > t o , I/ u(t, , -)I/ 3 E . It then follows that there is an xo E int Pil satisfying / j u(t, , xo)lI = E . From this, we deduce the inequality VE)
< L’(t1 , xo , 4 t , , xo)) < Z(t1 , xo) < b ( E ) .
This contradiction proves the equistability of the trivial solution of (10.6.1). On the basis of this theorem, we can formulate other stability results in this setup. We notice, however, that we now have the problem of knowing the stability behavior of partial differential equation (10.8.18). I n the cases where the function G(t,x,x, x, , x,,) is simple enough to know the behavior of its solutions by other methods, this technique is useful.
10.9. Conditional stability and boundedness I n this section, we shall consider the partial differential system of the type (10.9.1) U f = f(4 x, % u, , u,,), where U, =
(F
7-7
au2 __
au,
--
ax,
7
ax,
,*--,
au, __
ax, ’-*’
au, ax, ,-*., -)ax,
au, __
and
azu,
azu,
a2u,
ax, ax, ’...’ ax,z
’*..’
a2u, ax,2
**..’
w). a2uN
I t will be assumed that the first initial-boundary-value problem with respect to (10.9.1) admits the trivial solution and that all solutions exist on H . I n the sequel, a solution of (10.9.1) will always mean a solution of the first initial-boundary-value problem. Let k < N and M(N--k)denote a manifold of ( N - K ) dimensions containing the origin. Let S(a) and represent the sets, as before,
s(a)
S ( a ) = [ u : (1 u (1
< a]
and
S(a) = [u : 1) u (1
< a],
respectively. Parallel to the conditional stability and conditional boundedness definitions (C,) to (C16)of Sect. 4.4, we can formulate the definitions of conditional stability and boundedness of the trivial solution of (10.9.1.). Corresponding to (C,), we have
10.9.
CONDITIONAL STABILITY AND BOUNDEDNESS
20 1
DEFINITION 10.9.1. T h e trivial solution of the partial differential system (10.9.1) is said to be conditionally equistable if, for each E > 0 and to E J , there exists a positive function S ( t , , E ) , which is continuous in t n for each E, such that, if (i) ). M(N--k) (ii) $(t,4 c S ( E ) , ( t , 4 E aH, 9
9
then u(t, x)
“lo
c S(E),
>
( t , x) E H.
Sufficient conditions for the conditional stability of the trivial solution of (10.9.1) are given by the following result.
THEOREM 10.9.1. Assume that (i) g E C [ J x R+N,R N ] , g ( t , 0 ) = 0, and g ( t , y ) is quasi-monotone nondecreasing in y for each t E J ; (ii) V EC[H x R N ,R + N ] , V(t,x, u ) possesses continuous partial derivatives V t , V , , V,, , V , in H , and b(lI u 11)
where b E X , a (iii)
E
N
< 1 vik x,4 < 4 4 II u Ill, i=l
C [J x R+ , R+], and a E X for each t E J ;
f E C[H x RN x RNnx RNn2,R N ] ,G E C[H x R,N x RN x RN2,R N ] , G(t,2,
2, z:,
ZLJ
is elliptic, and
av
-
at
av + au .f(t, x,
u, u,
, uzz) < G(t,X, v(t,X, u ) , I/aci, VL);
(iv) G(t,x,x, 090) < g(t, 4, z 3 0; (v) Vi(t,x, u ) = 0 (i = 1, 2 ,..., k), k < N , if u E , where M(N--k) is an ( N - k)-dimensional manifold containing the origin. Then, if the trivial solution of the ordinary differential system y‘
= g(t,y),
= yo
>, 0,
to
2 0,
(10.9.2)
is conditionally equistable (in the sense of Definition 4.4.2), the trivial solution of the partial differential system (10.9.1) is conditionally equistable.
202
Proof.
10
CHAPTER
For any
> 0,
E
if
11 u 11
=
we have from assumption (ii) that
E,
(10.9.3) Suppose that the trivial solution of (10.9.2) is conditionally equistable. ) 0 and to E J , there exists a 6 = 6(to , c) > 0 such that Then, given b ( ~ >
c Yz(4 to ,Yo) < 44, N
t
2 to
(10.9.4)
9
2=1
provided v
C
y20
<6
and
( i = 1 , 2 )...,k),
0
uzo =
(10.9.5)
2=1
where y ( t , to , y o ) is any solution of (10.9.2). Suppose that u(t, x) is any solution of (10.9.1). It follows by Theorem 10.6.4 that N
N
V 2 ( 4
44 4)< 1 Y , ( t ,
.2^,
to
9
( t , 4 E R,
Yo),
2=1
2-1
(10.9.6)
whenever N
C VZ(4l
N
9
x, d t o
4) < 1Yro >
9
and
c N
V 2 ( 4
x , 4(t, .))
2=1
E
pto>
2=1
2=1
<
N
rz(t, t o ,yo),
(4 ).
E
2=1
aff,
where r ( t , to ,y o )is the maximal solution of (10.9.2). Choose y o such that N zi=lyiO /I $(to * ) / i P , o ) and ,). M(N-k) pto so that yio = Vi(to, x,$ ( t o ,x)) = 0 (i = 1, 2,..., k), by assumption (v). I n view of the properties of a(t, Y), there exists a 6, = 6,(to , e ) > 0 such that 7
Ild.(ro
9
9
7
.)/IPt0
< 61
and
4to
>
II d(to
7
.)llPJ
9
<8
at the same time. From the choice of yio and condition (ii), it follows n M(N-k), we have the inequality that, whenever + ( t o , x) C s(6,) (10.9.6). Suppose now that there exists a solution u(t, x) of (10.9.1) which satisfies
x, s(sl) M(N-I,) > (a) (b) +(4 4 c S(C--),( 4 x) E aH, 9
pto
9
10.9.
CONDITIONAL STABILITY AND BOUNDEDNESS
203
and has the property that u(t, x) $ S(E) for some t, > to and x E Ptl. Because of relation (b), there exists an xo E int Ptl such that 11 u(t, , x,,)lI = E. Hence, by (10.9.3), (10.9.4), and (10.9.6), we are led to the following absurdity:
< c Vi(t1 , xo N
b(c)
i=l
< i=l c Ti(t1 , to ,Yo) < b(4. N
9
, .a>>
U( t 1
Consequently, the trivial solution of (10.9.1) is conditionally equistable, and the proof is complete. On the strength of Theorem 10.9.1 and the parallel theorems on conditional stability and boundedness (in Sect. 4.4), we have the following
THEOREM 10.9.2. Assume that the hypotheses of Theorem 10.9.1 hold, and suppose that a(t, r ) = a(r), a E X . Then, one of the notions (C,*) to (C&) relative to the ordinary differential system (10.9.2) implies the corresponding one of the conditional stability concepts (C,) to (C16). T h e following example, in addition to demonstrating the conditional stability, serves to show that the system (10.9.1) need not be parabolic. Consider the system
au, at
au, at
where F, F, F,
+ cos t ) u , + (1
=
(1
=
(1 - ePt)ul
=
(cos t
~
cos t)u,
+ (cos t
-
l)u, ;
+ (1 + e+)u, + (ct- l)u, ;
&)ul
~
+ (ect - cos t)u, + (e+ + cos t)u, .
204
CHAPTER
10
Assume that the quadratic forms C;j=, aiih,hj , Crj=1bi& , and C:,i=, cijhihi are all nonnegative for arbitrary vector A. Choosing the vector Lyapunov function V = ( V l , V , , V,) such that
we observe that the functions b(r) and a(t, r ) reduce to b(r) = [(u12
+ u: + u1;3)1/2]2
Furthermore, the function G
and =
n(t, r )
=
5[(u,2
+ u: + u32)1/2]2.
(G, , G, , G,) takes the form
T h e differential system (10.9.2) can be reduced to
We find that g = (g, ,g, ,8,) fulfills the monotonic requirements. Choose k = 1. T h e n the solution ~ ( tto, ,yo) of (10.9.2) is given by rl(4 t o Yo)
= YlO
r2(t, to ,yo)
T
7
exp[4(t
-
yzoexp[sin t
to)],
-
sin to],
y3(t, to ,yo) = y30exp[4eKt0 - 4.81.
Here we have A!l(N-,i) = M , , the set of points u such that u2 - u3)2 = 0. I t is clear that the condition (C,") holds, which,
(ul
+
10.10 UNBOUNDED DOMAINS
205
in its turn, implies, from Theorem 10.9.1, the conditional equistability of the trivial solution of the system considered previously.
10,lO. Parabolic differential inequalities in unbounded domains Let D be a region in Rnfl of (t, x) space, satisfying the following conditions: (i) D is open and contained in the zone to < t < m; (ii) for any t, E [ t o , m], the intersection Stl of D with the plane t = t, is nonempty and unbounded; (iii) for any t , , Stl is identical with the intersection of the plane t = t, with the closure of that part of D which is contained in the zone to t < t, .
<
We denote by aD that part of the boundary of D which is contained in the zone to < t < GO. For any 4 E C[D,R ] , if the inequality
DEFINITION 10.10.1.
+(4 4
< M e x p w I/ x IlZ)
holds, where M , K are positive constants, we shall say that 4 E E2(M,K ) . Similarly, 4 E E2(M,K ) implies that
If 4 E E,(M, K ) 4 E &(M, K ) .
and
vq4 x) 3 -M
e x p w II x
4 E &(M, K )
simultaneously, we shall say that
l ”.
DEFINITION 10.10.2. A function - f C~[ D x R x Rn x Rnz,R] is said to satisfy an L? condition if there exist positive constants Li (i = 0, 1, 2, 3, 4) such that If(4 x, u, p , R) - f ( 4 x, u, p,R)I
+ (4II x /I + L,) c I p , n
,=I
THEOREM 10.10. I .
-
< Lo
p, I
c I R,, n
z,k=l
-
R,, I
+ ( L , /I x (I2 + L4) I
u
-
21 I.
Assume that
(i) u, v E C[D,R ] , the partial derivatives u t , u , , u,, , v l , v,, a,, exist and are continuous in D ;
206
CHAPTER
10
(ii) f E C[D x R x Rn x Rnz,R],f satisfies an 9 condition, the differential operator T is parabolic, and T[u]
< T[u]
on
D;
(iii) u E &(M, K O )and v E -? KO); $(&I, (iv) u(t, x) v ( t , X) on St0w ao.
<
Then, everywhere in D,we have (10.10.1)
4 4 x) 6 v(t,.)-
Proof. Let D , denote an open bounded domain separated from D by the cylindrical surface I', , with the equation I/ x I( = R. We shall denote by S 2 , aD, the parts of the surfaces S l 0 ,aD, respectively, lying inside and on Let C, represent the part of contained in D. Furthermore, let us designate by Dh,DRh, aDh, aD,h, CRhthe parts of the sets D, D,, aD, aD, , C, , respectively, contained in the strip to t <. to h . We shall prove the theorem for a domain Dh, where h will be chosen conveniently. Consider the function
r,.
r,
+
<
m = m(t, x)
Define u
=
=
u(t, x) - u(t, x).
UH, v = GH,where
is the growth damping factor. Here p = p ( K ) , y = y ( K ) are positive constants, which will be chosen later, for K > K O .Setting z we have m
=
-
v(t, x),
xH defined in D", where h
=
z(t, x)
C ( t , x)
It then follows that
< f ( t , x, u, uxH + u H = ,R ( l ) ) -f(t, X, iZxH + U H x , R"') ZL,
+ f ( t , x, u, iizH -f(t,
X, FO,
VxH
+ u H x , R(2))
+CHz,
R'3'),
(10.10.2)
10.10
207
UNBOUNDED DOMAINS
where
R'3' = [v,*,,H
+ v,,H,, + G'.,H,* + K * , j l .
Let {RU}be an increasing sequence such that R, -+co as fixed 01, consider the domain D i m .Denote A
01
-+ co. For
a
sup - z(t, x).
=
(t,x)sDh Re
-
+
+
+
Then, there exists a (tl , xl)E D i a = D t x St aDta C t a such that A = ~ ( t, ,xl). We shall show that the constants p, v, and h may be , 0. chosen such that ( t l , xl) E D t a S t a implies the inequality ~ ( t, xl) Indeed, supposing the contrary, we have, at ( t l , xl),
+
az/at (tl
7
x1)
<
3 0,
%(tl
, x1) = 0,
and
for an arbitrary vector A. This means that
Since f is elliptic, it follows that f(t1
+
, x1 , U(t1 ,x1), "a UH,,
w + uH,
9
R'9.
Hence, using the 9 condition and the relation (10.10.2), we get
-
"1
at
=
~ F H (say).
208
CHAPTER
10
From the definition of H , we deduce that 2KLon
+
{l
Setting L
=
PK II x 1 '
- p(t -
to)}'
-
4
1
-
p(t
-
to)
.
max[Li, L,], and, taking into consideration the inequalities
I xi I
I/ x /I < I/ x 11'
< I/ x 11,
+ 1,
1 - ~ ( -t 1,)
< 1,
we have FH
<
H
(1
--
P(t
-
to>Y
[4K2L,n2/ j x (1'
+ 2KLn + L3 I1 x 11' + L4
+ 2KL0n + 4KLn (1 x \I2
- pK
/I x 1 '
-
rU
- ~ ( -t to)}'].
Let us now choose Kp
=
+
4K2Lon2 4KLn
+ L, + A,
and h
=
(I
- P)/p,
A
> 0,
0
'This choice leads to the inequality /3
<1
- p(t -
to)
<1
in
Dh,
and, hence, letting
Y=
2KLon
we obtain QH
+ 2KLn + L, + N
< -AM
B'
/I x 11'
I
-NH
< -NH
N>0,
< 0.
'I'hus, we have a contradiction to the fact that az(tl , xl)/at 3 0. Hence, if ( t l , xl)E D t a LJ S i a , it follows that z(tl , xl) 0. Now, if ( t l , xl) E a D t a , it turns out, by assumption (iv), that z(tl , xl) 0. Finally, suppose that (tl , xl)E C i a . Then,
<
<
10.10.
209
UNBOUNDED DOMAINS
Let ( t , x) be an arbitrary point in Dh. Given E > 0, there exists an > a. , the point ( t , x) E D$ and the righthand side of (10.10.3) is less than E. Since z(t, x) z ( t ; , xl), it follows that z(t, x) E, if a > aO(e). Consequently, x(t, x) 0 in Dh, and, hence, a. = aO(e) such that, for a
<
<
m(t, x)
This implies that u(t, x)
<0
-
Dh.
in
< v(t, x)
<
-
D’L.
on
I n particular, we have the inequality (10.10.1) in the intersection of the closure of DIGwith the plane t to h. Since this intersection, by proposition (iii) in the definition of the domain D, is identical with S t , + b ,we have (10.10.1) for ( t , x) E . We may therefore repeat the arguments starting from the plane t = to h instead of t = to and obtain the inequality (10.10.1) in the intersection of D with the zone to h t to 2h. I n this way, the validity of (10.10.1) at any point of D follows after a finite number of steps. T h e proof is therefore complete.
+
+
+ < < +
THEOREM 10.10.2. Assume that (i) u, ZI E C[D,R ] , the partial derivatives exist and are continuous in D; (ii) fi,fi E C [ D x R x Rn x Rn2,R ] ,
u , , u,,
u,,
, zll,
ZI,, ,,zl
and one of the functions fi , f2 is elliptic and satisfies an 9 condition; (iii) u E E2(M,KO),71 E &(M, KO),and, in D,
(iv) u(t, x) < v(t, x) on S,, u aD. These assumptions imply u(t, x)
< v(t,x)
everywhere in
D.
By repeating the proof of Theorem 10.10.1 with obvious modifications, this theorem can be proved. We leave the details to the reader.
210
CHAPTER
COROLLARY 10.10. I.
10
Assume that
(i) v E C[D,R ] , v ( t , x) possesses continuous partial derivatives v t , v, , v,, in D , and v E &(M, KO); (ii) f E C [ D x R x Rn x Rn2,R ] , f satisfies an 2 condition, the differential operator T is parabolic, and T [ v ] >, 0 in D ; (iii) v ( t , x) 2 O on St0u aD. These hypotheses imply that v(t,x)
0
everywhere in
D.
Consider, as an example, the operator
in J2, = (0, GO) x Rn . Let T be parabolic, and, for i, k ( t , x) E Qo , suppose that
I a,k(t, I).
< Lo
7
44 x)
I
w, 41 < L, II II + L,
< L, /I x /I2+ L, -
x
=
1, 2,..., n,
9
Let us assume that T[u] 3 0 in Q, and that u E &(M, KO)in Q, where Q = [0, 00) x Rn. If u(0, x) >, 0 in Rn, then it follows from Corollary 10.10.1 that u(t, x) 3 0 in Q.
On the basis of the foregoing results, one can state and prove theorems
on comparison principle, bounds and error estimates for equations in
unbounded domains, and corresponding parallel results for systems. We leave them as exercises. 10.1 1. Uniqueness
We shall consider the parabolic differential system Ut = f ( t , x,
u, u:, UL),
(10.11.1)
where f E C[D x RN x R" x R"" RN].
DEFINITION 10.11.1. Given a function 4 E C [ s t oU aD,R N ] ,any function u E C[D,R N ]that has continuous partial derivatives ut , u, , u,, in D and that satisfies (10.1 I . 1) in D such that u(t, x) = + ( t , x) on s1,u aD is called a solution of the first boundary-value problem of Fourier.
10.1 1.
21 1
UNIQUENESS
DEFINITION 10.11.2. We shall say that a vector function f satisfies an 9 condition if, for each i = 1, 2,..., N , _ _ 1 f i ( t , x, u, Pi,RZ) - f i ( t , x, u, pi, Ri) I
c IR:k-R5kl n
n
t ( ~ , l l x l l t L , ) ~ / ~ ~ - ~ / i=1
j,k=l
+ (L3 II x 112 t L4) 2 I ui N
uj
i=l
I.
THEOREM 10.11.1. Iff. C [ D x R N x Rn x Rnz,RN]and satisfies an 2 condition, then the first boundary-value problem of Fourier possesses at most one solution u(t, x) such that each component of u(t, x) belongs to E,(M, KO). Proof. Suppose that there exist two such solutions, u(t, x) and v(t, x). Then the difference
m(t, x) = u(t, x)
-
v(t, x)
obeys the equation -am(t, ).
- f ( t , x,u(t, x),
at
U2(t,
x), uk,(t, x))
-f(4 x , v(t,4, V&> Define u = iiH, v
=
XI,
d,(t, x)).
CH, where
is the growth damping factor and K > KO. We shall retain the meaning of the symbols D, , r, , Sk, aD, , C, and Dh , DRh,aDk, aDRh, CRilas defined in the proof of Theorem 10.10.1. Set z
Then, we have m observe that az
at H
=
=
z(t, x) = G(t, x)
-
v(t,x).
x H defined in Dh, where h < p-l. Moreover, we aH
+zat = ut - vt
GH + G H , , R"') f ( t , x , iiH, u>H + u i H x ,R 2 ) ) +f ( t , X , GH, uTH + U"H,, R ( 2 ) ) - f ( t , x , c H , z H + GH,, R ( 3 ) ) ,
= f ( t , X , GH,
~
(10.11.2)
212
CHAPTER
10
Let {R,} be an increasing sequence, R, 4 GO as we consider the domain D:, . Let us denote A,
=
01
-+
00.
For a fixed a ,
max[ sup - I xi ( t , x)i]. (t,x)eDh Re
-
Then, there exists an index i, and a point ( t , , x,) E D i , such that A , = 1 zi.(t,, x,l). We shall show that p, y , and h may be conveniently chosen such that (t, , x,) E u Sta implies I zi=(t,, x,)l = 0. Let us suppose the contrary. Then, there are two cases to be considered: (a) zi"(t, , x,) (b) ziu(t, , x,)
> 0, and < 0.
I n case (a) holds, we have, for an arbitrary vector A,
and
Since f ie is elliptic, it follows that
CH + Z H x , I?"') , u(t, , x,), G H + .".Hx
f " ( t , , xu , ~ ( t ,,x,),
, X,
Hence, we get, from (10.11.2),
, R"').
10.12.
EXTERIOR
BOUNDARY-VALUE PROBLEM
213
<
because of the 2 ' condition and the fact I zj(t, x)I Nzia(t,, x,). We now proceed, as in the proof of Theorem 10.10.1, to show that azi.(tu , x,) at
< 0.
The choice p, h, and y would be as follows:
+
p K = 4K2Lon2 4KLn
and
h
Y=
=
(1
X
> 0,
0 < p < I,
- /3)/p,
2KLon
+ L,N + A,
+ 2KLn +L,N + Nl , B2
N, >O.
Thus, we have a contradiction to azia(t, , x,)/at >, 0. If zia(t,, x,) < 0, we repeat the foregoing reasoning with zia(t, x) = -zi-(t, x), and, again, we get a contradiction to azia(t, , x,)/at 3 0. If (tu , x), E aD; , $-(fa, xe) = 0 by assumption. Finally, if (tm,x,) E Cia, the;
-
<
M exp(KoR2) exP{[KR,2/1- to) ] -
+ rtm>.
Since I zi(t, x)I I zi=(t, , xu)\ in D;, , again - arguing as in Theorem 10.10.1, we can show that zi(t, x) = 0 in Dh. Consequently, it follows that mi(t, x) = 0 (i = 1, 2, ..., N ) in D, proving the uniqueness of solutions in Dh. The validity of the uniqueness of solutions at any point of D may now be proved after a finite number of steps as in Theorem 10.10.1. This proves the theorem. 10.12. Exterior boundary-value problem and uniqueness
Let A be a bounded and closed domain R n ,and let S be its complementary domain. Assume that the boundary ad of A is represented by the equation T(x) = 0, (10.12.1) where r(x) is a continuously differentiable function in 3, having bounded second derivatives in S and satisfying the condition
1 grad T(x)l 2 To > 0.
(10.12.2)
214
CHAPTER
10
We define, for t, >, 0, D
=
and
( t o , a)x S
c =
( t o ,m) x aA.
(10.12.3)
For every ( t , x) E u and each i = 1, 2, ..., N , let T~ be a straight half-line entering the interior of D and parallel to the plane t = t, . It is assumed that there exists a positive constant yo such that COS(Ti,
no)
yo
(i
1, 2,..., N ) ,
( t , X)
E
0,
where n, is the normal to u directed to the interior of D . We consider the partial differential system Ut =
where f E C[D x
RN
f(4 X,
u, u,i,
4m),
(10.12.4)
x Rn x Rn2,RN].
+
DEFINITION10.12.1. Given an initial function E C [ S ,R N ] ,a solution of (10.12.4) is any function u E C RN] satisfying the following conditions:
[o,
(i) ~ ( t, x , ) = +(x) on 5'; (ii) &(t, x ) / d exists ~ in (5, and [du(t, x )!&]
+ G(t,x, u(t, x))
=
0
on
u,
where G E C[u x RN,R N ] ; (iii) u(t, x) possesses continuous partial derivatives u t , u, and satisfies (10.12.4) in D.
, u, in D
T h e problem of finding a solution of (10.12.4) is called an exterior boundary-value problem. T h e following theorem concerns the uniqueness of solutions of such a problem.
THEOREM 10.12.1. Assume that (i) .f E C [ D x R N x Y' condition;
R n x Rn2,R N ] ,f is elliptic and satisfies an
(ii) G t C[u x K N , R N ] ,and, for each i
=
1, 2 ,..., N ,
Then, the exterior boundary-value problem has no more than one solution belonging to class E2(Ml , K O )in D .
10.12.
215
EXTERIOR BOUNDARY-VALUE PROBLEM
Proof. Suppose, if possible, there are two solutions u ( t , x) and u(t, x) of (10.12.4) belonging to E2(Ml , KO)in D. Then, m(t, x)
= u(t, x) - a(t, x)
belongs to E2(M,KO),where M We may assume that
2M1.
=
r(4= II x II
for
II x II > Ro
9
where R, is the radius of the sphere 11 x I I = Ro situated in Rn and containing the boundary ad in its interior. It then follows that there exist two constants A, B > 0 such that
I
rz,(x) j=1
I<
A
I
and
rzjS(x) j.k=l
We shall use the function H ( t , x; K) = exp
where
K [ r ( x )- PI2 - r.(t - to)
(1
p = 4KLoA2
and v = max
I
(KL,,B
I<
+ vt),
in
B
K
+ 2L1A + NL3K+ 1 ’
S . (10.12.5)
> K O , (10.12.6) (10.12.7) (10.12.8)
+ KLIAp +L,PN)~+ 2KLoA2+ L3p2N+ L4N + 1 9
Y2
y being arbitrarily chosen so that
0
(10.12.9)
First of all, we shall prove the theorem for the part
where We shall write
Dho = ( t o ,to
ha
+ ha] x S =
(1
- Y)/P.
of
D,
(10.12.10)
216
CHAPTER
10
It will be shown that the function H(t, x; K ) verifies the inequality FH
< -H(t,
x,K )
for
( t , x) E DhO.
(10.12.12)
Indeed, (10.12.10) yields O
(10.12.13)
I n view of (10.12.5), (10.12.6), and (10.12.13), we get
Hence, by (10.12.9), there results (10.12.12). If / / x 11 R,,then, from (10.12.8) and (10.12.14), it follows that
<
FH
<
H ~
t1
-
i.(t -
tdl2 H I r
~
P I - (KLOB
+ ~ A ( J 3+0 L2))I2
+ (KLoB + KA(&& + Lz))' + 2KL,A2 + N[L&,' + La)
~
vy'}.
By virtue of (10.12.9), we see that the inequality (10.12.12) is again verified. Let us introduce the functions u and such that u
=
z, =
Set z = z(t, x)
= zi
u(t, x )
=
CH(t, x; K ) ,
v(t, x) = uH(t, x; K ) .
-
a, so that
m
=
m(t, x) = z(t, x) H ( t , x;K ) .
(10.12.15)
(10.12.16)
Let us choose an increasing sequence {Ra}, R, > R, , and R, + w as Denote by Slo the part of the boundary of D lying on the plane t = t o . Furthermore, let D ~ and o S$ denote the parts of the
a + co.
10.12.
217
EXTERIOR BOUNDARY-VALUE PROBLEM
domains Dho and Sto contained inside the cylindrical surface C, with the equation 11 x I( = R, . Put Cto = Dho n C, . Define A,
=
max[ max 1 zi(t, .)I]. 2 (t,Z)€DhO
a
Observe that the sequence {A,} is nondecreasing and A, 0 for every 01. Hence, it is enough to show that A, + 0 as a -+ co. For this purpose, notice that, for any a, there exists an index i, and a point ( t , , x,) E 0,"s such that A, = I A ( t , , x,)/. T h e following cases are now possible: (i) (G 9 %>E $0 ( 4 ( t , , x,) E D? (iii) ( t , , x,) E aha, (iv) ( t , , x,) E C ~ O . 7
9
Evidently, in case (i), zi-(t,, x,) holds, we have either or
(iia) zi=(t,, x,)
=
(iib) zi,(t,, x,)
>0
(iic) zi-(t,,x,)
< 0.
or
=
0, and therefore A,
= 0.
I n case (ii)
0
Clearly, case (iia) implies A, = 0. If case (iib) is true, then, using a similar argument as in the proof of Theorem 10.10.1, we arrive at the inequality 0
<
aziu(ta
J
xrX)
at
H ( t , , x,; K )
< z'e(t,,
x,)FH(t,, x,; K ) ,
(10.12.17)
and thus, by (10.12.12), the impossibility of case (iib) follows. T h e case (iic) may be reduced to case (iib) by the substitution 2, = - A When case (iii) holds, we have again three possibilities: (iiia) zia(t, , x,) (iiib) .&(t, ,x,) (iiic) &(t, , x,)
=
0,
> 0, < 0.
If case (iiib) holds, then (10.12.18)
218
CHAPTER
10
since, otherwise, there would exist a point ( t l , xl) E int Dho such that z*=(t,, XJ > zi=(t,, x,), and this contradicts the definition of zi=(t, , x,). On the other hand, we obtain, by the relations (10.12.15), (10.12.16), and the condition (ii) in the Definition 10.12.1,
whence, according to assumption (ii) of the theorem, we get
(10.12.19)
The sign of T(x)in (10.12.1) may be so chosen that
9 1 grad r ( x ) i ax,
COS(X(,
no).
(10.12.20)
Hence, taking into account the relations (10.12.1), (10.12.2), (10.12.7), (10.12.19), and the fact COS(T~ , no) 3 yo , we derive the inequality
< z z N H -2Kproyo ( + NL) --zi"H
.~ -
< 0,
which contradicts (10.12.18). We can repeat the same reasoning for the function .9== -zi= to show that case (iiic) is also impossible. We have now proved that, in each of the cases (i), (ii), and (iii), A, = 0. When case (iv) holds, we obtain, by (10.12.16) and the fact that m E &(M, KO),the estimate
Consequently, in all cases, A, -+ 0 as a+ co. T h e validity of the uniqueness of solutions at any point of D may now be proved after a
10.13.
219
NOTES
finite number of steps as in Theorem 10.11.1. T h e theorem is therefore proved. 10+13. Notes
Lemmas 10.1.1 and 10.1.2 are new. For theorems of the type 10.1.1 and 10.1.2, see Friedman [22], Mlak [3], Nagumo and Simoda [I], Szarski [8], Walter [XI, and Westphal [I]. For a systematic use of inequalities in the theory of partial differential equation, see Picone [I]. Theorem 10.2.1 is taken from Lakshmikantham [5]. For Theorems 10.2.2 and 10.2.3, see Brzychczy [I], Mlak [6], Szarski [8], and Walter [8]. Theorem 10.2.4 is due to Maple and Peterson [l]. T h e results of Sects. 10.3 and 10.4 are adapted from similar results in Szarski [8] and Walter 181. Section 10.5 consists of the work of Maple and Peterson [I]. For the results analogous to the results of Sect. 10.6, see Szarski [8] and Walter [8]. T h e results of Sects. 10.9 and 10.10 are due to Lakshmikantham [5]. See also Bellman [2], Mlak [2], and Narasimhan [l]. Section 10.11 contains the work of Lakshmikantham and Leela 111. T h e rest of the chapter contains the results based on the work of Bes& 1-41, Krzyzanski [2, 81, and Szarski [4, 5 , 71. For related results, see Aronson [l-31, Aronson and Besala [I], Barrar [l], Brzychczy [2], Cameron [I], Ciliberto [2], Eidelman [I-51, Foias et al., [2], Friedman [I-241, Ilin et al. [l], Ito [I], Kaplan [I], K n m a t s i i rLI,)l l Krzvzannki rl-In1 r l l . 1 ax and Milgram L^ -",, Tadvzhenskaia J[I], Lees and Protter [l], Lions and Malgrange [l], McNabb- [I], Milicer-Gruzewska 11, 21, Mizohata [l-31, Mlak [5, 6, 8, 10, 161, Miirakami [ I , 21, Nickel [l-31, Nirenberg [I, 21, Pini [l-31, Pogorzelski [ l , 21, Prodi [l-31, Protter [2], Serrin El], Slobodetski [I], Smirnova [l], Vishik [2], and Zeragia [I, 21. I-____._
---I
---J
L-37
This page intentionally left blank
Chapter 11
11 .O. Introduction This chapter is concerned with hyperbolic differential equations. Main results associated with hyperbolic differential inequalities are discussed, including under and over functions. Certain uniqueness criteria, growth, and error estimates are treated utilizing the comparison principle.
11.1. Hyperbolic differential inequalities Let f E C[R, x R3, R ] , where R, is the rectangle defined by = [O x a, 0 y b ] . We shall denote by P+ the expression 4,?/ - f(X, y , 4, 4% 4J in what follows. A fundamental result in hyperbolic differential inequalities is the following.
< <
R,
< <
7
THEOREM 11 . 1 . 1 . Assume that (i) u, v E C[R, , R ] , the partial derivatives u, , uI , uzy , v, , vy , vxy exist and are continuous in R, ; (ii) f E C[R, x R3,R ] , f(x, y , u, p , q) is monotonic nondecreasing in u, P , q, and, in R, PU < pv; 9
0
(iii) u(0,O) < v(0, 0), uz(x,0)
<x
< vz(x, 0),
and u,(O, y )
< $0, y ) for
Th e n we have, on R , , @,Y)
< V(.,Y),
.,(.,Y)
< VZ(X,Y), 221
UY(X,Y)
< V,(.,Y).
(11.1-1)
222
CHAPTER
11
Pmof. I t is enough to show that, in R, , u,(.y.,
Y)
and
< %(.Y Y )
u&, Y )
< %(X, Y ) .
Assume that this is false, and let to be the greatest lower bound of numbers t > x + y such that the two inequalities are satisfied for x -1y < to . T h e n there exists a point (xo,yo) with xo yo = to for which one of the inequalities is not true. Let us suppose that
+
( I 1.1.2)
u d x n Yo) = a d ~ 7oYo>-
Clearly, by assumption (iii), y o > 0. We have yo
zL,(xg,
and 0.4
-
< v z ( x o , yo
h)
< a ( ~ oyo),
,yo)
h
- A),
< nu,(xn,Yo).
u , ( ~ ,yo) o
9
> 0,
( 11.1.3)
(1 1.1.4)
From (11.1.2) and (11.1.3), we obtain (11.1.5)
u d . y o 7 Ya) 2 azdxo , Yn).
Furthermore, since f is monotonic nondecreasing in u , p , q, it follows, from ( I 1.1.4), that f(.yn
>yo u ( . ~ oyo), u d x o , Yo), uu(xn Y O ) ) : ’ : : f(.yo yo .(.yo Yo), u,(xo yo), au(xo yo)). 7
3
7
7
I
9
7
< Pv,yields
This, together with ( I 1.1.5) and the assumption Pu f(xn
,Yn
7
Yo), uA.yn yo), ndxn Y O ) ) , yo 9 4 , y o ,yo), ar(-xo yo), a,,,(xo yo)),
.(.yo
,:’f ( . y o
7
7
t
which is an absurdity, in view of ( I 1.1.2). It is clear that, if we suppose that u u ( x o ,y o ) = uu(xo,yo) instead of (1 I . 1.2), an argument similar to the foregoing leads to a contradiction. Hence, the desired result (1 1 . I . 1 ) is true. ’
DEFINITION 1 1.1.1. A function u E C[H, , R ] , possessing continuous partial derivatives u , ,~ u!, , u , ~ ! ,in KO and satisfying the hyperbolic differential inequality f ( . T , Y , u , 11,. uy) Zi,,, 7
in I<,, , is said to be an under .filmtion with respect to the hyperbolic differential equation ~ ~ . / ; 1 f(.r, Y , 21, z4.z > .,I. ( 11.1.6)
11.2.
223
UNIQUENESS CRITERIA
If, on the other hand, u satisfies the reversed inequality, it is said to be an over function. T h e following result can be proved by a repeated application of Theorem 11.1.1.
THEOREM 11.1.2. Let u, w be under and over functions with to the hyperbolic differential equation (1 1.1.6), respectively. Suppose that v is a solution of (1 1.1.6) such that 4090) < @,O)
a x , 0) < UY(0,
%(X,
Y ) < QY(0,
< 4 0 , o>, 0) < %(X, o>, Y ) < WY(0, Y).
Then we have, on R, ,
THEOREM 11.1.3. Assume that (i) u, v E C[R, , R], the partial derivatives u, , uy , uXy, v, , vy , vxy exist and are continuous in R, ; (ii) f E C [ R , x R2, R ] , f(x, y , u , p ) is monotonic nondecreasing in u, and, in R, , Pu < Pv,where P+ = $xy - f(x, y , 4, 4,); (iii) u(0,O) < v(0,0) and u,(x, 0) < v,(x, 0) for 0 x a.
< <
Then we have, on R , , .(.,Y)
< +,Y)
and
%(.,Y)
< .,(.,Y)-
This theorem is a special case of Theorem 11.1.1. However, it is to be noted that the function f needs to be monotonic in u alone, instead of u and p , as one might expect. T h e reason is obvious, if we follow the proof of Theorem 11.1.1. 11.2. Uniqueness criteria We wish to consider a general uniqueness theorem of Perron type for the hyperbolic differential equation u y ,
= f(X, Y , u, u,
9
UY),
(1 1.2.1)
224
where f
CHAPTER
E
11
C[R, x R3, R], subject to the conditions U(X, 0) = U(X),
U(0)
U(0,y ) = T( y ) ,
=
T(0) = Ug ,
the functions u and T being Lipschitz continuous on 0 0 y b, respectively.
< <
< x < a,
THEOREM 11.2.1. Assume that (i) f E C[R, x R3, R] and
I f ( X , Y , u, P , 9 ) -f(x, Y,u, P, 411 < g ( x , y , l u Ul, I P -1; I, I 9 4 0 , where g E C[R, x R+3,R,], g(x, y , 0, 0,O) E 0, g is monotonic nondecreasing in z , p , q and bounded; ~
~
(ii) z(x, y ) 5 0 is the only solution of the hyperbolic equation Z,u
such that
x(0,O)
=
0,
= g(x,
(1 1.2.2)
Y,x,z, 4, 9
z(x, 0) = 0,
x(0,y ) = 0.
(1 1.2.3)
Then, there is at most one solution for Eq. (1 1.2.1). Proof. Suppose that there exist two solutions u ( x , y ) and v(x,y) for Eq. (1 1.2.1) on R, . We define A h Y ) = I .(.,Y) W , y ) = I .,(.,Y)
-
4x,Y)l,
- %(x,Y)/,
C(X,Y) = I .,(x,r)
-
%(.,Y)l.
Since we have u(x, 0)
=
v(x, 0)
=
a(x),
u(0,y) = v ( 0 , y ) = .(Y),
u,(x, 0)
U,(O,Y)
= =
zt,(x, 0)
v,(O,y)
= a‘@),
=
T’(Y),
it follows that A(0,O) = 0,
B(x, 0 ) = 0,
Furthermore, by condition (i), we obtain
C(0,y ) = 0.
0 0
< x < a, < Y < b,
1 1.2.
225
UNIQUENESS CRITERIA
We note that A ( x , y ) is continuous, B ( x , y ) and C ( x , y ) are uniformly Lipschitz continuous in y and x, respectively. Let us define the sequence of successive approximations to the solution of (1 1.2.2) and (1 1.2.3) as follows:
r>, p n ( X , t ) , Y n ( X , t > ) d t ,
pn+1(.,
Y > = J y g(x, t , an(.,
Yn+l(x,
Y > = Sog(s,Y * an(s, Y ) ,pn(s, Y ) ,Yn(s, Y ) ) ds*
<
<
< y1 , and g is nondecreasing in z , p , q, it
Since a. a l , Po P1 , yo follows by induction that an
<
mn+19
Pn
<
Pn+1>
Yn
<
yn+l*
Also, the functions a n , Pn , yn are uniformly bounded in view of the fact that g is assumed to be bounded. Hence, we get limn+man(x, y) = Y(X, Y ) , uniP(x, Y ) , and 1imn-m Yn(x,y ) a(x, y ) , limn+rn Bn(x, y ) formly on R, . It is easy to see that a(%,y ) is a solution of ( I 1.2.2) and (1 1.2.3). Consequently, on Ro ,
< .(.,Y), B(x7 Y ) < P(x, Y ) , Y ) < r(x9 Y ) .
+,Y)
C(X9
By assumption, identically zero is the only solution for the problem
(1 1.2.2) and (11.2.3). This proves A(%,y ) 5 0, B(x, y ) = 0, and C(x, y) = 0. T h e theorem is therefore proved. T h e next uniqueness theorem is under Nagumo's condition, and the interest is rather in the elementary method employed in the proof.
THEOREM 11.2.2. Assume that (i) f ( x , y , u, p , q) is defined for (x,y ) E Ro and u, p , q E R, and xy If(.,Y,
u1
< "(X,Y)
,Ply
41)
-f(.,Y,
I u1 - u2 I
u2
+&Y)
,Pi?,9211 I Pl
-P2
I + Y(X9Y)Y I 41
~
42 I,
226
CHAPTER
where a , p, y n+P+y=l; (ii) x
and
11
3 0 are continuous functions on R, such that
< I 41
I f ( X , 0, u,P, 41) - f @ , 0, u, P,9z)l
G
Ylf(O,Y,u,P,,4)-f(O,Y,u,P,,4)l
IP1
-
92
I,
-P,I-
Then, there is at most one solution on R, for Eq. (11.2.1). Proof. Suppose that u(x,y ) and u(x, y ) are two solutions of (1 1.2.1) existing on R, . Then, u(x, 0)
=
v(x, 0)
= u(x),
u,(x,
0)
=
0)
?I&,
=
u’(x),
0
< x < a.
It can be shown that u y ( x ,0) = zfy(x, 0)
For, putting y
=
for
0
< x < a.
0 in the partial differential equations satisfied by
u(x, y ) and v(x, y ) , we get the ordinary differential equation
w’= f(x, 0, u(x), o ‘ ( x ) , w).
(11.2.4)
We observe that uJx, 0) and uul/(x, 0) are solutions of (1 1.2.4) satisfying the initial condition u,(O, 0) = u,(O, 0) = ~’(0).Furthermore, the function f of (1 1.2.4) verifies the following condition: x
1 . k O,o(x),
a’(x),
4 -f(xt
0, d x ) , d x ) ,
Wl)l
< I w - w1 I.
This being exactly Nagumo’s uniqueness condition, it follows that u,(x, 0 ) == vy(x, 0)
as desired. Similarly, we have
for
0
< x < a,
11.2.
227
UNIQUENESS CRITERIA
We next consider the function defined by
<
<
a, 0 < y 6. We shall show that F ( x , y ) is continuous for 0 < x a, 0 < y b. Hence, it on R, . Clearly, it is continuous for 0 < x remains only to verify that lim
(X.Y)-ttP.U)
<
F ( x , y ) = 0,
<
O<x
and that Since a , p, y are continuous on R, functions
<
-<
, it suffices to prove that the three
where 0 < x a, 0 < y b, approach zero as (x,y ) 0 x a, or as (x,y ) (0, p), 0 7 6. We have, by mean value theorem,
< <
< <
-
(X,0), with
Similarly, it results that
where 0
< f < x,0 < 7 < y . Proceeding
to the equalities
in a similar way, we are lead
and
But the function uzU- vZyis continuous on R, and vanishes when x
=
0
228
CHAPTER
11
and y = 0. Thus, three functions given in (11.2.5) tend to zero as asserted, and this implies the continuity of the function F ( x , y ) as desired. Suppose now, contrary to what we want to prove, that u(x,y ) - v(x, y ) is not identically zero on R, . Then, F ( x , y ) must have a positive 6, and, maximum at a point (xo , yo) such that 0 < xo a, 0 < y o furthermore, F ( x , y) < F(xo ,yo), whenever 0 x y < xo yo with 0 x a and 0 < y 6. At this stage, we get a contradiction by applying the mean value theorem twice to the 01 term of F(xo ,y o ) ,once to each of the other two terms of F(xo , yo),and then using the Nagumo condition that is satisfied by f. We obtain, for example,
< <
where 0
< < +
<
<
+
< 6 < xo , 0 < r ] < y o . Similarly,
and
Putting together the last three inequalities and using the definition of F , we obtain F(x0 ,yo)
< 0.4
,Yo)F(5,?1)
+
+
+
P(X0
,Yo)F(.o
>
9*)
+
+
Y(X0
?
Yo)F(t** Yo),
where 6 + 7, xo q*, [* yo are less than xo y o . This contradicts the choice of (xo ,yo),since, by definition of (xo ,yo),we must have F(E9 7) F(X" 7 ?*)
F(5*, Yo)
I
< Q.0
1
Yo),
11.3.
UPPER BOUNDS AND ERROR ESTIMATES
while, at the same time, and 4 x 0 Yo) "0 Yo) T h e proof is complete. 7
+
9
+
Ax0
229
yo) 3 0, P(xo,y o ) 3 0, Y ( ~ oY,O )3 0, Yo) = 1. 3
1 1.3. Upper bounds and error estimates Let us first prove some results that give a priori estimates of the solutions cf hyperbolic equation ( 1 1.2.1).
THEOREM 11.3.1. Assume that (i) f E CIRo x R3, R], f ( x , y , u, p , q) is monotonic nondecreasing in u, P, 9, and (I 1.3.1) If(., y , u, P , dI d i ., Y,I u I P I, I ?! 0%
e
I7
where g E CIRo x R+3,R,]; (ii) z E CIRo ,R,], x(x, y ) possesses continuous partial derivatives 9 %z/ > 0,and xx ,
Y) > A x , Y,
ZZY(X7
4x7
Y), Zr(% Y)* XY(%
A);
(iii) u(x,y ) is any solution of (1 1.2.1) existing on R, such that
I UO I < Z(O,O), for 0
<x
0
I +)I
< %(X, O),
I .'(Y)I
< ZY(O,Y),
< b.
Then we have, on R,
,
Proof. T h e proof proceeds as in the first part of Theorem 11.1.1. It is enough to prove that, on R, ,
I %(.,Y)I
< .x(%Y)
and
I UY(X*Y)I < ZY(X,Y).
Assume that this is false. Let to be the greatest lower bound of numbers t >x y such that the inequalities are satisfied for x y < to . Then, there exists a point (x, ,y o ) with xo yo = to for which one of the two inequalities is not true. Suppose that
+
+
+
230
CHAPTER
11
I t then follows that
I .(.tY)I 1 U Y ( X , Y)l for x
< .(.,Y), < zv(x, Y )
< zz(x,Y)7
I &>Y)I
(1I .3.3)
+ y < t, . I n view of the condition (1 1.3.1), we have
I u,(xo
,Yo); <
+ J"
I U'(xo)i
4-J"
< I +")I :-: zz(xo
, o ) -i-
yo
I f(.xo , t , u(x0 , t ) , uz(xo , t ) , u,(xo
YO
g(xo
I
yo
> f,
I .(.o,
t)l, I u,(xo,
t ) ) dt ~
,I u,(xo,
t)l
t)l) dt
.
g(xo 9 t , z(xo 9 t ) , zz(xo , t ) ,zy(xo 7 t ) >dt
because of condition (iii), the relations (1 1.3.3), and the monotonic character of g in u, p , q. Assumption (ii) now leads to the inequality
I4 x 0
roll
< zdxn
9
yo),
which contradicts ( I 1.3.2). On the other hand, if we suppose that 1 u?,(x,, y,)l = x z I ( x oy, o ) , proceeding similarly, we get the inequality I u,(xO ,yo)/< x,(xO ,y o ) , which is again a contradiction. Hence, the stated estimates are true, and the theorem is proved. Using a similar argument with obvious modifications, we can prove the next theorem, which offers an estimate for the difference of any two solutions of ( I I .2.1).
THEOREM 1I .3.2. Assume that hypotheses (i) and (ii) of Theorem 11.3.1
hold except that the condition (1 1.3.1) is replaced by If(X>
Y9 u1 Pl Sl) - A x , Y ?uz Pz sz>l <,.(g Y , I u1 - uz I> I P, - P z I, I Q1 - Qz 1). 7
7
9
9
Let u(x,y ) and ~ ( xy,) be any two solutions of (1 1.2.1) satisfying
such that
u(0,O) = uo ,
n(0,O) = % ,
u(.x,O) = o z ( x ) ,
U(0,Y)
I uo - no I < 4 0 , O), I Tl'(Y) ~
=
.l(Y),
4%0) = 4 4 , 40,Y)
=
I 01'(x) - 02'(x)I < z&,
.z'(r)l < Z,(O,Y)
Tz(Y),
O),
11.3.
for 0
23 1
UPPER BOUNDS AND ERROR ESTIMATES
< x < a, 0 < y < b. Under these assumptions, we have, on R, ,
Next we shall consider the hyperbolic differential inequality
where f E C[R, x R3,R] and S
E
C[R, R,].
DEFINITION 11.3.1. Any function u E C[R, , R],possessing continuous partial derivatives u, , uv , uxv in R, and satisfying (1 1.3.4) in R, , is said to be a 6-approximate solution of (11.2.1), if it verifies the boundary conditions specified in Sect. 11.2. T h e following theorem estimates the error between a solution and a 6-approximate solution (1 1.2.1).
THEOREM 11.3.3. Assume that (i) u ( x , y ) is a solution and v ( x , y ) is a S-approximate solution of (1 1.2.1); (ii) z E C [ R , ,R,], z(x, y ) possesses continuous partial derivatives x, , zv, zxv > 0 in R, , and
(iii) f E C[R, x R3, R], S nondecreasing in u, p , q, and
If(.?Y,
E
C[R, , R,], f ( x , y , u, p , q) is monotonic
u1 , P l y 41)
B g(x, y , I u1
-f(%Y, - u2
u2 ,P2 9
I, I Pl
- P2
9211
I, I 41 - 42 I>,
where g E C[R, x R+3,R+]. Then, the inequalities
I VO - uo I < 4 0 , O),
I v,(O,y) - +)I
< x,(O,y)
I %(X, 0 ) - 441 < X,(% O), for
0
< x < a,
0
< y < b,
232
CHAPTER
11
imply
I %(X, Y )
on R, .
-
u&, Y)I
< %J(x,Y ) ,
Proof. We proceed as in Theorem 11.3.1 to show that there exists a to = xo yo such that either
+
or and, in either case, we get
for x
+y
< t, . T h e n we
have, successively,
contradicting the fact that I v,(xo , y o )- uz(xo , y o ) \= xz(xo , y o ) . A similar contradiction occurs in the other case, too. Consequently, the desired result follows.
11.4.
NOTES
233
11.4. Notes
See Walter [8] for the hyperbolic differential inequalities given in Sect. 11.1. Theorem 11.2.1 is taken from Shanahan [1], whereas Theorem 11.2.2 is due to Diaz and Walter [8]. See aiso Lakshmikantham [2]. For the estimates of the type given in Sect. 11.3, see Walter [8]. For the global existence theorems using Tychonoff’s fixed point theorem, see Aziz and Maloney [l]. See also Bielecki [l]. Concerning periodic solutions of hyperbolic differential equations, refer to Aziz [2], Cesari [l, 21, and Hale [ll]. For the application of contraction mapping theorem in generalized metric spaces to existence and uniqueness theorems of a particular type of hyperbolic differential equation, see Wong [ l , 21. For related results, see Alexiewicz and Orlicz [I], Aziz [l], Aziz and Diaz [l], Chu [l], Ciliberto [I, 3-51, Conlan [ l , 21, Conti [l], Diaz [l, 21, Guglielmino [ 1 4 , Kisynski [ l , 31, Lakshmikantham [2], Palczewski [l], Palczewski and Pawelski [l], Pelczar [I, 2, 4, 51, Phillips [l], Protter [l], Santoro [l], Szarski [8], Szmydt [l-51, Volkov [l], and Walter [l-6, 81.
This page intentionally left blank
DIFFERENTIAL EQUATIONS IN ABSTRACT SPACES
This page intentionally left blank
Chapter 12
12.0. Introduction A study of differential equations in abstract spaces is the content of this chapter. A variety of results on existence, uniqueness, continuous dependence, and method of averaging are given. A major part of the chapter is devoted to nonlinear evolution equation. A number of results are obtained regarding existence, estimates on approximate solutions, Chapligin's method, asymptotic behavior, and stability and boundedness of solutions. 12.1. Existence Let E be a Banach space, and, for any u E E, let 1 u I denote the norm of u. Suppose that f ( t , u ) is a mapping from [to , to a] x E to E. We consider the differential equation
+
(12.1.1)
u' = f(t, Id).
DEFINITION12.1.1. Any function u(t) is said to be a solution of (12.1.1) if the following conditions are verified:
+
(9 E C"t, 7 t, 4 7 El; (ii) u(to) = uo ; (iii) u(t) is strongly differentiable in t for to < t Eq. (12.1.1) for t, < t \c to a. We shall now prove a local existence theorem.
+
THEOREM 12.1.1. Assume that (i) f E C [ [ t o ,to a] x S, , El, where
+
S , = [ U E E :I
u-u~I
237
< to + a and satisfies
238
CHAPTER
12
and f ( t , u ) maps bounded sets into bounded sets so that, for any b there exists a positive number M such that
> 0,
+
(ii) V EC [ [ t , , t, u ] x S, x sb , R,], V(t,u, v) > 0 if u # v, V ( t ,u, a) = 0 if u = a,a(t, u, , TI,) -+ 0 implies u, - 71, -+ 0 for each t ; (iii) V(t,u, 71) has continuous (bounded additive FrCchet) derivatives
and
(iv) for any positive number M , aV/at, (aV/au)x, (aV/av)x are continuous in ( u , a) uniformly for ( t , U , v) E [ t o ,t o u] x Sh x Sb and I x j M.
+
<
Then, Eq. (12.1.1) possesses a solution on [ t o ,t, number satisfying CUM 6 .
<
Proof.
Let d be a subdivision of [ t o ,to to
For tk-l
< t , < ... < t ,
+
=
CU],
to
+
CU], cu
> 0 being the
namely,
+ a.
< t < t,<,we define
Let d and A , be two subdivisions of [ t o ,to
+ a ] , and
consider ~ $ ~ ( t )
12.1.
239
EXISTENCE
and $Ado(t)as before. If t is not a subdivision point of either A or A , , < t , and t9-1 < t < tjo, say, then
tkPl < t
d
dt [ V t ,+ d ( t ) , +d,(t)>l
=
where
av aP (aL at
-I
+
aP av aP aP at + [(a , a,> +a,]
av -- a q t , +A(t), +d,(t)) _ at
at
aP a v t , +d(tk-l), _ at
9
+A,(tjO-l)>
at
,
with similar expressions for aV/au, aV/av. By assumption (iv), for any E > 0, there exists a 6 > 0 such that, if we take I A I = max(t, - tkPl) < 6 and I A , I = max(tjO- t9-1) < 6, then we have
and Thus, we deduce that d V ( t , +A(t), dt
+d,(t))
< E.
Hence, it follows that
and this proves that there exists a d ( t ) such that
240
CHAPTER
12
Let now t E [to , to + a] be fixed. Then, for each subdivision A , there t t, . Furthermore, we obtain exists a k such that tk-l
< <
which shows that 44( tk-1)
Consequently, =1
f(t,d’d(tk--l)) and, by dominated convergence,
The proof is therefore complete. In Theorem 12.1.1, we have assumed thatf(t, u) is continuous. It can be shown that the conclusion of Theorem 12.1.1 is true even under weaker hypotheses on f. For this purpose, we need the following
+
DEFINITION 12.1.2. Let f ( t , u ) be a mapping from [to! to a] x E to E. We say that f ( t , u) is demicontinuous if it is continuous from
[ t o ,to
+ a] x
E with the strong topology to E with the weak topology.
THEOREM 12.1.2. Suppose that E is a reflexive Banach space. Then, the conclusion of Theorem 12.1.1 remains valid if we replace the continuity of f by demicontinuity, other assumptions being the same. (In this case, the differentiation is in the sense of weak topology.) Proof. Notice that aV/au and aV/lav are bounded additive functionals. Therefore, by an argument similar to that used in the proof of Theorem 12.1.1 , we can prove that fd(t)
so that we have
where
-
-
f(4+(%
+ ' ( t ) = f ( t 9 C(t)>,
means weak convergence.
REMARK12.1.1. Suppose now that E is a Hilbert space and that f ( t , u)satisfies the monotonicity condition, that is, Re(f(t,
u ) -f(t,
v), u - v)
<M I u
-
1)'
( 12.1.2)
12.2. NONLOCAL
24 1
EXISTENCE
where we denote the scalar product by ( x , y ) and the norm by1 x I = (x, 3 0. Then V(t, u, v) = e-2Mtl u - v l2 satisfies all the hypotheses of Theorem 12.1.1, and, consequently, the conclusion of Theorem 12.1.1 is true. We shall now give an example to show that, even whenf(t, u ) does not satisfy the monotonicity condition, there does exist a V(t, u, v) satisfying the hypotheses of Theorem 12.1.1. Consider the example du - = f ( t , u) =
dt
u
< 0,
assuming that E = R. Clearly (12.1.2) is not fulfilled. However, there exists a V ( t ,u, v) defined as follows:
I(";
V(t,u, n) =
I
-
dV
-
log(1
+ dU) + log(1 + dVy,
(G- log(1 + 4;)- &v)2, (4. - dV + log(1 + dV))Z,
\a<.
- v)2,
u
< 0,
u u
3 0, < 0,
2,
u
3 0, n >, 0,
< 0,
n 3 0,
v < 0.
12.2. Nonlocal existence We shall use the functional method of Leray-Schauder to prove the nonlocal existence of solutions of the differential equation (12.1.1).
DEFINITION 12.2.1. Let A and B be completely continuous operators defined for u E S , , where S , = [u E E : I u I p ] , with values in E and Au E E, Bu E E for u E S, . Then we say that A and B are homotopic if there exists an operator T(u,A), which is completely continuous on E x [0, 11 such that T(u,0) = Au and T(u, 1) = Bu for u E S, and T(u, A) # u for I u I = p. We need the following lemma of Leray-Schauder.
<
LEMMA 12.2.1. Let the completely continuous operator A be homotopic to the operator identically equal to zero. Then, there exists at least one solution u of the equation Au = u such that I u I < p . Let V(u) be a functional that possesses FrCchet differential L. Then, we have V(u
where w(u, h)/l h I + O
+ h)
~
V(u)= L(u, 4
as I h 1-0.
+ 4%4,
242
12
CHAPTER
As an application of Lemma 12.2.1, we prove the following existence result.
THEOREM 12.2.1. Assume that (i) there exist functionals V,(u) possessing FrCchet differentials Li such that (i = 1 , 2,..., 4, (12.2.1) ~ , ( u , f ( t ,4)< g,(4 V&), ..., V&)>
+
wheref(t, u ) is completely continuous for t E [to , to a], u E E ; (ii) g E C[to, to u] x R+n,Rn], g ( t , y ) is quasi-monotone nondecreasing in y for each t , and the maximal solution r(t, to ,yo ) of
+
+
Y’
= g(t,y),
Y@o)= Yo
exists on [to , to a ] ; (iii) @(u) = maxi V,(u) and @(u)-+ co as I u
I -+ co.
Then, for every uo E E, there exists at least one solution u(t) of the differential equation (12.1 .l) defined on [to , to u] such that u(to)= uo .
+
Proof. Let us consider the space B of all continuous functions u(t) with values lying in E, continuous on the interval [ t o ,to u]. Define
1u1
=
max,oGIGl,+a I u ( t ) / .Also, consider the operator
+
which maps B into itself. Denote this operator by T(u,A). Sincef(t, u ) is completely continuous, it follows that T(u,A) is completely continuous on B x [0, I]. Suppose now that u(t, A) is a solution of the equation ~ ( t= ) XU,
+X
f ( ~u(s)) , ds. t0
The last inequality implies that
12.3.
243
UNIQUENESS
choosing yo = V(Au,). Write K
max ri(t, t o , Vl(Au,), ..., Vn(Auo)).
=
i.d,t
Then, it follows that Vi(u(t,A))
< K,
t
+ a ] , 0 < X < 1. (12.2.2) < M for 0 < X < 1 and to < t < to + a.
E
[ t o , to
There is such an M that I u(t, A)l Indeed, if it were not so, then there would exist a sequence A, E [0, I] and a sequence t, E [ t o ,to u] such that 1 u(tn , A,)[ + co. Hence, @ ( u ( t nA,)) , -+a.This contradicts (12.2.2). Thus, 1 u(t, A)l M, 0 A 1 , t E [ t o , t, a]. We conclude that T(u,A) # u for 0 X 1 and 1 u 1 = M E, f > 0. It is easily seen that T(u, I) is homotopic to zero in the region I u 1 M E. By Lemma 12.2.1, we find that, in the space B , there exists at least one solution of the equation
+
<
+
< <
< <
+
< +
4 t ) = uo
+ fp,4s)) ds.
This completes the proof.
12.3, Uniqueness
+
Let E be a normed space, Q t ) , t E [ t o , to u] a subset of E, and D = [(t,u ) : t E [ t o ,to a ] , u E D(t)].We consider Eq. (12.1.1), where f ( t , u)is defined on D. We note that f(t, u)need not be continuous.
+
THEOREM 12.3.1. Assume that ___
(i) u,, E D(O),f(t, u)is defined on the set D ; (ii) V E C[Q,R,],where
9
= [(t,
u, 4 : ( t , ).
E D,
(4 n) E DI,
V ( t ,u, u ) > 0 if u f v, V ( t ,u, v) = 0 if u = v ; (iii) V ( t ,u, v) has continuous (bounded additive FrCchet) derivatives and
av
-at+
avm 4 au
~
avxt,a) aV
< 0.
244
CHAPTER
12
Then, there is at most one solution of (12.1.1). Furthermore, if V(t,u, u ) satisfies the condition V ( t , u, , z)J
implies
+0
u,
-
for each
vn + 0
t,
(12,3.1)
then the solution u(t)depends continuously on uo for each t. Proof. Suppose that there are two solutions u ( t ) , u ( t ) of (12.1.1) such that u(t,) = u(t,) = uo . Then, we derive that
<0
for
> t,.
t
Hence, it follows that
v(t,4 t h v(t>)< q t o
9
u(to), 4 t o ) ) .
(12.3.2)
Since V(t,, u(t,), u(t,)) = 0, we see that V ( t ,u(t),~ ( t )= ) 0, and, consequently, u ( t ) = u(t). This proves uniqueness. Since V ( t ,u, u)is continuous, 1 u(t,) - v(t,)l 40 implies
-
v(to u(tn), v(to)) 9
0.
This implies that V(t,u(t),u(t))-+ 0 by (12.3.2). If V ( t ,u, u) satisfies (12.3.1), we have
I u ( t ) - v(t)l when I u(to)- u(t,)I 40. T h e proof of the theorem is complete.
0
Consider the following simple parabolic equation, U t = %z
+ F(4 x,4,
(12.3.3)
+
in a region bounded by t = to , t = t, a, x = h,(t), and x = h,(t), where h,(t) < h2(t) for to < t to a. T h e initial and boundary conditions are as follows: u = g ( x ) on t = t o , u = hl(t), u = h2(t) on x == h,(t), x = h,(t), respectively, where g, h, , and h, are continuous and g(Xi(tn)) hi(to), gdhdto)) = hAto)- Suppose that
< +
Then, we can use Theorem 12.3.1 to prove uniqueness of solutions of (12.3.3). Let D(t) be a set of functions x that are continuous on
12.3.
245
UNIQUENESS
< <
A,(t) x A,(t), twice continuously differentiable on Al(t) < x < A2(t), and take values h,(t), h,(t) at x = A,(+ x = A,(t), respectively. Define
and f(4
4 =).(,,.
+qt,
X,
4.))
for u E D(t). Then, we have
< 0, so that the solution is unique by Theorem 12.3.1.
THEOREM 12.3.2. Let E be a Banach space and f ( t , u ) be defined and to a, u E El for which demicontinuous on a set D C [(t,u ) : to < t we assume that D(t) = [ u : ( t , u ) E D ] is closed in E for each t E (to , to u ] . Suppose further that
< +
+
If(4
41 < w t >
on
D,
where M ( t )is a summable function. Assume that there exists a real valued function V ( t ,u, v) satisfying conditions (ii), (iii), and (12.3.1) of Theorem 12.3.1. Let u = $(t) be continuous in D for which SUP,^^(^) V ( t ,$(t),u ) 0 as t -+ to . Assume also that, for every t, , to < t , < to a, (12.1.1) has a solution starting from (tl , $(tl)) and reaching the plane t = to a in D. Then, there is a unique solution u ( t ) of (12.1.1) with u(t,) = uo = $(to).
+
---f
+
Proof. Denote by u = +(t;t l ) a solution starting from (tl , $(tl)). Then, if we take to < t , < t, , we have
246
12
CHAPTER
so that
<
V ( t ,# ( t ; tl), # ( t ; t 2 ) )
=
-
,+(tz ; tl), d(t2 ; tz)> V(t2 ,4(& ; tl), +(Q) 0
q
2
as
t2
-
to .
Thus, 1 $(t; t l ) - +(t;ta)l + 0 if t , , t, -+ to . Let +(t) = limll+to$ ( t ; t l ) . Then, ( t ,$ ( t ) )E D and +(t)is a solution of (12.1.1), since, for arbitrary Z > t o , we have
4 ( c tl)
-
(6(t'; t l )
=
f-tcs, +(s; 5 ) )ds,
to
i
< tl < t ,
and hence, by letting t, -+ to , we get 4(t)
-
442)
=
Jt.f(.,
&)> ds-
This means that +(t)is a solution of (12.1.1) in D.Uniqueness of solutions is obvious, since, if +*(t)is such a solution, then
c-(4 4(t; tl), C * ( t ) )
< V(tl
-
= V(tl
as t, + t,, shows that
4(c tl) This completes the proof.
7
-
4*(tl)> 4*(tl)) 0
4(tl ; tl),
> +(tl)>
4*(9
Let ff be d I-lilbert space and f ( t , u ) be a continuous function on Q -- [ ( t ,zi): t,, ,t t, a , j u - uo 1 c] taking values in H . Suppose that
<
< +
(i)
f ( t , u ) + 0 as
( t , 21) -+ (to , zio) in Q;
(ii) Re( f ( t , 2 1 ) - f ( t , v),zi
- 71)
< (t
-
t,,)-l I u
-
v
on
a.
'Then, there is a unique solution $(t) of (12.1.1) such that +(t)+ u,, as t -* t,, . I n fact, take 1 ( t , ZL,
Then, for ( t , u ) , ( t , z7)
E
dl'
=
[ ( t - to)")-' 1 u
- z1 12.
Q,
~-1
Pt
21)
'
a l- f ( t ,Zl) avf(t v) o. -.+--L ~
?ZL
iiV
< rn if < cl.
Let $ ( t ) u0 . By condition ( I ) , we may assume that i f ( t , u)i c1 , c1 c, and ma, to t -< to -L o1 u1 to 1 u, 1 u - u,, 1 Set 1
<
~
A(t)
=
sup[/f(t, u ) ' : j u
<
1 < ?dt
- U,)
<
t")].
12.4.
Then, A ( t )
-+
247
CONTINUOUS DEPENDENCE
0 as t -+ to . Writing
we see that
Since there exists a solution starting from (tl , $ ( t l ) ) for each t , such that to < t, < to a , , which reaches the plane t = to a, , by Theorem 12.1.1, the desired uniqueness is a consequence of Theorem 12.3.2.
+
+
12.4. Continuous dependence and the method of averaging Let us consider the differential equation
where f is a function with values in a Banach space E, defined in a set J x H x A , where H is an open subset of E and A an arbitrary metric space. Throughout this section, we shall assume that the following conditions hold: (i) for each X E A , the mapping ( t , u ) +f ( t , u, A ) is continuous in
J x H;
(ii) for some A, E A , there exists a solution u,(t) of (12.4.1) which is defined on J and has its values in H ; (iii) there is a neighborhood I', of A, in A such that, for every X E T o , (12.4.1) admits a solution u ( t ) with u(0) = uo(0),which exists in some interval [0, T(h))C J. We recall that condition (i) alone does not guarantee the existence of a solution of (12.4.1) unless E is finite dimensional. Then, we can prove the following results using the arguments similar to the proofs of Lemma 3.20.1 and Theorem 3.20.1. We do not give the details.
248
CHAPTER
LEMMA 12.4.1.
12
Assume that
(i) V EC[J x E, R,], V(t,0) = 0, V(t,u ) is positive definite and satisfies a Lipschitz condition u for a constant M > 0; g(t, 0) = 0, and, for any step function v(t) on J (ii) g E C [J x R, , R], with values in H and for every t E J , u E H , D + U ( t ,u, A,)
=
lim sup h-l[ V(t h-O+
+ h,F(t, v(t),A,)
+ h { f (4 74th A,)
where
< A t , r.(4 4 t ) F ( t , v ( t ) ,A)
-
=
4
- f(4 u, A,)}) 1
u
~
q t , F(t, 4 t h A,)
-
~
u>1
7
40)
+ J k s , m,A) ds.
Then, given any compact interval [0, T,] C J and an E > 0, there is a 6 = 6 ( ~> ) 0 such that, for any step function v ( t ) in [0, .T,,], with v(0) = uO(0)and 1 v ( t ) u,,(t)l < 6 in [0, T,], there follows ~
1 It
If(S,
4%A,) - A s ,
uo(s), A")] ds
1<
€9
for every t E [0, T,], u,(t) being any solution of u' = f(4 u, A,),
defined for t E J. We assume that, for each t
E
J and u E H ,
lim / ' f ( s , u, A) ds
n-.n,
,
=
s:
f ( s , u, A,) ds.
(12.4.2)
I t then follows that, given any compact interval [0, T,] and any step function v(t) in [0, T,,] with values in f1, lim
&A. a
/ ,k s ,
'u(s),
A) ds
f(S, +J(.V),
A,) ds,
0
uniformly in [0, T,,]. Hence, if the assumptions of Lemma 12.4.1 hold, there exists, for every E > 0, a constant 6 = S ( E ) > 0 such that, whenever v(t) is a step function in [0, To] with v(0) = uo(0) and I v(t) u,,(t)I ,-: 6 in [0, To],there is a neighborhood F = r ( E ) C A of A, for which h E r implies ~
12.5.
249
EXISTENCE
t E [0, To].Thus, we have
which will be used in the main theorem that follows, as in Theorem 3.20.1.
THEOREM 12.4.1. Suppose that (i) V E C[J x E , R+], V(t,0 ) = 0, V ( t ,u ) is positive definite and satisfies a Lipschitz condition in u for a constant 111 > 0; (ii) g E C[J x R, , R], g(t, 0 ) = 0, and r ( t ) -=0 is the maximal solution of Y’ = Y) passing through (0, 0); (iii) for any step function v(t) on J , with values in H and for every t E J, u ~ H , y e A , D+U(t,u, 4
< g(t, V ( t ,v ( t ) 4); -
(iv) the relation (12.4.2) holds. Then, given any compact interval [0, To]C J and any E > 0, there exists a neighborhood T ( E )of ho such that, for every h E T(E),(12.4.1) admits a unique solution u ( t ) with u(0) = uo(0),which is defined on [0, To] and satisfies
I 141) - u,(t)l < E ,
t
E
[O, To].
12.5. Existence (continued) Hereafter, we shall be concerned with the nonlinear evolution equation u’
=
A ( t ) u +f(4 u),
(12.5.1)
where A(t) is a family of densely defined closed linear operators on a Banach space E andf(t, u ) is a function on [ t o ,to a] x E taking values in E. First of all, we shall summarize some of the known results for the linear equation u’ = A(t)u + F ( t ) , (12.5.2)
+
where F ( t ) is a function on [to, to is unbounded.
+ u] taking values in E. Usually A(t)
250
12
CHAPTER
Let us make the standing assumptions that there exists an evolution operator U ( t ,s) associated with A(t).This means that ( U ( t ,s)} is a family of bounded linear operators from E into E defined fort, s t to a, strongly continuous in the two variables jointly and satisfying the conditions
< < < +
U(S,s)
U ( t ,s) U(s,r ) = U ( t ,r ) ,
au(t,s) u = A ( t ) U ( t ,s)u, 3s
I,
for u in a subset of E , specified in each case.
at
___ au(t9 s, 1L =
=
U(t,s) A(s)u
A function u(t) defined on [to,to
DEFINITION 12.5.1.
+ a] is said to be
a strict solution of (12.5.2) with the initial value uo , if u(t) is strongly continuous on [ t o ,to $- a ] , u(to) = uo , strongly continuously differentiable, and satisfies (12.5.2) on ( t o ,to a).
+
IfF(t) is continuous, any solution of (12.5.2) is of the form
DEFINITION 12.5.2. A function u ( t )is said to be a mildsolution of (12.5.2) with the initial value u,, if u(t) is continuous on [ t o ,to a] and satisfies (1 2.5.3).
+
12.5.3. T h e family (A(t)} of operators is said to be DEFINITION uniformly parabolic if (i) the spectrum of A ( t )is in a sector S,, /(A
= ~
[ z : 1 arg(z A(W’
I
- T)\
< w < n/2],
< Mil A I,
and I(A(t))-’
I
$ s, ,
< M,
where u and M are independent o f t ; (ii) for some h = n-l where n is a positive integer, the domain of A(t)”is independent of t , that is, D[A(t)”]= D and
I A(t)hA(s)-hI < M , I A(t)hA(s)-h- I I < M I t for t, s E [to, to
+ a], 1
-
h
< I, M
- s jk,
being independent of t.
12.5.
25 1
EXISTENCE
DEFINITION 12.5.4. T h e family {A(t)}of operators is said to be hyperbolic if A(t), for each t E [ t o ,to a ] , is the infinitesimal generator of a contraction semigroup, D[A(t)]is independent of t , and A(t)A(to)-l, which is a bounded linear operator, is strongly continuously differentiable. It is known that, if the family {A(t)}of operators is uniformly parabolic, there exists a unique evolution operator with the following properties:
+
(i) U ( t ,s) E C D[A(t)]for s < t ; (ii) U(t,s)u is Holder continuous in t and s, for u E D = DIA(to)h]; (iii) the mild solution is the strict solution of (12.5.2), ifF(t) is Holder continuous on [ t o , to a ] , where uo is an arbitrary element of E.
+
If, on the other hand, the family {A(t)}is hyperbolic, there exists a unique evolution operator such that
u(4s) W ( s ) l c D[A(t)l,
+
and, if F ( t ) E D [ A ( t ) ]for every t E ( t o ,to a), the mild solution is the strict solution of (12.5.2). We shall now prove an existence theorem for mild solutions of the nonlinear evolution equation (12.5.1), which is parallel to Theorem 12.1.1.
THEOREM 12.5.1. Assume that
+
(i) f~ C [ [ t o to , a] x E , El and f ( t , u) maps bounded sets into bounded sets SO that, for any b > 0, there exists a positive number M such that If(4 41 <
+
<
for t E [ t o ,to a ] , u E s b = [u E E: I u - uo I 61; (ii) V E C[[t,, to a] x E , R,], V ( t ,u,v) > 0 if u # a, V ( t ,u,v) = 0 if u = v, V ( t ,u, , v,) + 0 implies u, - v, -+0 uniformly in t ; (iii) V(t,u, v) has continuous (bounded additive FrCchet) derivatives and
+
(iv) for any positive number M , aV/at, (aV/au)x,(aV/av)x are continuous in (u, v) uniformly ( t , u, v) E [ t o ,to a] x Sb x Sb and 1x1 < M ; (v) (avlau)A(t)u (av/av) A(t)v 0.
+
+
<
252
CHAPTER
12
Then, the evolution equation (12.5.1) has a mild solution on [ t o ,to where (Y is a positive number
+ a],
(a) when A ( t ) is ,parabolic, u,, E D[A(t,)] and f ( t , u ) is Holder continuous in t and u ; (b) when A ( t ) is hyperbolic, uo E D[A(t,)] and A ( t ) f ( t ,u(t)) is defined and strongly continuous whenever u(t)E D [ A ( t ) ] . Furthermore, in the latter case, the mild solution thus obtained is actually a strict solution if E is reflexive and if we assume, in addition, that A ( t ) f ( t ,u ) is bounded whenever u remains in a bounded set. Proof.
Choose a positive number cz so small that
+
where I U I = sup1 U ( t, s)l, s, t E [ t o ,to m ] . If 1 t - to I be a subdivision of [to , to 011, namely, t,, < t , < -.- < t, For tk-l t t , , we define
< <
+
< m, ==
let d to + a.
Then, we can write
where
Hence, $ d ( t )E S, . Suppose that A ( t ) is parabolic, w,, E D[A(t,)], and f ( t , w) is Holder continuous in t and u. Then, we have $d(tkpl) E DIA(tkpl)], and U(t, trcpl)$d(trc-l) is Holder continuous in t for tk-1 t tk . Hence, (bd(t)is differentiable in ( t k - l , t k ) , and, as a consequence,
< <
If, on the other hand, A(t) is hyperbolic, w,,E D[A(t,,)],f(t,u) E D [ A ( t ) ] whenever u € D [ A ( t ) ]and , A(t)f(t,U ( t ,tkpl)$d(tk--l))is strongly continuous, we can again deduce (12.5.4). Let d and A , be two subdivisions of [ t o ,to + a]. Then, if t is not a subdivision point of either d or A , ,
12.5.
we can take k and j such that tkPl < t obtain
'
where
L\
av at
aP at
-
< t , and
< t < ti‘ and
ilv J
ilv
_ -
253
EXISTENCE
av
at (t,$ A (
av
= -( t ,
at
U(
with similar expressions for Since I+A(t) - W t ,
6 M l U and, similarly, we see that {$d(t))is a strongly convergent family. Thus, there exists a $ ( t ) such that $ d ( t )- + + ( t ) .Furthermore, by assumption (ii) we find that this convergence is uniform. Hence, fd(4
and consequently
4(t) =
s)
-
u(t,4 f ( S , #(s)),
w , + fl, u(4s ) f ( s , 4(4)ds, to)%
which shows that + ( t ) is a mild solution of (12.5.1).
254
CHAPTER
12
When A(t) is hyperbolic, we can prove that +(t)thus obtained is a strict solution, under the assumption that I A ( t ) f ( t ,I). is bounded whenever u remains in a bounded set. T h i s means that there is a positive integer L such that
I A ( t ) f ( t ,U ( t , tJC-1)
&I(tlc--l))l
Then,
+ jt +
< L.
A ( f ) U( t, A(s)-’A(s)f(s, U(s, 4-11
4d(tk-l))
ds
‘k-1
f(t,
u(4 4-1)
4d(tk-l))
shows that
I sho’(t)l G (1
4 f o ) u oI +L(t
~
to))A
+M,
where A = sup/ A(t) U(t, s) A(s)-l 1. Thus, {A(t)+,(t)}is bounded. Since A(t) is closed, this implies that $(t)E D [ A ( t ) ]when E is reflexive. It then follows that $(t) is a strict solution of (12.5.1) because of the assumption that A ( t ) f ( t ,{ ~ ( tis ) )strongly continuous in t if $(t)E D[A(t)]. T h e proof is therefore complete.
12.6. Approximate solutions and uniqueness Suppose that ( A ( t ) }is a one-parameter family of closed linear operators defined for each t c J. We shall assume that the domain D [ A ( t ) ]= D of A ( t )is independent of t. Assume also that, for each t E J , the resolvent set of A ( t ) includes all sufficiently small positive real numbers 01 and that thc domain of the resolvent K ( T , A(t)) [I - c~A(t)]-lis dense in E. We observe that, since A ( t ) is closed, D [ R ( x ,A(t))]= E, when 01 is in the resolvent set. Let us consider the differential inequality 1 s’ - A(+
f(t,
)I.
3.q ( t , 1 s
I),
(12.6.1)
w h e r e f E c‘[J<, B, El, w 1E C [ J x R, , R,]. h F I X i T I O N 12.6.1. Let the function x(t) be strongly continuous and defined f o r t F J . Suppose further that x ( t ) has strong derivative x’(t)
12.6.
255
APPROXIMATE SOLUTIONS AND UNIQUENESS
and that x(t) E D. Assume that x(t) satisfies the inequality (12.6.1) for t E J - S , where S is at most a denumerable subset of J . Then we shall say that x ( t ) is a solution of (12.6.1). If wl(t, u) = E, x ( t ) is said to be an +approximate solution of the differential equation x' = A ( t ) x +f(t, x).
(12.6.2)
THEOREM 12.6.1. Let w2(t,u ) be a scalar function defined and continuous on J x R,. Suppose that r ( t ) is the maximal solution of the scalar differential equation u'
where
=
w(t, u),
4,4 =
u(tg) =
u)
,
(12.6.3)
u),
(12.6.4)
ug
+ %it,
existing to the right of t o . Assume that, for each t
E
J, x E E,
lim R(h, A(t))x = x,
(12.6.5)
+ hf(4 .)I
(12.6.6)
h-O+
and
I R(h, A m
< I x I + hwz(t, I x I),
for all sufficiently small h > 0 depending on t and x. Let solution of (12.6.1) such that 1 x(to)l u,, . Then
<
I x(t)l
< r(t),
t
2 t, *
+ h)
+ + w, + +
I x(to)l
+ +w ,
< uo .
+w ,
I x(t h)l - I R(h, 4 t ) )x(t) x(t))l I 4 t ) )x ( t ) x(t))l < I x(t h) - R(h, A ( t ) )x ( t ) - hf(4 x(t))l I R(h, A@))x ( t ) x(t))l.
=
be any (12.6.7)
Proof. Let x(t) be any solution of (12.6.1) such that Define m ( t ) = I x(t)l. For small h > 0, we have m(t
x(t)
(12.6.8)
Since, for each t E J , A(t) is closed and D[R(h,A(t))]i s dense in E, it follows that D[R(h,A(t))]= E and R(h, A(t))(l- hA(t))x = x for every x E D.Hence,
+
+
R(h,A ( t ) ) ~ (= t )~ ( t ) h A ( t ) x ( t ) h[R(h,A(t))A ( t ) ~ (t )A ( t ) x ( t ) ] . (12.6.9)
From (12.6.8) and (12.6.9), we obtain the inequality lim sup h-l[m(t h-O+
+ h) - m ( t ) ] < w(t, m(t)),
2 56
CHAPTER
12
because of the relations (12.6. I), (12.6.4), ( 1 2 . 6 3 , and (12.6.6). Now, an application of Theorem 1.4.1 implies the stated result. We can prove an analogous result for lower bounds.
THEOREM 12.6.2. Let w 2 ( t ,u ) be a scalar function defined and continuous on J x R , . Suppose that p(t) is the minimal solution of the scalar differential equation u' =
-w(t, u),
p(to) == p,
> 0,
existing on [to , 00). Assume that, for each t E J and x E E, lim R(h, A(t))x = x
h-O+
and
I R(h7 4 t N x + h ? ( t ,41 2 I x I - hwz(t, I x I) for all sufficiently small h > 0 depending on t and x. Let x(t) be any solution of (12.6.1) such that 1 x(to)\ 3 po . Then, for all t for which p ( t ) 2 0, we have IX(t>l
3P(9
Proof. Defining m(t) = I x(t)I as before, it is easy to obtain the inequality lim inf h-l[m(t h-O+
+ h)
-
m ( t ) ] 2 -w(t, m(t)).
This is enough to prove the stated result using an argument essentially similar to that of Theorem 1.4.1. For various choices of w1and w 2 ,Theorems 12.6.1 and 12.6.2 extend many known results in ordinary differential equations to abstract differential equations. Suppose that w 1 = E and that x ( t ) is an E-approximate solution of (12.6.2). Let w 2 = Kzi, K > 0. Then, Theorem 12.6.1 gives an estimate of the norm of +approximate solution, namely,
I x(t)l '2 1 x(t,)/ eK-)
+ (c/K)(eK(t-Q
--
l),
t
2 to,
whereas Theorem 12.6.2 yields a lower estimate,
Again, suppose that w1Y 0 and that x(t) is a solution of (12.6.2) existing on [t, , 00). Let w2= A(t)g(u), where g(u) > 0 for u > 0 and
12.6.
APPROXIMATE SOLUTIONS AND UNIQUENESS
257
X E C [ J ,R ] . Then, we obtain the following upper and lower bounds of the norm of a solution, namely,
Jz,
where G(u) = [g(s)]-lds, u,, 3 0. If we suppose that wl = v(t)u, v(t) >, 0 is continuous on J , we have a variant of Theorem 12.6.1 which offers a sharper estimate.
THEOREM 12.6.3. Let the assumptions of Theorem 12.6.1 hold except that the condition (12.6.6) is replaced by
+
< I x I (1 - ah) + hw2(t, I x 1
I R(h, A(t))x hf(t, .)I where
01
> 0. Then (12.6.7) is replaced by &to)
Proof,
e--u(t-to), (12.6.10)
I x(t)l
e r(t),
t
3 to.
Let R(t) be the maximal solution of R'
= -&
+ w(t, Rp(t-to)), - E ( t - f o ) ,
<
such that I x(t,)l R(to).Then, it is clear from (12.6.10) that Theorem 12.6.1 can be used to obtain the inequality Ix(t)l
< R(t),
t
2 to
1
But R(t) = r(t) & - l o ) , where r ( t ) is the maximal solution of (12.6.3) such that R(t,) = u o . Verification is just the method of variation of parameters. Hence the result follows. We next prove a uniqueness result analogous to Theorem 2.2.8.
THEOREM 12.6.4. Suppose that x(t) and y ( t ) are two solutions of the differential equation (12.6.2) with the initial condition x(0) = y(0) = 0. Let the condition lim I x(t> B(t)
=0
t+o+
be satisfied, where the function B(t) is positive and continuous on
258
CHAPTER
12
< t < CO, with B(0) = 0. Let g ( t , u ) 3 0 be continuous on J x R+ . Suppose that the only solution u(t) of
0
on 0
u'
< t < co such that
= g ( t , 24)
is the trivial solution. Assume that, for each t E J , lim R(h, A(t))x = x
h-O+
for every x E E and that
1 R(h, A ( W A ( t ) ) y+ h [ f ( 44 -f(t,Y)lI 5: I x - y I -1hg(4 I x - y I) ~
wt
for each t E (0, GO), each x, y E E, and for all sufficiently small h > 0, depending on t and x. Then, there exists at most one solution of (12.6.2) on J .
Proof. Suppose that there are two solutions x(t) and y ( t ) of (12.6.2) on J , with the initial condition x(0) = y(0) = 0. Let m ( t ) = 1 x ( t ) - y(t)i. Then, m(0) = 0. Now, using an argument similar to that of Theorem 12.6.1, we obtain D+m(t) < g(t, W ) . From now on, we follow the proof of Theorem 2.2.8 with appropriate changes to complete the proof. 12.7. Chaplygin's method
By the one-parameter contraction semigroup of operators, we mean a one-parameter family { T(t)),t 3 0, of bounded operators acting from E to E , such that
+
(i) T ( t , t z ) = T(t,) T(t,) for t , , t, >, 0; (ii) 1irnjb->,,T(h)x = x for x E E ; (iii) I T(t)I 1 for t E J.
<
T h e infinitesimal generator A of T ( t )is defined by Ax
=
lim
w-0
T(h) - z X h
12.7.
259
CHAPLYGIN’S METHOD
for every x, for which the limit exists. T h e limits mentioned previously, of course, are strong limits. T h e domain D[A] of A is dense in E, A is closed, and, for h > 0, I R(h, A ) ] 1. It is well known that, if A is 1, then there exists a closed and densely defined and if I R(h, A)l unique contraction semigroup {T(t)}such that A is its infinitesimal generator. For x E D [ A ] ,the function x(t) = T(t)x satisfies the equation
<
x’(t) =
x(2)
Ax(t),
<
= x,
t
2 0.
Notice that, for t , h 3 0,
I T(t
+ J.9” I < I T(h) T(t>xI e I T(h)l I T(t)x I
Hence, it follows that
e I T ( t ) x I. I x(t
+ 41 e I
x(t>l,
that is, the norm of the solution x(t) is a decreasing function. We observe that limb+, R(h, A ) x = x for every x E E, if A is closed, D ( A ) is dense in E, and limb,, sup/ R(h, A)I < 00. I n view of this fact and on the basis of Theorem 12.6.1, we can prove the following
THEOREM 12.7.1. Assume that
<
1 for (i) A is closed with dense domain such that I R(h, A)I h >O; ( i i ) g E C [ J x R + , R + l , f ~ c [ JE ,xE l , a n d l f ( t , x ) l < g ( t , I x l ) f o r t E J and x E E ; (iii) r(t) is the maximal solution of u’ = g(t, u),u(t,) = u,, existing on J. Then, if x(t) is any solution of x’ =
such that I x(t,)l
Ax + f ( t , x)
< u,, existing on J,
we have
I x(t)l
t
< +>
(12.7.1)
>to.
We shall now prove a result that generates the Newtonian method of approximations in a version given by Chaplygin.
THEOREM 12.7.2. Suppose that (i) A is an infinitesimal generator of contraction semigroup;
260
CHAPTER
12
(ii) f(t, x) is FrCchet differentiable in x t o j , ( t , x) and
I f Z ( 4 Y)-
fZ(4
41 e gl(4 I Y
-2
0,
where g, E C [ J x R, , R,] and gl(t, u ) is nondecreasing in u for each t s J; (iii) the sequence of functions {xn(t)} such that I xn(t)1 M, t E [O, a ] , n = 0, I , 2,... satisfies
<
4,l(t)
=
~,(t)),
+f(t,
(iv) suPl€[O,all
f X t 9
+f Z ( 4
&L+l(t)
Xvt(t))[Xn+dt) -
x,(O)
=
4t)l
xo,
t
E
(0, a ] ;
(12.7-2)
O>l -<
Then, x,(t) converges uniformly on [0, u ] . Furthermore, if x(t) is a solution of (12.7.1) such that x(0) = x,, , then there exists a well-defined sequence { ~ , ( t )such ) that 1x70) - 4t)l
where 1 x l ( t ) - x(t)l
< wl(t),
e w,(t),
(12.7.3)
t E [0, a ] , and
K being given by
If wrL(t)is equibounded, then x,(t) [O, .I. Proof.
Consider the sequence
which satisfies the equation
Using assumption (ii), we obtain U
+x ( t )
as n
+ 00
uniformly on
12.7.
26 1
CHAPLYGIN'S METHOD
On the strength of Theorem 12.7.1, we get
+
setting g ( t , u ) = Ku gl(t, I ~,-~(t)l)l znpl(t)I.Because of the mono2M, we have tonic character of g, and the fact I z,(t)l
<
< g k , 2M) < MO ,
I g,(s, I %-,(S)l)
t
E
[O, a],
where M,, is a suitable bound. Hence,
where = Moexp(Ka). On the other hand, I zl(t)l consequently, by (12.7.5),
< 2M,
and
I t then follows that {x,(t)} is uniformly convergent. If x(t) is a solution of (12.7.1), under the assumption, x ( t ) is uniquely determined. Also, notice that the inequality (12.7.3) is true for n = 1. Furthermore, w,(t) is well defined on [0, u] because
Suppose now that
I Xn-dt)
-
4t)l
<
(12.7.6)
Wn-lW.
Observe that
Then, assumption (ii) implies that
I PTl I
< K I %it)
-
4t)l
+ gl(4 I
Xn-l(t>
-
W) I X n - l ( t )
~
Since gl(t, u ) is nondecreasing, (12.7.6) and (12.7.7) yield
w.
262
CHAPTER
Theorem 12.7.1, with g ( t , u ) = Ku
12
+ gl(t, ~ , - ~ ( twnPl(t), )) shows that
= wn(t),
which proves (12.7.3). Suppose that Q n = 1, 2 ,..., is finite. Then,
w,d9
=
sup/ ~ , ( t ) l for t E [0, a],
exp[WNwn-1(4 ds,
where N = maxsg,(s, a). Hence, wn(t)+ 0 uniformly on [0, a]. T h e proof of the theorem is therefore complete. Another version of Theorem 12.7.2, which depends on Theorem 1.4.4, will be given next.
THEOREM 12.7.3. Suppose that (i) f(t, x) is FrCchet differentiable in x tofz(t, x) and
where g, E C [ ] x R, , R,] and gl(t, u ) is nondecreasing in u for each t g
J;
(ii) z(t) = T ( t )z o , where { T ( t ) ) is a contraction semigroup with generator A and x, E D[A]; (iii) G, F E C[[O, a ] , R,], and, for t E [0, a],
(iv) r ( t ) is the maximal solution of U' =
3G(t)u
+ 3g1(t, + F ( t ) , U)U
existing on [0,a ] ; (v) y ( t ) is a solution of the equation
such that y(0) = T(0)x,
= x,,
.
~ ( 0= ) 0,
12.7.
263
CHAPLYGIN'S METHOD
Then, the inequality
I 4) - @)I
=
I x ( t ) - W)%I < r ( t ) ,
t fz[O, a ] ,
implies
I Y ( t ) - 4t)l Proof.
< r(t),
t E [O, a].
We have, in view of the assumptions,
By Theorem 12.7.1, it follows that
I y ( t ) - z(t)l
<Wt),
(12.7.8)
t E [O, a ] ,
where R(t) is the maximal solution of u' = 2gi(t, ~ ( t )r )( t )
+ 2G(t)r(t> +F ( t ) + [gi(t, r ( t ) ) + G(t)]u,
existing on [0, a ] . Setting g ( 4 u, 4 = 2g,(t, v)v
+ 2G(t)v + F ( t ) + [g,(t, v) + G(t)lu,
we see that all the assumptions of Theorem 1.4.4 are satisfied, and, as a result, R(t) = r ( t ) on [0, a]. T h e assertion of the theorem then follows from (12.7.8).
REMARK12.7.1. I t follows from the preceding theorem that, for a sequence of solutions xn(t) of xA+,(t) = Axn+1(t)
+f&
xn-l(t))[xn+l(t)
-
4 0 )=%EaAI,
the estimate
I x,(t)
-
holds, provided z(t) = xl(t).
< r(t),
t
E
[O,aI,
+fk x n w ,
264
CHAPTER
12
12.8. Asymptotic behavior
We shall now suppose that the norm in E is differentiable in the sense of Gateaux, namely, lim 1 A
1
+
4
h
-
1
'
=
(rx,h),
where r x = grad I x 1 and (1, x) is the value of the linear functional 1 E B*,the conjugate space of E, at an element of E. It is easy to check that acts from E into E* and that
r
DEFINITION 12.8.1.
Consider a function y ( t ) for which the estimate (%
holds for all t
E
J and x E D
494 < A t ) I x I =
D [ A ( t ) ] .Introduce the notation
Q~ = lim sup
t-1
t-m
and
Q
=
(12.8.1)
f
y(s) ds,
0
inf Q, ,
where the inf is taken over all functions y ( t ) . T h e number Q is called the central characteristic exponent.
THEOREM 12.8.1. Assume that (i) Q is the central characteristic exponent; (ii) f ( t , x) allows the estimate ( T x , f ( t ,x))
< 8 I x 1,
Then, given an E > 0, there exists a 6 of (12.6.2) admits an estimate lx(t)l
> 0.
(12.8.2)
> 0 such that
< I 4 ) l c exp[(Q + 2EY1,
where the constant C depends on
Proof.
6
t
any solution x(t)
2 0,
(12.8.3)
E.
Let m ( t ) = I x ( t ) i . Then, using (12.8.1) and (12.8.2), we have
12.8.
265
ASYMPTOTIC BEHAVIOR
so that
By Theorem 1.4.1, we get
From the definition of SZ, given E > 0 there exists a function y ( t ) such that (12.8.1) is satisfied and, at the same time, Qy < Q
I n other words,
+
€.
where C i s a constant depending on E . 'Then, taking 6 < E and considering the last inequality, we obtain from (12.8.4) the desired inequality (12.8.3), and the theorem is proved. As an application of Theorem 12.8.1, we prove the following theorem, which gives sufficient conditions for the asymptotic stability of the null solution of (12.6.2).
THEOREM 12.8.2. Suppose that (i) the central characteristic exponent 52 is negative; (ii) the functionf(t, x) satisfies
4) < P I x
(&f(t,
01
I1+Or,
> 0.
(12.8.5)
Then, the trivial solution of (12.6.2) is asymptotically stable. Proof. Choose a X function
> 0 such that 9, = 52 + X < 0, and consider the x ( t ) = exp( -At) y ( t ) -
Then where
Y"t>
=
[4t)
+ A11 Y ( t ) + A t , Y ( t > ) ,
(12.8.6)
g(t, Y ) = e x p ( A t ) f ( t , e.p(--ht)y).
It then follows from the properties of Y = Y(t),
r and
(123.5) that, setting
(TY, g(4 Y ) ) = eAt(rY,f(4 = e A t ( T ( e c n t y ) , f ( ecAty)) t,
< eAt/3 1 ecAty
J1+a,
266
CHAPTER
12
(12.8.7)
+
Hence, the central characteristic exponent of the operator A(t) h l is equal to Q, = 9 A, and, therefore, choosing an E > 0 such that Q, 2~ < 0, we can find a function y ( t ) that satisfies (12.8,l) and Q < 9, E < 0, because of the definition of 9. Let 6 > 0 be such that 6 <: E . Take the initial time to > 0 so large that, for t to and small I y I, we get, by (12.8.7),
+
+
+
(T Y ,g(4 Y ) )
e 8 I Y I.
Thus, the operator g ( t , y ) verifies the hypotheses of Theorem 12.8.1, and hence
+
I Y(t)l < I Y(t0)l c exp[(Q,
+ 2+19
t
2 to .
Since Q, 2~ < 0, the null solution of (12.8.6) is asymptotically stable, and, as a result, the trivial solution of (12.6.2) is also asymptotically stable. T h e theorem is proved. Another set of conditions for the asymptotic stability is given by the following
THEOREM 12.8.3. Assume that (i) g E C[J x R, , R ] , and the solutions u(t) of the scalar differential equation u’
=
g(t, u),
u(to) = uo 3 0,
(12.8.8)
are bounded on [ t o ,a]; (ii) for each t E J , x E E, lim R(h, A(t))x = x
and
h-Of
I ~ ( h~ ,( t ) ) + x hf(t, x ) ~,< 1 x j ( I
where
O(
~
ah)
+ hg(t, 1 x I ea(t-tO))e--a(*--fJ,
> 0, for all sufficiently small h > 0 depending on
Then, the trivial solution of (12.6.2) is asymptotically stable.
t and x.
12.9.
Proof.
LYAPUNOV FUNCTION AND COMPARISON THEOREMS
267
Following the proof of Theorem 12.6.3, we obtain
I x(t)l
< r ( t ) exp[--or(t
-
to)],
(12.8.9)
t >, t o ,
where r ( t ) is the maximal solution of (12.8.8) and x(t) is any solution of (12.6.2). By assumption, r(t) is bounded on [ t o , 001. Hence, the asymptotic stability of the trivial solution of (12.6.2) is immediate from the estimate (12.8.9). T h e proof is complete.
12.9. Lyapunov function and comparison theorems We shall continue to consider the differential equation (12.6.2) under the same assumptions on the family of operators {A(t)}as in Sect. 12.6. Let us prove the following comparison theorems.
THEOREM 12.9.1. Assume that (i) V E C [ J x E, R,] and
I q t , .I>
-
q t ,4 1 < c ( t ) I 2 1 --
x2
(12.9.1)
I>
for t E J , x1 , x2 E E, c ( t ) 3 0 being a continuous function on J ; (ii) g E C[J x R, , R], r ( t ) is the maximal solution of the scalar differential equation u‘
= g(t, u),
existing on J , and, for t L)+l’(t, x)
=
u(tn) = zco
t J,
>, 0,
to
(12.9.2)
0,
x E E,
lim sup /z-l[V(t + h, R(h, A ( t ) ) x + /zf(f, x)) h-O+
s g ( t , r -(t,~y));
-
l’(t, x)]
(12.9.3)
(iii) for each t E J , lim,,,, R(h, A(t))x = x, x E E, and x ( t ) is any solution of (12.6.2) existing on [to, co) such that V ( t , , .%,(to))< ug .
Under these assumptions, we have V(t, x ( t ) )
< r(t),
t
2 to.
(12.9.4)
satisfying Proof. Let x(t) be any solution of (12.6.2) existing on [to, a), V ( t o ,x(to)) u,,. Consider the function
<
m(t) = V(t,. Y ( t ) ) ,
268
CHAPTER
so that nz(t,) m(t
+
--
12
< uo . Furthermore, for small h > 0,
m(t)
< c(t)[l
x(t
+ h ) - R(h, 4 t ) )x ( t )
-
hf(t, x(t))l]
+ v(t + h, R(h, 4 t ) )4 t ) + hf(4 x ( t ) ) )
~
because of (12.9.1). Since, for every x E D [ A (t)] ,
V(t,x ( t ) ) , ( 12.9.5)
R(h, A(t))[l- h A ( t ) ] x= x ,
it follows that
which, together with (12.93, implies that m(f
+4
-
m(t)
< c(t)[l x ( t + 4 - x ( t ) - h [ A ( t )x ( t ) +f(4 x(t))lll
+
h[l R(h, 4 t ) ) 4f) x(t)
-
A(t)4t)Il
+ L'(t + h, R(h, A ( t ) ) + hf(4 x ( t ) ) ) x(t)
-
V(t,x ( t ) ) .
Using the relations (12.6.2), (12.9.3), and assumption (iii), we obtain the inequality D+m(t)
< g(t, 40).
An application of Theorem 1.4.1 now yields the stated inequality (1 2.9.4), and the proof is complete.
THEOREM 12.9.2. Let the assumptions of Theorem 12.9.1 be satisfied except that the condition (12.9.3) be replaced by p ( t ) ntq44
+ "(t, ).
D+p(t)
< g(t, V ( t ,x ) p ( t ) ) ,
(12.9.6)
where p ( t ) > 0 is continuous on J . Then, whenever
~ ( t dV t n 1 4 t n ) ) G uo the inequality (12.9.4) takes the form p ( t ) qt,X ( t ) )
P ~ o o f . DefineL(t, x)
< r(t),
t
2 to *
p ( t ) V ( t ,x). Then, using (12.9.Q we have
1
12.10.
where
E +
269
STABILITY AND BOUNDEDNESS
0 as h + 0; a rearrangement of the right-hand side gives
It then follows that D+L(t,x)
=
lim sup h-l[L(t h a +
+ h, R(h, A(t))x + hf(t, x))
-
L(t, x)]
which implies that Theorem 12.9.2 can be reduced to Theorem 12.9.1 with L(t, x) in place of V ( t ,x). Hence we have the proof.
12.10. Stability and boundedness Let M be a nonempty subset of E containing {0}, and let d(x,M ) denote the distance between an element x E E and the set M . Denote the sets [x: d(x, M )< q] and [x: d(x, M ) 771 by S ( M , 7)and s ( M , q), respectively. Suppose that x ( t ) is any solution of (12.6.2) existing in the future. Then, we may formulate the various definitions of stability and boundedness with respect to the set M and the differential system (12.6.2) corresponding to the definitions (S,) to (Slo)and (B,) to (Bl,,) given in Chapter 3. As an example, (S,) would run as follows.
<
DEFINITION 12.10.1. T h e set M , with respect to the system (12.6.2), is said to be (S,) equistable if, for any E > 0, to E J , there exists a S = S ( t o ,E ) that is continuous in to for each E , such that x(t)
c S ( M , 4,
t
>, t o >
provided that x(to)E s ( M , 6). T h e following theorem gives sufficient conditions for stability.
THEOREM 12.10.1. Assume that (i) g E C [ J x R,
, R] and g(t, 0) = 0 ;
270
12
CHAPTER
(ii) V E C [ J x S ( M , p), I?+], V ( t ,x) is locally Lipschitzian in x, and, for ( t ,x) E J x S ( M , p),
)< V t , 4 d 4, 4 x 9 MI),
b(d(.x, W
where b E 2,a E C [ J x [0, p), R+],a E X for each t E J ; (iii) for ( t , x) E J x S ( M , p), D'Vt,
(iv) limh+O+ R(h, A(t))x
=
4 d g(4 vt,4);
x for t E J and x E E.
Then, the equistability of the. null solution of (10.2.1) implies the equistability of the set M with respect to the system (12.6.2).
P Y O O ~Let . 0 < E < p, to E J be given. Assume that the trivial solution of (10.2.1) is equistable. Then, given b ( ~ > ) 0, to E J , there exists a 6 = S ( t o , E ) that is continuous in to for each E such that u(t, t o , uo)
< b(4,
t
2 to,
(12.10.1)
<
provided uo 6, , where u(t, t o , uo) is any solution of (10.2.1). Choose uo = V ( t o ,x(to)). Because of the hypothesis on a(t, u), there exists a 6, = S,(to , E ) satisfying the inequalities +(to),
M ) < 81
and
4 t 0 ,
d(x(to), M ) ) d 8
at the same time. We claim that, if x(to)E s ( M , Sl), x(t) C S ( M , E ) , t >, to . Suppose that this is not true. Then, there would exist a solution x(t) with .(to) E s ( M , S,) and a t, > to such that and
d(x(tl), M ) = E
so that b(6)
d(x(t),
<W
l
M ) < E,
t
E
, "W).
[to , tl],
(12.10.2)
Since this implies that x(t) E S ( M , p), t E [to , tl], the choice uo = V ( t o x(to)) , and condition (iii) yield, by Theorem 12.9.1, the estimate V ( t ,4 9
< r(t, to , uo),
t
E
[ t o , tll,
(12.10.3)
where r(t, t o , uo) is the maximal solution of (10.2.1). I t is easy to see that the relations (12.10.1), (12.10.2), and (12.10.3) lead us to the following contradiction: b(E)
d
qt1,
X(tl)>
d
9
t o , .o>
< b(4.
Hence, the stated result is true, proving the theorem.
12.10.
STABILI'PZ AND BOUNDEDNESS
27 1
THEOREM 12.10.2. Let the assumptions of Theorem 12.10.1 hold except that the function a(t, u ) in condition ii is independent of t, that is, a(t, u ) = a(u), where a E 37. Then, one of the stability conditions of the trivial solution of (10.2.1) implies the corresponding one of the stability conditions of the set M with respect to the system (12.6.2). On the basis of the proof of Theorem 12.10.1 and that of the parallel theorems in Chapter 3, the proof of Theorem 12.10.2 may be constructed easily. We therefore omit its proof. As an application of Theorem 12.9.2, we shall give a result that offers sufficient conditions for equi-asymptotic stability.
THEOREM 12.10.3. Let the hypotheses of Theorem 12.10.1 hold except that assumption (iii) is replaced by p ( t ) D + W ,).
+ v(4).
D+P(t) d g(t, V
t , 4P ( t ) ) ,
where p ( t ) > 0 is continuous on J and p ( t ) -+00 as t + CO. Then, the equistability of the trivial solution of (10.2.1) assures the equi-asymptotic stability of the set M with respect to the system (12.6.2). Proof. The proof of this theorem is analogous to the proof of Theorem 3.4.7, if we introduce the necessary changes. Finally, we state a theorem giving conditions for various boundedness notions, the proof of which may be constructed on the basis of the proof of Theorem 12.10.1 and the corresponding boundedness results of Chapter 3.
THEOREM 12.10.4. Assume that (i) Y ECCJ x E, R,], Y(t,x) is locally Lipschitzian in x, and, for (t, x) E J x E, &(x,
W )<
w,4 e
a(+,
V),
where a, b E X and b(u) co as u -+ 00; (ii) g E C[J x R , , R], and, for ( t , x) E J x E, ---f
(iii) limh+O+R(h, A(t))x
=x
for t E J and x
E E.
Then, one of the boundedness conditions with respect to the scalar differential equation (10.2.1) implies the corresponding one of the boundedness conditions of the set M with respect to the system (12.6.2).
272
CHAPTER
12
12.11. Notes
T h e existence theorems 12.1.1 and 12.1.2 of Sect. 12.1 are due to Murakami [3]. See also Browder [2] and T. Kato [4]. Theorem 12.2.1 is taken from the work of Mlak [5]. Theorems 12.3.1, 12.3.2, and the example that follows are due to Murakami [3]. The results of Sect. 12.4 are due to Antosiewicz [l]. Section 12.5 consists of the work of Murakami [3]. T h e results of Sect. 12.6 are taken from Lakshmikantham [4]. For the work contained in Sect. 12.7, see Mlak [8, 131. Theorems 12.8.1 and 12.8.2 are due to Mamedov [2], whereas Theorem 12.8.3 is adopted from the work of .Lakshmikantham [4]. The results of Sects. 12.9 and 12.10 are due to Lakshmikantham [3]. For the existence of solutions of functional equations using Lyapunov functions, see Hartman [11, where, applying analogous arguments, existence theorem for an initial value problem for ordinary differential equations in Hilbert spaces is discussed. Concerning abstract differential inequalities, see Cohen and Lees [l] and Edmunds [I]. Differential equations on cones in Banach spaces are treated in Coffman [l-31, Chandra and Fleishman [l, 21, and Szarski [8]. For an excellent treatment on linear differential equations and function spaces, see Massera and Schaffer [l]. For related work, see Agmon and Nirenberg [l], Browder [4], Foias et al. [3], T. Kato [1-4], Kato and Tanabe [I], Kisynski [2], Krasnoselskii [2], Krasnoselskii, et aE. [I, 21, Krein and Prozorovskaya [l], Lees [l], Lions [l], Lyubic [l], Minty [ l , 21, Mlak [5, 10, 12, 14, 15, 181, Ramamohan Rao and Tsokos [l], Sobolevski [l, 21, Taam and Welch [I], and Tanabe [ l , 21.
COMPLEX DIFFERENTIAL EQUATIONS
This page intentionally left blank
Chapter 13
13.0. Introduction An important extension of the initial value problem of ordinary differential equations is to the case where the time variable t and the space variables x may be complex. We treat in the present chapter various problems connected with this type of differential systems. We consider the existence and uniqueness problems, obtain error estimates of approximate solutions, discuss the singularity-free regions of solutions, and derive a priori bounds. Introducing Lyapunov-like functions, we prove comparison theorems, which may be used for stability criteria. All our considerations depend crucially on the technique of reducing the study of the behavior of solutions of complex differential systems to that of certain ordinary differential equations.
13+1+Existence, approximate solutions, and uniqueness Let Cn denote complex euclidean n space, represent C1. For any element y E Cn,we mean norms 11 y /I = CL I yi I, I/ Y II = ( y * i = 1 , 2, ..., n. Let f be an analytic complex defined on the domain D
=
[ ( x , Y )E Cn+' : 0
<jz i
and we shall use C to by 11 y 11 one of the usual and II y /I =: maxi yi I , valued vector function
< U , j / y / / < b, a, b > 01.
We consider the initial value problem of complex differential system given by
2 = f(z,39, dz
Y(0) = Y o ,
Yo 6 C".
(13.1. l )
T h e existence and uniqueness of the solutions of (13.1.1) may be 275
276
13
CHAPTER
inferred from the method of successive approximations. We merely state this well-known result.
THEOREM 13.1.1. Suppose that the function f is regular-analytic and bounded in D.Let Ilf(z,y)iI = M and n: = min(a, b/M). Then, there exists a unique regular-analytic function y ( z ) defined on 0 < I x I < a, which satisfies (13.1.1). We now consider the following complex differential inequality,
(13.1.2)
where f is regular-analytic in
D.and E > 0 is a real number.
DEFINITION 1 3.1.1. Any complex valued vector function y ( x , E ) is said to be an +approximate solution of (13.1.1) if the following conditions are satisfied: (i) y ( z , C) is regular-analytic in 0 I z 1
<
If E = 0 in (13.1.2), it is to be understood that y ( x ) is a solution of (1 3.1 .I). T h e following result gives the upper and lower bounds of the norm of the difference between any two approximate solutions of (13.1.1).
THEOREM 13.1.2. Let g E C[[O, a ) x R, , R,], f ( z ,y ) be regularanalytic in D,and
< g(l 2 I> I/Y1
IIf(z, Yl) -f(%Y2)ll
-
Y2 1 ).
(13.1.3)
Suppose further that y ( z , el) and y ( z , c2) are el- and E,-approximate solutions of (13.1.1) such that 0
< P(0)
< I/ Yl(0,
€1) - Y2(0,.2)iI
(13.1.4)
G Y(O)*
Assume that r ( t ) and p ( t ) are maximal and minimal solutions of ordinary differential equations r'
-
f'
=
respectively, existing on 0 0<1 x I a, we have < _
g(4 .)
+ + +
-t€1
-k(t,P)
< t < a.
€2,
€1
(13.1.5)
€21,
Then, on each ray z
=
teie and
13.1.
277
EXISTENCE, APPROXIMATIONS, UNIQUENESS
Proof. Let y ( z , and y ( z , c 2) be the approximate solutions of (13.1.1) satisfying (13.1.4). Let z = teie and arg z = const. Define m(t) = )Iy(teie,el)
For small h
-
y(teis,c2)11.
> 0, we obtain
1 m(t
+ h) - m(t)l < II[y((t + h)eis, -
y((t
el)
-
+ h) eie, c2)
y(teis,el)
-
(13.1.7)
y(teis,c2)]11.
Also, we have _________
dt
=
1 dy(z,dz
cl)eis -__dy(z, c,)eie
dz
1 (1 3.1.8)
Hence, using (13.1.7) and (13.1.8), we get
where m+’(t) is the right-hand derivative of m ( t ) with respect to t. Moreover, dz
+ IIf(z9Y k , 4)-f(z, Y(.,
4l.
It therefore follows, in view of (13.1.2) and (13.1.3), that
I m+Yt)l e g(t, 4 t ) ) + €1 + €2 and, consequently, - [ g ( t , m(t))
1
+ + < m+’(t) < g(4 4 0 ) + + €1
€21
€1
€2
*
I t is now easy to prove the stated estimate (13.1.6) by repeating the proof of Theorem I .2.1 with appropriate changes.
278
CHAPTER
13
> 0, is admissible
COROLLARY 13. I . 1. T h e function g(t, u ) = ku, K ?‘heorem 13.1.2. Indeed, this choice of g yields
p(o)e
-I
L t.L k (e-lLt E
‘
~
1)
< 11 y(teZe,
E ~ )-
in
y(teZe,E.J~
which corresponds to a well-known inequality in ordinary differential equations. LVe now provc a general uniqueness theorem for complex differential equations which extends a similar result in ordinary differential equations, namely, Theorem 2.2.8. Suppose that y,(z) and y > ( z )are any two solutions of (13.1.1). As before, let arg z = const, 1 z 1 = t , and
74t)
= IlYl(4
( 13.1.9)
-Y2(4ll.
I,et the function B(t) > 0 be continuous on 0 Suppose that
and B(O+)= 0. ( 13.I. 10)
THEOREM 13.1.3. Let y l ( z ) and y 2 ( z )be any two solutions of (1 3.1.1)
satisfying (13.1.10), where B(t) is the function just defined. Let g ( t , Z L ) 2 0 be continuous on 0 -.t --.a and zi 3 0 and g ( t , 0) = 0. Assume that the only solution u(t) of (13.1.11)
u‘ -- g ( t , u )
such that
( 13.1 .12 ) IS the
tribial solution. Suppose further that
lif(9,rd for 0
, 2 a , ijyl 11, sollltlon of (13. I.I ) on 0
f(2,
~ $ 1 G A1 I, /i ~1
- yz
II)
i / y 211 , b. Then, there exists at most one
<
z
,
a.
Ptoof. Suppose that there exist t n o solutions y,(z) and y (z)of (13.1.1) and that ( I 3.1.10) holds. J’rocceding as in the proof of ‘I’hcorem 13.1.2 114 rnoditications, it I S easy to obtain the differential inequality 111
’(f)
y(f,m(t)),
0 .I
a,
13.2.
SINGULARITY-FREE REGIONS
279
where m(t) is the function defined by (13.1.9). T h e rest of the proof follows closely the proof of Theorem 2.2.8. Hence the proof is complete. REMARK13.1.1. If B(t) == t , Theorem 13.1.3 reduces to an extension of Kamke’s general uniqueness theorem 2.2.2 for complex differential equations. Let us consider the following example. Let y’
=
2y z ,
y(0)
-
0.
This has a unique solution y ( z ) = 0 if we consider the solutions in the classical sense, whereas, for each complex K , y ( z ) = Kz2 represents a solution in the extended sense. I n the first case, taking B(t) = t2, we see that the condition (13.1.10) is verified. Moreover, since the differential equation (13.1.1 1) takes the form u’ = 2u/t, all the assumptions of the uniqueness theorem are fulfilled. On the other hand, when we consider the latter case, even though the choice of B(t) = t Z p f ,E > 0 being an arbitrarily small number, satisfies the condition (13.1.1 I), the assumption (1 3.1.12) is violated, since the solutions of (13.1.I I ) are u(t) = Lt2, where L is an arbitrary nonnegative constant.
13.2. Singularity-free regions and growth estimates Let us first of all be concerned with the solutions of the complex equation
3 + F ( z , y ) = 0, dz2
(13.2.1)
where F ( z , y ) is an entire function of y and regular-analytic in z for 1 z 1 < a. I n addition, we assume that
for all y , j x I < a, where G E C[[O,u ) x R, , R,] and G(t, u) is nondecreasing in u for each t. DEFINITION13.2.1. A region S is said to be singularity-free for a solution y ( z ) if y ( z ) is regular-analytic in S.
THEOREM 13.2.1. Assume that F ( z , y ) is subjected to the conditions
280
CHAPTER
13
stated previously. If there exists a function u(t), which is defined and continuous in [0, a ) such that u'(0) = p >, 0,
2 0,
u(0) = a:
and that zt"(t)
0
>, G(t, u(t)),
012
+
p2
> 0,
< a,
(13.2.3)
then any solution of (13.2.1) for which
is regular for J z
1 < a.
Proof. Suppose that arg z integral equation
= 6' = const.
Then, from (13.24, we have the
where the integration is carried out along the ray 6'
=
const. Then,
(13.2.4) where s
=
I 5 1, t
1 z 1. Also, since u(t) satisfies (13.2.3), it
=
follows that
(13.2.5)
Setting m ( t ) 13.1.2, that
=
1 y(teis)Jfor fixed
8, it is easily checked, as in Theorem
Hence, in particular,
I i
1 m+'(O)\ < dY(0) dz
=
/3
<6 +S
for every
S > 0.
We consider the function u(t, 8) that satisfies (13.2.2) and u(0, S)
= a:,
u'(0, 6 ) = (6
+ s.
13.2. SINGULARITY-FREE REGIONS
It is easy to show that m(t) < u(t, 6 ) for 0 contrary, there exists a t, > 0 such that
28 1
< t < a.
Supposing the (13.2.6)
and (13.2.7)
<
Notice that u(t) < u(t, a), 0 t < a . Using this fact, the monotony of G(t, u ) in u, the inequalities (13.2.5) and (13.2.7) lead to
I n other words,
which contradicts (13.2.7). Thus, we have m ( t ) = I y(teie)l < u(t, S),
0
< Iz I
< a.
Hence, letting 6 40, we arrive at the inequality
IY(4l
< 4%
0
< I z I < a-
(13.2.8)
Since u(t) is assumed to exist on [0, a), the stated result follows from (13.2.8), and the proof is complete.
COROLLARY 13.2.1. T h e complex equation ~
d2Y +p(z)yn dz2
=
has solutions, which are regular in I z
Ip(z)I
< a(1
-
1z
0,
n
> 1,
(1 3.2.9)
I < 1 if (Y-2,
E
> 0.
(13.2.10)
Proof. I n view of Theorem 13.2.1, it is sufficient to show that there exists a solution of
(13.2.11)
which is continuous on [0, 1). We consider whether (13.2.11) can have
282
CHAPTER
a solution of the form /?(I we must have
- t)-p,
p,
p > 0. If
(1
__ qz+nu--e
+
P b 1) -~
(1
qu+2
~
13
u p 1
such a solution exists,
'
that is to say, p(p
Since n
+ 1)
and
= u p 1
p = np
- c.
> 1, we obtain >0
E
p =
n-1
E(2+ 1) n-1
'
12-1
= up-1.
If /? is determined by the last equation, u(t) = p(1 - t)-* will be a solution of (13.2.11) for which u(0) and u'(0) are positive. T h e proof is complete. We shall next consider the complex differential system (13.1. I), wherefis regular in z , 0 1 z I .< a and entire in y E Cn.T h e following theorem gives an upper bound of the norm of solutions of (13.1.1) along each ray z : tei8.
<
THEOREM 13.2.2. Assume that y
<
(i) f E Clz, f ( z ,y) is regular-analytic in z , 0 1z I Ci8, and, for each fixed 9, 0 0 < 277, and z = teie,
<
E
< a,
entire in
Ilf(Z>Y)ll G dl I. I/ Y II),
(13.2.12)
where g E C[[O,0) x R, , R,]; (ii) v(t) is the maximal solution of the scalar differential equation 24'
u(0)
= g(t, u),
== 21"
(13.2.13)
:> 0,
whose maximal interval of existence is [O, b(9)), b(0)
< a. < uo , is
'Then, every solution y ( z ) of (13.1.1), such that 0 i (1 y(O)[l regular-analytic in a region that contains the set E
-.:
[z : 2
teiO,0
~-:
Furthermore,
1I Y ( W I l on each ray z
=
te", 0
< t < b(0).
< 469,0
< r(t)
< 6 < 2711.
13.2. Proof.
Let z
= teie, and
SINGULARITY-FREE
283
REGIONS
< 0 < 2n. Define
fix 0, 0
m(t> = I/Y(teis)ll,
where y ( z ) is any solution of (13.1.1) such that 0 < 11 y(0)ll 6 u,, . Then, as in Theorem 13.1.2, we derive the differential inequality
< g(l z I>/IY(.)l) = g(4
Moreover, m(0)
< u,, . Hence,
m(t>>.
by Theorem 1.4.1, we have
<
on
m ( t ) = (/y(teis)/) r ( t )
Therefore, y ( z ) is regular for z the proof is complete.
=
teie, 0
[0, b(0)).
< t < b(0), 0 < 0 < 2n,
and
We state a corollary that generalizes Theorem 13.2.1.
COROLLARY 13.2.2. If, in addition to the hypotheses of Theorem 13.2.2, we assume that the solutions of (13.2.13) are unique and there exist continuous functions u ( t ) on [0, ,(0)), .(0) > 0, 0 < 0 < 2 ~ , such that u ’ ( t ) 2 g(t, u ( t ) ) , 4 0 ) = uo , on [0, a(0)), then every solution of (13.1.1) for which IIy(0)jI regular in a region containing the set E
=
[z : x
=
teis, 0
< ~ ( 0 )0,
< u,, is
< 6 < 2~1.
COROLLARY 13.2.3. Let hypothesis (i) of Theorem 13.2.2 hold. Suppose that p ( t ) is the minimal solution of u’ = -g(t, u), u(0) = uo > 0 existing on [0, b(0)). Then, every solution y ( z ) of (13.1.1) such that / / y(0)Ij 3 u,, satisfies IIy(tei8)ll 2 p ( t ) ,
0
< t < WO).
THEOREM 13.2.3.
Let the assumptions of Theorem 13.2.2 hold, except that the condition (13.2.12) is replaced by a weaker condition, IIY
+ hf(%Y)ll < IIY I/ + hg(l z I, IlY II) + O(h),
284
CHAPTER
13
for sufficiently small h > 0. Then, the conclusion of the Theorem 13.2.2 remains valid. I t is easy to see that the weaker condition of Theorem 13.2.3 is sufficient to get the inequality D+m(t)
< g(4 4)),
as in the proof of Theorem 13.2.2. I n the present case, since g ( t , u)need no longer be nonnegative, the growth estimates are sharper. 13.3. Componentwise bounds
We shall first prove a result that offers componentwise bounds for solutions of the complex differential system (13.1.1).
THEOREM 13.3. I . y
Assume that
(i) f E Crl,f ( z ,y ) is regular-analytic in z , 0 C", and, for each fixed 8, 0 B < 2 ~ and , z
<
t
I%Y)l
<1z I
< a,
entire in
= tei9,
(13.3.1)
G g(l z I, I Y I),
whereg E C[[O,a ) x R f n ,R+"], g(t, u)is quasi-monotone nondecreasing in u for each t E [0, a ) ; (ii) the maximal solution r ( t , 0, uo) of the ordinary differential system u'
40)
g(t, u ) ,
=
uo
exists on [0, a ) .
2 0,
Then, every solution y ( z )of (13.1. I ) such that I y(0)I where 1 z
I
I y(tePS)I< r ( 4 0, U"), =
t.
Proof. I,et z function
--
tetB,
and we fix 8, 0 m(t> =
t
< uo satisfies
3 0,
< 8 < 2n.
I Y(teze>l,
where y(x) is any solution of (13.1 . l ) such that I y(0)I
Define the vector
< uo . Proceeding
as in the proof of Theorem 13.1.2 with obvious modifications, it is
easy to obtain the differential inequality Dfnz(t)
< g(t, m ( t ) ) .
Corollary 1.7. I now assures the stated componentwise bounds.
13.3.
COMPONENTWISE BOUNDS
285
Analogous to Theorem 13.2.3, we can state a theorem for componentwise bounds which yields sharper bounds in some situations.
THEOREM 13.3.2. Let the condition (13.3.1) in Theorem 13.3.1 be
replaced by
IY
+ hf(z,Y>l < I Y I + hg(l
I, I Y I) -1 W )
for all small h > 0, where g E C[[O, a ) x R+" , Rn], and g ( t , u ) is quasimonotone nondecreasing in u for each t E [0,a), other assumptions remaining the same. Then, the conclusion of Theorem 13.3.1 is true. Instead of the complex differential inequality (13.1.2),we shall consider the system of inequalities (13.3.2)
where E is a positive vector. Definition 13.1.1 has to be slightly modified in an obvious way. Corresponding to Theorem 13.1.2, we have the following
THEOREM 13.3.3. Let g E C[[O, a ) x R+", R f n ] g, ( t , u ) be quasi-monotone nondecreasing in u for each t E [0, a), and r(t) be the maximal solution of the system u'
= g(t,
u)
+
E,
>0
u(0) = uo
existing on [0, u). Suppose further thatf(z, y ) is regular-analytic in D and
If yl(z, el) and y z ( z ,c2) are such that
el-
I YdO, €1) then we have, on each ray z
and <,-approximate solutions of (13.1.1) -
Y z P , 4< uo
= teis and
0
<1z 1
9
< a,
When = 0, c2 = 0, y l ( z )and y,(z) are any two solutions of (13.1.1), and Theorem 13.3.3 gives componentwise upper bounds for the difference of these solutions.
286
CHAPTER
13
13.4. Lyapunov-like functions and comparison theorems
Let D denote the region of the complex plane I z 1 3 a and
(?I
< arg z < /3, where a, a,/3 are real numbers. We consider the complex
differential system
Y'
= f ( z ,Y ) ,
y(zo) = Yo ,
zo E
D,
(13.4.1)
where y and f are n vectors, and the function f(z,y ) is regular-analytic in z on L) and entire in y E C". As before, we shall denote 1 z I by t and arg z by 19. Moreover, let
1tI
= (I
5, I, I (2
I,..., I
t n I).
Let a function V ( x , y ) , where V and y are n vectors be defined, regular-analytic in z on D and entire in y E C". T h e ith component of V will be denoted by Vi(z,y ) or synonymously by Vi(z,y1 ,y 2 ,..., yn) whenever necessary. We define (13.4.2)
With respect to the functions defined previously, we state the following comparison theorems.
THEOREM 13.4.1.
Assume that
(i) g F C [ / x R , n, R ,"1, g(t, a ) is quasi-monotone nondecreasing in u for each t , and
) r ( t , to,u,,) is the maximal solution of the ordinary differential (ii) ~ ( t= system 21' = g ( t , a), u(t,) = 240, (13.4.4)
existing on [ t o , a]; (iii) y ( z ) is any solution of (13.4.1) such that (13.4.5)
Then w e have, for z
E
D, (13.4.6)
where 1 z 1
=
t.
13.4.
287
LYAPUNOV-LIKE FUNCTIONS
THEOREM 13.4.2. Let assumptions (i) and (ii) of Theorem 13.4.1 hold except that the condition (13.4.3) is replaced by
wherep(2:) is regular-analytic in 2: on D. Ify(z) is any solution of (13.4.1) satisfying
I v ( z o > Y ( Z 0 ) ) P(.o>l
then
I J+,
Y(Z))P(Z)l
< r(to>,
< r(t>,
z
I z o I = to
E D,
I2 I
( 13.4.8)
?
=
4
(13.4.9)
for all t 3 t o . We shall prove below Theorem 13.4.2, since Theorem 13.4.1 can be deduced from Theorem 13.4.2 by taking p(2:) = 1. We have stated Theorem 13.4.1 separately, as it is a basic comparison theorem by itself.
Proof of Theorem 13.4.2. Define
wherey(2) is any solution of (13.4.1) verifying (13.4.8). For each fixed 8, set nz(t) = I L(teie,y(te"))l.
Then, if h m(t
> 0 is sufficiently small,
+ h)
~
m(t)
.< I L((t + h)eie,y ( ( t + h)eiO))
-
L(teis,y(teie))l.
We can easily verify that
Also, dL(teie,y(teie))
I=)
dL(z,y(z))eie dz
288
CHAPTER
13
It therefore follows from the foregoing considerations that
< g ( 4 %(q,...,m,(l)). Now a straightforward application of Corollary 1.7.1 yields the desired inequality (13.4.9).
13.5. Notes T h e results of Sect. 13.1 are due to Deo and Lakshmikantham [l]. Theorem 13.2.1 and Corollary 13.2.1 are taken from the work of Das [l]. See also Das [4]. Theorem 13.2.2 is due to Wend [2], whereas Theorem 13.2.3 is new. T h e results of Sects. 13.3 and 13.4 are adapted from the work of Kayande and Lakshmikantham [l]. For further results, see Deo and Lakshmikantham [2] and Kayande and Lakshmikantham El], where stability and boundedness criteria are discussed.
Bibliography
AGMON,S., AND NIRENBERG, L. [l] Properties of solutions of ordinary differential equations in Banach spaces, Comm. Pure Appl. Math. 16 (1963), 121-239. ALEXIEWICZ, A., AND ORLICZ, W. [I] Some remarks on the existence and uniqueness of solutions of the hyperbolic equation a2z/ax ay = f (x, y , z, a z / a x , &lay), Studia Math. 15 (1956), 201-215. ANTOSIEWICZ, H. A. [l] Continuous parameter dependence and the method of averaging, Proc. Int. Symp. Nonlinear Oscillations, 2nd, Izd. Akad. Nauk. Ukrain, SSR, Kiev, 1963, pp. 51-58. ARONSON, D. G. [l] On the initial value problem for parabolic systems of differential equations, Bull. Amer. Math. SOC.65 (1959), 310-318. [2] Uniqueness of solutions of the initial value problem for parabolic systems of differential equations, /. Math. Mech. 11 (1962), 403-420. [3] Uniqueness of positive weak solutions of second order parabolic equations, Ann. Polon. Math. 16 (1965), 285-303. ARONSON, D. G., AND BESALA, P. [I] Uniqueness of solutions of the Cauchy problem for parabolic equations, Anal. Appl. 13 (1966), 5 16-526.
/. Math.
AZIZ,A. K. [l] A functional integral equation with applications to hyperbolic partial differential equations, Duke Math. J. 32 (1965), 579-592. [2] Periodic solutions of hyperbolic partial differential equations, Proc. Amer. Math. SOC.17 (1966), 557-566. AZIZ,A. K., AND DIAZ,J. [l] On a mixed boundary value problem for linear hyperbolic partial differential equations in two independent variables, Arch. Rational Mech. Anal. 10 (1962), 1-28. 289
290
BIBLIOGRAPHY
Aziz, A. I<., AND MALONEY, J. P. [I] An application by Tychonoff’s fixed point theorem to hyperbolic partial differential equations, Math. Ann. 162 (1965), 77-82. BARRAK, I<. B. [I] Some cstimates for solutions of parabolic equations, (1961), 373-397.
1.Math.
Anal. Appl. 3
BELLRIAN, R. [ 11 On the boundedness of solutions of nonlinear differential and difference equations, Irans. Amev. Mnth. SOC.62 (1947), 357-386. [2] On the existence and boundedness of solutions of nonlinear partial differential equations of parabolic type, Tvans. Amer. Math. Soc. 64 (1948), 21-44.
,.
[3] On the existence and boundedness of solutions of nonlinear differential-difference equations, Ann. of Math. 50 (1949), 347-355. [4] T h e stability theory of differential-difference equations, RAND Corp. Paper No. P-381, 1953. [5] Terminal control, time lags, and dynamic programming, Proc. Nat. Acad. Sci. U.S.A. 43 (1957), 927-930. [6] Asymptotic series for the solutions of linear differential-difference equations, Re77d. Circ. Mat. Palernzo [2] 7 (1958), 1-9. [7] On the computational solution of differential-difference equations, /. Math. Anal. Appl. 2 (1961), 108-110. BELLRIAN, R., AND COOKE, K. L. [I] Stability thcory and adjoint operators for linear differential-difference equations, Trans. Amw. Math. Soc. 92 (1 959), 470-500. [2] Asymptotic behavior of solutions of differential-difference equations, Mem. Amer. Math. SOC.35 (1959). [3] O n the limit of solutions of differential-difference equations as the retardation approaches zero, Proc. Nut. Acad. Sci. U.S.A. 45 (1959), 1026-1028. [4] “Differential-Difference Equations.” Academic Press, New York, 1963.
R., BELLMAN,
AND
DANSKIN, J. M.
[I] T h e stability theory of differential-difference equations, Proc. Symp. Nonlinear Civcnit A77nl., New York, 1953, pp. 107-123. [2] A survey of mathematical theory of time lag, retarded control, and hereditary processes, RAND Corp. Rep. No. R-25\5 1956. BELLMAN, I<., DANSKIN, J. M., AND GLICKSDERG, I. [11 A bibliography of the theory and application of differential-difference equations, RAND Corp. Res. Memo. No. KM-688, 1952. BBSALA, P. (see Aronson, D. G.) [l] On solutions of nonlinear parabolic equations defined in unbounded domains, Biill. Acad. Polon. Sci. Sir. Sci. Math. Astronom. Phys. 9 (1961), 531-535. [2] On solution of Fourier’s first problem for a system of nonlinear parabolic equations in an unbounded domain, Atrir. Pol. Math. I3 (1963), 247-265.
BIBLIOGRAPHY
29 1
[3] Concerning solutions of an exterior boundary-value problem for a system of non-linear parabolic equations, Ann. Pol. Math. 14 (1964), 289-301. [4] Limitations of solutions of non-linear parabolic equations in unbounded domains, Ann. Pol. Math. 17 (1965), 25-47. BHATIA, N. P.
[l] Stability and Lyapunov functions in dynamical systems, Cont. Dzflerentiul Equations 3 (1964), 175-188. [2] On exponential stability of linear differential systems, J. SIAM Control. Ser. A. 2 (1965), 181-191. [3] Weak attractors in dynamical systems, Bol. SOC.Mat. Mexicunu 11 (1966), 56-64. [4] O n asymptotic stability in dynamical systems, Math. Systems Theory 1 (1967), 113-127. BHATIA, N . P., AND SZECO, G. P. [I] Weak attractors in R“, Muth. Systems Theory I (1967), 129-133. [2] “Dynamicdl Systems: Stability Theory and Applications” (Lecture Notes in Math., No. 35). Springer, New York, 1967. BIELECKI,A. [ I ] Une remarque sur l’application de la methode de Banach-Cacciopoli-Tikhonov dans la thkorie de 1’Cquation s = f (x, y , z, p , q), Bzill. Acud. Polon. Sci. CL. 1114 (1956), 265-268. BOGDANOWICZ, W. M., AND WELCH,J. N. [I] On a linear operator connected with almost periodic solutions of linear differential equations in Banach spaces, Math. Ann. 172 (1 967), 327-335.
F. E. BROWDER, [I] Non-linear parabolic boundary value problems of arbitrary order, Bull. Amer. Math. SOC. 69 (1963), 858-861. [2] Nonlinear equations of evolution, Ann. of. Muth. 80 (1964), 485-523. [ 3 ] Strongly non-linear parabollc boundary value problems, Amer. /. Muth. 85 (1964), 339-357. [4] Nonlinear monotone operators and convex sets in Banach spaces, Bull. Amer. Math. SOC.71 (1965), 780--785. S. BRZYCHCZY, [l] Some theorems on sccond order partial differential inequalities of parabolic type, Ann. Polon. Muth. 15 (1964), 143-151. [2] Extensions of Chaplighin’s method to the system of non-linear parabolic equations in a n unbounded domain, BidI. Acud. Polon. Sci. St+. Sci. Muth. Astronom. Phys. 13 (1965), 27-30.
BULLOCK, I<. 31. [ l ] A uniqueness theorem for a delay-differential system, SIAA%ZRev. 9 (1967), 737-740.
292
BIBLIOGRAPHY
CALDJXON. A. P [ l ] Uniqueness in the Cauchy problem for partial differential equations, Amer. J . Math. 80 ( I 958), 16-36. CAMERON, R. H [I] Nonlinear Volterra functional equations and nonlinear parabolic differential systems, 1.Aizalyse Math. (1956/57), 136-182.
CESARI, L. [ I ] Existence in the large of periodic solutions of hyperbolic partial differential equations, Arch. Rational Mech. Anal. 3 (1965), 170-1 90. [2] Periodic solutions of hyperbolic partial differential equations, Intern. Symp. Nonlinear Eqs. rind Nonlinear Mechanics, Academic Press, New York, 1963. pp. 36-57.
J., AND ~iT.EISHMAN, B. A. [I] Positive solution of nonlinear equations, Proc. U.S.-Japan Sem. Differential and Fiinctional Eqs. Minneapolis, Benjamin, New York, 1967, pp. 435-442.
CHANDHA,
[2] Bounds and maximal solutions of nonlinear functional equations, Bull. Amer. Math. SOC.74 (1968), 512-516. CHANC, H. [I] On the stability of systems of differential equations with time lag, Acta Math. Sinica. 10 (1960), 202-21 1. CHU,
S.
c.
[ I ] On a mixed boundary value problem for linear third order hyperbolic partial differential equations, Rend. Circ. Mat. Palermo [2] 13 (1964), 199-208. CILIBRRTO, C. [ I ] Sul problema di Darboux per I’equazione s = f ( x , y , z , p , q), Rend. Acad. Sci. Pis. Mat. Napoli 22 (1956), 221-225. [2] Sulle equazioni quasi-lineari di tipo parabolic0 in due variabili, Ricerche Mat. 5 (1956b), 97-125. [3] Sull’approxximazione delle soluzioni del problema di Darboux per l’equazione s = f ( x , y , z , p , q), Ricerche Mat. 10 (1961), 106-138.
[4] Teoremi di confronto e di unicita per le soluzioni del problema di Darboux, Riccvche Mat. 10 (1961a), 214-243. [5] S u l problema di Darboux, Atti Acnd. Naz. Lincei. Rend. CI. Sci. Fir. Mut. Nattty. 30 ( I 961b), 460-466. COFFRIAN, C . V. [ I ] Linear differential equations on cones in Banach spaces, Pacific J . Math. 12 ( I 962), 69-75. [21 Nonlinear differential cquations on cones in Banach spaces, Pacific 1.Math. 14 (1964), 9-16. [3] Asymptotic behavior of solutions of ordinary difference equations, Trans. Amer. M ~ t h SOC. . 110 (1964), 22-51.
BIBLIOGRAPHY
293
COHEN,P., AND LEES,M. [l] Asymptotic decay of solutions of differential inequalities, Pacific J. Math. 11 (1961), 1235-1 249. CONLAN, J. [l] T h e Cauchy problem and the mixed boundary value problem for a nonlinear hyperbolic partial differential equation in two independent variables, Arch. Rational Mech. Anal. 3 (1959), 355-380. [2] An existence theorem for the equation uZYs= f,Arch. Rational Mech. Anal. 9 (1962), 64-76. CONTI,R. [I] Sul problema di Darboux per l’equazione zzY= f (x, y , z, z, , z,,), Ann. Univ. Ferrara Sez. VZZ (N.S.) 2 (1953), 129-140. COOKE,K. L. (see Bellman, R.) [l] T h e asymptotic behavior of the solutions of linear and nonlinear differentialdifference equations, Trans. Amer. Math. SOC.75 (1953), 80-105. [2] T h e rate of increase of real continuous solutions of algebraic differential-difference equations of the first order, Pacific J. Math. 4 (1954), 483-501. [3] Forced periodic solutions of a stable nonlinear differential-difference equation, Ann. ofMath. 61 (1955), 381-387. [4] A symbolic method for finding integrals of linear difference and differential equations, Math. Mag. 31 (1958), 121-126. [5] Functional-differential equations: some models and perturbation problems, Proc. Intern. Symp. Differential Eqs. and Dynamical Systems, Puerto Rico, 1965, Academic Press, New York, 1967, pp. 167-183. [6] Functional differential equations close to differential equations, Bull. Amer. Math. SOC.72 (1966), 285-288. [7] Functional differential equations with asymptotically vanishing lag, Rend. Circ. Mat. Palermo 16 (1967), 39-56. [8] Asymptotic theory for the delay-differential equation u’(t) = -au(t - r(u(t))), J. Math. Anal. Appl. 19 (1967), 160-173. CORDUNEANU, C. [I] Theorkmes d’existence globale pour les systemes diffbrentielles B argument retarde, Studii Cere. Mat. Zasi 12 (1961), 251-258. [2] Sur la stabilitb des systemes perturbes B argument retarde, An. Sti. Univ. “Al. Z. Cuza”Zasi 11 (1965), 99-105.
DAS,K. M. [l] Singularity-free regions for solutions of second order non-linear differential equations, J. Math. Mech. 13 (1964), 73-84. [2] Comparison and monotonicity theorems for second order non-linear differential equations, Acta Math. Acad. Sci. Hungar. 15 (1964), 449-456. [3] Properties of solutions of certain non-linear differential equations, J. Math. Anal. Appl. 8 (1964), 445-451. [4] Bounds for solutions of second order complex differential equations, Proc. Amer. Math. SOC.18 (1967), 220-225.
294
BIBLIOGRAPHY
DEO,s. G.,
AND LAKSRMIKANTHAM,
v.
[I] On complex differential inequalities, Bol. SOC.Mat. Sao Pazilo 18 (1966), 3-8. [2] Complex differential inequalities and extension of Lyapunov method, Proc. Nut. Acad. Sci. India Sect. A 36 (1966), 217-242. [3] Conditional stability and functional differential equations. Proc. Symp. Anal., Annamalai UTziv.,India, Math. Student (Special Issue) (1966). [4] Functional differential inequalities, Bull. Calcutta Math. SOC.58 (1966), 89-95.
DIAZ,J. B. (see Aziz, A. K.) [I] On an analogue of the Euler-Cauchy polygon method for the numerical solution of u Z v= f (x, y , u, u , , u v ) ,Arch. Rational Mech. Anal. 1 (1958), 357-390. [2] On existence, uniqueness, and numerical evaluation of solutions of ordinary and hyperbolic differential equations, Ann. Mat. Pura Appl. 52 (1960), 163-181. R. D. DRIVER, [l] Delay-differential equation and an application to a two-body problem of classical electrodynamics, Tech. Rep., Univ. of Minnesota, Minneapolis, 1960. [2] Existence theory for a delay-differential system, Contr. Differential Equations 1 (1961), 317-336. [3] Existence and stability of solutions of a delay-differential system, Arch. Rational Mech. Anal. 10 (1962), 401-426. [4] Existence and continuous dependence of solutions of a neutral functionaldifferential equation, Arch. Rational Mech. Anal. 19 (1965), 149-166. [5] On Ryabov’s asymptotic characterization of solutions of quasi-linear differential equations with small delays. SIAM Rev. 10 (1968), 329-341. EDMUNDS, D. E.
[I] Differential inequalities, Proc. London Math. SOC.15 (1965), 361-372. S. D [I] Bounds for solutions of parabolic systems and some applications, Mat. Sb. 33 (1953), 359-382. [2] On the connection between the fundamental matrix-solution of parabolic and elliptic systems, Mat. Sb. 35 (1954), 57-72. [3] On the fundamental solution of parabolic systems, Mat. Sb. 38 (1956), 51-92. [4] The fundamental matrix of general parabolic systems, Dokl. Akad. Nauk SSSR (N.S.), 120 (1958), 980-983. [ 5 ] On the fundamental solution of parabolic systems, Mat. Sb. 95 (1961), 73-136.
EIDELMAN,
L. E. EL’SGOL’TS, [l] An approximate method of solution of differential equations with a retarded argument, Uspehi Mat. Nauk 6 (1951), 1-30. [2] On approximate integration of differential equations with retarded argument, Prikl. Mat. Meh. 15 (1951), 771-772. [ 3 ] Some questions in the theory of differential-difference equations, Uspehi Mat. NU& 7 (1952), 147-148.
BIBLIOGRAPHY
295
[4] Stability of solutions of differential-difference equations, Uspehi Mat. Nauk 9 (1954), 95-112. [ 5 ] On the integration of linear partial differential equations with retarded argument, Moskov Gos. Univ. Ucen. Zap. Mat. 181 (1956), 57-58. [6] On the theory of stability of differential equations with perturbed arguments, Vestnik Moskov. Univ. Ser. Mat. Meh. Astronom Fiz. Him. 5 (1959), 65-72. [7] Qualitative methods in mathematical analysis (trans. from Russian), Amer. Math. SOC.Providence, 1964. EL’SGOL’TS, L. E., AND MYSHKIS, A. D. [I] Some results and problems in the theory of differential equations, Russian Math. Surveys 22 (1967), 19-57. EL’SGOL’TS, L. E., MYSHKIS, A. D., AND SHIMANOV, S. N. [I] Stability and oscillations of systems with time lag, Int. Union Theoret.Appl. Mech. Symp. Nonlinear Vibrations,Ukr. SSR Acad. Sci. Publ., Kiev, 1961. EL’SGOL’TS, L. E., ZVERKIN, A. M., KAMENSKII, G. A., AND NORKIN, S. B. [I] Differential equations with a perturbed argument, Russian Math. Surveys 17 (1962), 61-146. 121 Differential equations with retarded arguments, Uspehi Mat. Nauk 17 (1962), 77-164.
FODCUK, V. I. [l] Existence and properties of an integral manifold for nonlinear systems of equations with retarded argument and variable. coefficients. Approximative methods of solving differential equations, Izv. Akad. Nauk SSSR (1963), 129-140.
FOIAS, C., Gussr, G., AND POENARU, V. [l] Une mkthode directe dans l’etude des equations aux derivtes partielles hyperboliques quasilineaires en deux variables, Math. Nachr. 15 (1956), 89-1 16. [2] Verallgemeinerte Losungen fur parabolische quasilineare systems, Math. Nachr. 17 (1958), 1-8. [3] Sur les solutions generalistes de certaines equations lineaires et quasi-IinCaires dans l’espace de Banach, Rev. Math. Pure Appl. 3 (1958), 283-304. FRANKLIN, J. [I] On the existence of solutions of functional differential equations, Proc. Amer. Math. SOC.5 (1954), 363-366. FRIEDMAN, A. [I] Classes of solutions of linear systems of partial differential equations of parabolic type, Duke Math. J , 24 (1957), 443-442. [2] On the regularity of solutions of nonlinear elliptic and parabolic systems of partial differential equations, J . Math. Mech, 7 (1958), 43-60. [3] Interior estimates for parabolic systems of partial differential equations, J. Math. Mech. 7 (1958), 393-418. [4] Boundary estimates for second order parabolic equations and their applications, J. Math. Mech. 7 (1958), 771-792.
296
BIBLIOGRAPHY
[ 5 ] On quasi-linear parabolic equations of the second order, J . Math. Mech. 7 (1958), 793-809. [6] Remarks on the maximum principle for parabolic equations and its applications, Pacific J. Math. 8 (1958), 201-211. [7] Convergence of solutions of parabolic equations to a steady state, J. Math. Mech. 8 (1959), 57-76. [8] Asymptotic behavior of solutions of parabolic equations, J. Math. Mech. 8 (1959), 387-392. [9] Free boundary problems for parabolic equations. I. Melting of Solids, J. Math. Mech. 8 (1959), 499-518. [lo] Parabolic equations of the second order, Trans. Amer. Math. SOC.93 (1959), 509-530. [I21 On thc uniqueness of the Cauchy problem for parabolic equations, Amer. /. Math. 81 (1959b), 503-511. [ 131 Free boundary problems for parabolic equations. 11. Evaporations or condensation of a liquid drop, J. Math. Mech. 9 ( 1 960), 19-66. [I41 Free boundary problems for parabolic equations. 111. Dissolution of a gas bubble in liquid, J. Math. Mech. 9 (1960), 327-345. [I51 On quasi-linear parabolic cquations of the second order. 11, J. Math. Mech. 9 (1960a), 539-556. [I 61 Remarks on Stefan-type free boundary problems for parabolic equations, J . Math. Mech. 9 (1960), 885-904. [17] Mildly nonlinear parabolic equations with applications to flow of gases through porous media, Arch. Rational Mech. Anal. 5 (1960), 238-248. [ 181 Asymptotic behavior of solutions of parabolic equations of any order, Acta Math. 106 (196l), 1-43. 1191 A strong maximum principle for weakly subparabolic functions, Pacific J. Math 1 1 (1961), 175-184. [21] “Generalized Functions and Partial Differential Equations.” Prentice-Hall, Englewood Cliffs, New Jersey, 1963. [22] “Partial Differential Equations of Parabolic Type.” Prentice-Hall, Englewood Cliffs, New Jersey, 1964. [23] Asymptotic behavior of solutions of parabolic differential equations and of integral equations, Proc. Intern. Symp. Differential Eqs. and Dynamical Systems, Pirerto Rico, 1965, pp. 409-426. [24] Remarks on nonlinear parabolic equations, Proc. Symp. Appl. Math. 17 (1965), 3-23. G t r lL tA N O,
1,.
[ I ] Sull’unicita della soluzione per una classe di equazioni differenziali alle derivate parziali, paraboliche, non lineari, Atti Accad. Naz. Lincei Rend. Cl. Sci. Fis. Mat. Natnr. 12 (1952), 260-265. GUCLIELMINO, F. [I] Sulla risoluzione del problema de Darboux per l’equazione Boll. Un. Mat. Ital. 13 (1958), 308-318. [2] Sul problema di Darboux, Richerche Mat. 8 (1959), 180-196.
s = f ( x , y , z),
BIBLIOGRAPHY
297
[3] Sull’esistenza delle soluzioni dei problemi realtive alle equazioni non lineari di tip0 iperbolico in due variabili, Matematiche 14 (1959), 67-80. [4] Sul problema di Goursat, Ricerche Mat. 8 (1960), 91-105.
HAHN,W. [l] “Stability of Motion.” Springer, New York, 1967. HALANAY, A. [l] Some qualitative questions in the theory of differential equations with a delayed argument, Rev. Math. Pures AppZ. 11 (1957), 127-144. argument i retarde, C. R. Acad. [2] Solutions pkriodiques des systlimes IinCaires ? Sci. Paris Se‘r. A-B (1959), 2708-2709. [3] T h e method of averaging in equations with delay, Rev. Math. Pures Appl. 4 (1959), 467-483. [4] Periodic and almost periodic solutions of systems of differential equations with retarded arguments, Rev. Math. Pures Appl. 4 (1959), 685-692. [5] Stability criterions for systems of differential equations with a delayed argument, Rev. Math. Pures Appl. 5 (1960), 367-374. [6] Integral stability for systems of differential equations with a delayed argument, Rev. Math. Pures Appl. 5 (1960), 541-548. [7] Sur les systlimes d’kquations diffkrentielles lineaires B argument retarde, C. R. Acad. Sci. Paris Sir. A-B 250 ( 1 960), 797-798. [8] Solutions pkriodiques des systemes gbnkraux a retardement, C. R. Acad. Sci. Paris Sir. A-B 250 (1960), 3557-3559. [9] Almost periodic solutions of systems of differential equations with a delayed argument, Rev. Math. Pures Appl. 5 (1960), 75-79. [lo] Differential inequalities with time-lag and an application to a problem of the stability theory for systems with time-lag, Corn. Acad. R. P . Romaine 1 1 (1961), 1305-1310. [ l l j Periodic solutions of linear systems with delay, Rev. Math. Pures Appl. 6 (1961), 141-158. [I21 Stability theory of linear periodic systems with delay, Rev. Math. Pures Appl. 6 (1961), 633-653. [I31 Perturbations singulitres des systlimes a retardement, C. R . Acad. Sci. Paris Ser. A-B, 253 ( I 961), 1659-1 650. [14] Singular perturbations of systems with time lag, Rev. Math. Pures A p p f . 7 (l962), 627-63 1. [ 151 Asymptotic stability and small perturbations of periodic systems of differential equations with retarded arguments, Uspehi Mat. Nauk 17 (1962), 231-233. [16] Periodic and almost-periodic solutions of some singularly perturbed systems with time lag, Rev. Math. Pnrcs Appl. 8 (1963), 397-403. [17] Almost-periodic solutions of linear systems with time lag, Rev. Math. Pures Appl. 9 (1964), 71-79. [18] S y s t h e s a retardement. RCsultats et p r o b l h e s , 3rd Conf. Nonlinear Vibrations, East Berlin, 1964, Akademie-Verlag, Berlin, 1965. [I91 Periodic invariant manifold for a class of systems with time lag, Rev. Roumaine Math. Pnres Appl. 10 (1965), 251-259.
29 8
BIBLIOGRAPHY
[20] Invariant manifolds for systems with time lag, Proc. Intern. Symp. Dzfferential Eqs. and Dynamicnl Systems, Puerto Rico, 1965, pp. 199-213. [21] An invariant surface for some singularly perturbed systems with time lag, J. Differential Equations 2 (1966), 33-46. [22] “Differential Equations.” Academic Press, New York, 1966.
H A L E , J. K. [ I ] Asymptotic behavior of the solutions of differential-difference equations, RIAS Tech. Rcp. No. 61-10, 1961. [2] Functional-differential equations with parameters, Contr. Differential Equations 1 (1962), 401-410. [3] A class of functional-differential equations, Contr. Differential Equations 1(1962), 41 1-423. [4] Linear functional-differential equations with constant coefficients, Contr. Differential Equations 2 (1963), 291-317. [5] A stability theorem for functional-differential equations, Proc. Nut. Acad. Sci. U.S.A. 50 (1963), 942-946. [6] Periodic and almost periodic solutions of functional-differential equations, Arch. Rational Mech. Anal. 15 (1964). 289-304. [7] Geometric theory of functional-differential equations, Proc. Intern. Symp. Dtfferential Eqs. and Dynamical Systems, Puerto Rim, 1965, pp. 247-266. [8] Sufficient conditions for stability and instability of autonomous functionaldifferential equations, J. Diflerential Equations 1 (1969, 452-482. [9] Averaging methods for differential equations with retarded arguments and a small parameter, J. Differential Equations 2 (1966), 57-73. [ 101 Linear asymptotically autonomous functional differential equations, Rendi. cinc. Matem. Palermo 15 (1966), 331-351. [ I I] Periodic solutions of a class of hyperbolic equations containing a small parameter, Arch. Rational Mech. Anal. 23 (1967), 380-398. HALE,J. K., AND PERBLLO, C. [ 11 T h e neighborhood of a singular point of functional-differential equations, Contr. D[ffererrtiul Equations 3 (1954), 351-375. HARTMAN, P.
[ I ] Generalized Lyapunov functions and functional equations, Ann. M a t . Pura Appl. 69 (1965), 305-320. [2] On homotopic harmonic maps, Canad. J. Math. 19 (1967), 673-687. [3] O n hyperbolic partial differential equations, Amer. J. Math. 74 (1952), 834-864. HASTINGS, S. P. [I] Variation of parameters for non-linear differential difference equations, Proc. Amer. Math. Sac. 19 (1968), 1211-1216. [2] Backward existence and uniqueness for retarded functional differential equations, /. Differeutial Eqs. 5 (1969), 441-451. ILIN,A. M., KALASHNIKOV, A. S., AND OLEINIK, 0. A. [I] Linear second order parabolic equations, Uspehi Math. Nuuk 17 (1962), 3-146.
299
BIBLIOGRAPHY ITO,
s. [l] Fundamental solutions of parabolic differential equations and boundary value problems, Japan J. Math. 27 (1957), 55-102. G . S. [I] T h e existence of periodic solutions of f ‘ ( x ) = f ( x - 1)(1 f ( x ) ) , /. Math. Anal. Appl. 5 (1962), 435-450. [2] Asymptotic fixed point theorems and periodic systems of functional-differential equations, Contr. Differential Equations 2 (1963), 385-405. [3] Periodic motions in Banach space and applications to functional-differential equations, Contr. Differential Equations 3 (1964), 75-106. [4] Hereditary dependence in dynamics, Proc. Intern. Symp. Dzfferential Eqs. and Dynamical Systems, Puerto Rico, 1965, pp. 189-198. [5] Hereditary dependence in the theory of differential equations, I, 11, Univ. of Maryland Reps 1965.
JONES,
KAKUTANI, S., AND MARKUS, L.
+
+
[I] On the nonlinear difference-differential equation y ’ ( t ) = [ A By(t Contr. Theory Nonlinear Oscillations, Princeton 4 (1958), 41 1-418.
-
y)] y ( t ) ,
KAMENSKII, G. A. (see El’sgol’ts, L. E., Zverkin, A. M., and Norkin, S. B.)
[I] Asymptotic behavior of solutions of linear differential equations of the second order with lagging argument, Moskov. Gos. Univ. UZen. Zap. Mat. 165 (1954), 195-204. [2] Existence and uniqueness of solutions of differential-difference equations of neutral type, Moskov. Gos. Univ. Uten. Zap. Mat. 181 (1956), 83-89. [3] On the general theory of equations with perturbed argument, Dokl. Akad. Nauk SSSR 120 (1958), 697-700. [4] Existence, uniqueness, and continuous dependence on initial values of the solutions of systems of differential equations with deviating argument of neutral type, Mat. Sb.(N.S.) 55 (1961), 363-378. [5] Differential equations with a perturbed argument, Russian Math. Surveys 17 (1962), 61-416. KAPLAN, S. [l] O n the growth of solutions of quasi-linear parabolic equations, Comm. Pure Appl. Math. 16 (1963), 304-330. KARASIK, G. J. [I] On conditions for the existence of periodic solutions of difference equations, Izv. Vyss. UZebn. Zuved. 4 ( 1 9.59), 70-79. [2] On the existence of periodic solutions of a system of differential equations with retarded argument, Sibirsk Mut. Z . 2 (1961), 551-555.
KATO,T. [I] Asymptotic behaviors in functional-differential equations, Tohoku Math. /. 18 (1966), 174-215,
300
BIBLIOGRAPHY
T [ I ] Integration of the equation of evolution in Banach space, J . Math. SOC.Japan 5 (1953), 208-234. [21 On linear differential equations in Banach spaces, Cotrrm. Pure Appl. Math. 9 (l956), 479-486. [ 3 ) Abstract evolution equations of parabolic type in Banach and Hilbert spaces, N U ~ O JMath. O J . 19 (1961), 93-125. [41 Nonlinear evolution equations in Banach spaces, Mimeographed notes, Univ. of California, Berkeley (1964).
KA.I.o,
T., AND TANABE, 11. [I) O n thc abstract evolution equation, Omka f . Math. 14 (1962), 107-133.
KATO,
A. A , , A N D lAAI<SHRIIKANTHAXI, v. [ I ] Complex differential systems and extension of Lyapunov’s method. J. Math. Algal. Appl. 13 (1966), 337-347. [21 General dynamical systems and conditional stability, Proc. Cambridge Philos. SOL.53 (1967), 199-207. [3] Lyapunov functions defined over sets, to appear.
KAYANDE,
KISYNSKI, J. [ I ] Sur l’existence et l’unicitk des solutions des problkmes classiques rklatifs A 1’6quation s = F ( x , y , z , p , q ) , Ann. Univ. Mariue Curie-Sklodowska. Sect. A 11 (19S7), 73-1 12. [21 Sur les Cquations diffkrentielles dans les espaces de Banach, BdZ. Acad. Poton. Sci. 7 (1959), 381-385. [3] Sur a convergence des approximations successives pour l’bquation a2z/ax ay = f (x,y , Z , a ~ j a x azjay), , Ann. of. Math. 7 (l960), 233-240.
r-I. KOMATSU, [ I ] Abstract analyticity in time and unique continuation property of solutions of a parabolic equation, J . Fac. Sci. Univ. Tokyo 9 (1961), 1-1 1. M. A. KHASNOSELSKII, [I] An alternative principle for the existence of periodic solutions for differential equations with retarded argument, Dokl. Akad. Nauk S S R 152 (1963), 801-804. [2] “Positivc Solutions of Operator Equations,” Noordhoff, (1964). M. A., K H R I N , s. G., A N D SOBOLEVSKI, P. E. [l] On differential equations with unbounded operators in Banach spaces, Dokl. Akad. N[zrrk SSSR 1 1 1 (1956), 19-22. [2] On differential equations with unbounded operators in a Hilbert space, Dokl. Akad. Nrilrk SSSR 112 (1957), 990-993.
KRASNOSELSKII,
KRRSOVSKII, N. N. [ I ] On the application of Lyapunov’s second method to equations with time lags, Prikl. Mat. Meh. 20 (1956), 31 5-327. [2] On the asymptotic stability of systems with after effects, Prikl. Mat. Meh. 20 (1956), 513-518.
BIBLIOGRAPHY
301
[3] On periodical solution of differential equations involving a time lag, Dokl. Akad. Nauk S S S R 114 (1957), 252-255. [4] T h e stability of quasilinear systems with after effects, Dokl. Akad. Nauk SSSR 119 (1958), 435-438. [5] Some problems in the theory of stability of motion, Goz. Izd. Fix.-Mat. Lit., Moscow, 1959. English transl. Stanford Univ. Press, Stanford, California, 1963.
KREIN, S. G., AND PROZOROVSKAYA, D. I. [ 11 Analytic semigroups and incorrect problems for evolutionary equations. Dokl. Akad. Nauk. S S S R . ( N . S.) 133 (1960), 277-280. KRZYZANSKI, M. ti] Sur les solutions des tquations du type parabolique dttermintes dans une rtgion illimitte, Bull. Amer. Math. SOC.47 (1941), 91 1-915. [2] Sur les solutions de I’kquation IinCaire du type parabolique dtterminkes par les conditions initiales, Ann. SOC. Polon. Math. 18 (1945), 145-156. [3] Note compltmentaire, Ann. SOC.Polon. Math. 20 (1947), 7-9. [4] Sur les solutions de I’kquation lintaire du type parabolique dhermintes par les conditions initiales, Ann. SOC.Polon. Math. 20 (1948), 7-9. [5] Sur I’allure asymptotique des solutions d’kquation du type parabolique, Bull. Acad. Polon. Sci. CI. 111 4 (1956), 247-251. [6] Evaluations des solutions de I’Cquation aux dtrivtes partielles du type parabolique, dkterminkes dans un domaine non bornt, Ann. Polon. Math. 4 (1957), 93-97. [7] Recherches concernant I’allure des solutions de l’tquation dy type parabolique lorsque la variable du temps tend vers l’infini, Atti Accad. Naz. Lincei. Rend. CE. Sci. Fis. M a t . Natur. 23 (1957a), 28-32. [8] Certaines inkgalitts relatives aux solutions de I’Cquation parabolique lintaire normal, Bull. Acad. Polon. Sci. S i r . Sci. Math. Astronom. Phys. 7 (1959), 131-135. [9] Sur l’unicitk des solutions des second et troisieme probkmes de Fourier relatifs a l’tquation linkaire normale du jpe parabolique, Ann. Polon. Math. 7 (1960), 201-208. [lo] Une proprittt de solution de l’kqbation lintaire du type parabolique ? coefficients i non born&, Ann. Polon. Math. 12 ( I 962), 209-212. LADYZHENSKAJA, 0. A. [I] On the uniqueness of solutions of the Cauchy problem for linear parabolic equations, M a t . Sb. 27 (1950), 175-184. LAKSHMIKANTHAM, V. [l] Lyapunov function and a basic inequality in delay-differential eq iations, Arch. Rational Mech. Anal. 10 (1962), 305-310. [2] On the uniqueness and boundedness of solutions of hyperbolic differential equations, Proc. Cambridge Philos. SOC.58 (1962), 583-587. [3] Differential equations in Banach spaces and extensions of Lyapunov’s method, Proc. Cambridge Philos. SOC.59 (1963), 373-381. [4] Properties of solutions of abstract differential inequalities, Proc. Londo: Math. SOC.14 (1964), 74-82.
302
BIBLIOGRAPHY
[ 5 ] Parabolic differential equations and Lyapunov-like functions, J. Math. Anal. Appl. 9 (1964), 234-251. [6] Functional-differential systems and extensions of Lyapunov’s method, J. Math. Anal. Appl. 8 (1964), 392-405. [7] Some results in functional differential equations, Proc. Nut. Acad. Sci. Zndia 34 (1964), 299-306.
LAKSHMIKANTHAM, V., AND LEELA,S. [I] Parabolic differential equations and conditional stability, Rend. Circ. Mat. Palermo 15 (1966), 179-192. [ 2 ] Conditional stability of conditional invariant sets and functional differential equations, Rend. Circ. Mat. Palermo 15 (1966), 251-272. [3] Almost periodic systems and differential inequalities, J. Zndian Znst. Tech. 1 ( I 967), 286-301. LAKSHMIKANTHAM, V., AND SHENDGE, G. R. [l] Functional differential inequalities, An. Acad. B r a d Ci.39 (1967), 7-14. LAX,P. D., AND MILGRAM, A. [I] Parabolic equations, “Contributions to the Theory of Partial Differential Equations.” Pp. 167-190, Princeton Univ. Press, Princeton, New Jersey, 1954. LEES,M. (see Cohen, P.) [l] Energy inequalities for the solution of differential equations, Trans. Amer. Math. SOC.94 (1960), 58-73.
M. H. LEES,M., AND PROTTER, [ I ] Unique continuation for parabolic differential equations and inequalities, Duke Math. J. 28 (1961), 369-382. LEVIN,J. J., AND NOHEL,J. A. [l] On a nonlinear delay equations, J. Math. Anal. Appl. 8 (1964), 31-44.
L. H. LIBERMAN, [I] Stability of solutions of differential-operator equations with retarded times under perturbations bounded in the mean, Sibirsk. Mat. Z. 4 (1963), 138-144. LIONS,J. L. [l] “Equations Diffhrentielles. OpCrationelles et Probltmes aux Limites.” Springer, New York, 1961.
LIONS, J . L., AND MALGRANGE, B. [I] Sur I’unicitC rhtrogarde dans les probltmes mixtes paraboliques, Math. Scand. 8 (1960), 277-286.
LYWBIC, Yu. I. [l] Conditions for the uniqueness of the solution to Cauchy’s abstract problem, Dokl. Akad. Nauk. S S S R (N.S.) 130 (1960), 969-972.
BIBLIOGRAPHY
303
MAMEDOV, JA. D. [l] One-sided estimates in examination of solutions to the Cauchy limit problem for differential equations with unbounded operators, Sibirsk Mat. 2. 7 (1966),
31 3-317. [2] One-sided estimates in the conditions for asymptotic stability of solutions of differential equations involving unbounded operators, Dokl. Akad. Nauk SSR, 166 (1966), 533-535. MAPLE,C. G., ANDPETERSON, L. G. [l] Stability of solutions of nonlinear diffusion problems, J. Math. Anal. Appl. 14
(1966), 221-241.
MASSERA, J. L., AND SCHAFFER, J. J. [I] “Linear Differential Equations and Functions Spaces.” Academic Press, New
York, 1966.
MCNABB,A. [l] Notes on criteria for the stability of steady state solutions of parabolic equations, J. Math. Anal. Appl. 4 (1962), 193-201. MILICER-GRUZEWSKA, H. [I] Le theorkme d’unicitk de la solution du systkme parabolique des equations lintaires avec les coefficients holdkriens, Bull. Acad. Polon. Sci. Sir. Sci. Math. Astronom. Phys. 7 (1959), 593-599. [2] Le second thCoreme d’unicitk de solution d’un systkme parabolique d’kquations linkaires avec les coefficients holdkriens, Bull. Acad. Polon. Sci. Sir. Sci. Math. Astronom. Phys. 7 (1959), 719-720. MILLER,R. K. [l] Asymptotic behavior of nonlinear delay-differential equations, J. Differential Equations 1 (1965), 293-305. MINORSKY, N. [l] “On Certain Applications of Difference-Differential Equations.” Stanford Univ. Press, Stanford, California, 1948. MINTY,G. J. [I] Monotone (non-linear) operators in Hilbert space, Duke Math. J. 29 (1962),
341-346. [2] On a “monotonicity” method for the solution of non-linear equations in Banach spaces, Proc. Nat. Acad. Sci. U.S.A. 50 (1963), 1038-1041.
MITRYAKOV, A. P. [I] On periodic solutions of the nonlinear hyperbolic equation, Trudy Znst. Mat. Meh. Akad. Nauk Uzbek. SSR 7 (1 949), 137-149. MIZOHATA, S. [I] Le probleme de Cauchy pour les Cquations paraboliques, J . Math. SOC.japan 8
(1956), 269-299.
304
BIBLIOGRAPHY
[2] HypoellipticitC des Cquations parabolique, Bull. SOC.Math. France 85 (1957), 15-49. [3] Le problizme de Cauchy le passe pour quelques equations paraboliques, Proc. Japan Acad. 34 (1958), 693-696. MLAK,W. [l] The epidermic effect for partial differential inequalities of the first order, Ann. Polon. Muth. 3 (1956), 157-164. [2] Remarks on the stability problem for parabolic equations, Ann. Polon. Math. 3 (1957), 343-348. [3] Differential inequalities of parabolic type, Ann. Polon. Math. 3 (1957), 349-354. [4] Differential inequalities in linear spaces, Ann. Polon. Math. 5 (1958), 95-101. [5] A note on non-local existence of solutions of ordinary differential equations, Ann. Polon. Muth. 4 (1958), 344-347. [6] Limitation of solutions of parabolic equations, Ann. Polon. Math. 5(1958/59), 237-245. [ 7 ] The first boundary-value problem for a non-linear parabolic equation, Ann. Polon. Math. 5 (1958/59), 257-262. [8] Limitations and dependence on parameter of solutions of non-stationary differential operator equations, Ann. Polon. Math. 6 (1959), 305-322. [9] On a linear differential inequality of parabolic type, Bull. Acad. Polon. Sci. Sir. Sci.Muth. Astronom. Phys. 7 (1959), 653-656. [lo] Differential inequalities with unbounded operators in Banach spaces, Ann. Math. Pol. 9(1960), 101-111. [l 11 Parabolic differential inequalities and Chaplighin’s method, Ann. Polon. Math. 8 (1960), 139-153. [ 121 Integration of differential equation with unbounded operators in abstract (L)-spaces, Bid. Acad. Polon. Sci. Sir, Sci. Math. Astronom. Phys. 8 (1960), 163-168. [13] Note on abstract differential inequalities and Chaplighin method, Ann. Polon. Math. 10 (1961), 253-271. 1141 A note on approximation of solutions of abstract differential equations, Ann. PO^. Math. 10 (1961), 273-278. [15] Note on abstract linear differential inequalities, Rev. Math. Pures Appl. Acad. Rep. Pop. Roumaine 6 (1961), 655-657. [16] Estimates of solutions of hyperbolic systems of differential equations in two independent variables, Ann. Polon. Math. 12 (1962), 191-197. [17] An example of the equation ut = u,, f (x, t , u) with distinct maximum and minimum solutions of a mixed problem, Ann. Polon. Math. 13 (1963), 101-103. [ 181 Integration of linear differential inequalities with distributions, Contr. to Dzfferential Equations 2 (1963), 265-268.
+
MURAKAMI, H. [l] On non-linear partial differential equations of parabolic types, Proc. Japan Acad. 3 (1957), 530-535, 616-627. [2] Semi-linear partial differential equations of parabolic type, Funkcial. Ekvac. 3 (1960), 1-50.
BIBLIOGRAPHY
305
[3] On non-linear ordinary and evolution equations, Funkcial. Ekvac. 9 (1966), 151-162.
MYSHKIS, A. D. (see El’sgol’ts, L. E., and Shimanov, S. N.) [l] Hystero-differential equations, Uspehi Mat. Nauk 4 (1949), 190-193. [2] General theory of differential equations with a retarded argument, Uspehi Mat. Nauk 4 (1949), 99-141. [3] Supplementary bibliographical material to the paper “General theory of differential equations with retarded argument,” Uspehi Mat. Nauk. 5 (1950), 148-154. [4] Linear homogeneous differential equations of the first order with retarded argument, Uspehi Mat. Nauk 5 (1950), 160-162. [5] O n the solutions of linear homogeneous differential equations of the first order of unstable type with a retarded argument, Dokl. Akad. Nauk S S S R 70 (1950), 953-956. [6] “Linear Differential Equations with Delay in the Argument.” Gostehizdat, 1951. [7] A differential functional inequality, Uspehi Mat. Nauk 15 (1960), 157-161.
M., AND SIMODA, S. NAGUMO, [I] Note sur I’inCgalitC diff6rentielle concernant les Cquations du type parabolique, Proc. Japan Acad. 27 (1951), 536-539. NARASIMHAN, R. [l] On the asymptotic stability of solutions of parabolic differential equations, J. Rational. Mech. Anal. 3 (1954), 303-313. NICKEL,K. [l] Fehlerabschatzungen bei parabolischen Differentialgleichungen, Math. 2. 71 (1959), 268-282. [2] Parabolic equations with applications to boundary layer theory. “Partial Differential Equations and Continuum Mechanics,” Univ. of Wisconsin Press, Madison, Wisconsin, 1961. [3] Gestaltaussagen uber Losungen parabolischer Differentialgleichungen, J. Reine Angew. Math. 211 (1962), 78-94.
L. (see Agmon, S.) NIRENBERG, [l] A strong maximum principle for parabolic equations, Comm. Pure Appl. Math. 6 (1953), 167-1 77. [2] Existence theorems in partial differential equations, New York Univ. Notes. NOHEL,J. A. (see Ergen, W. K., and Lipkin, H. J . ; Levin, J. J.) [l] A class of nonlinear delay differential equations, J. Math. Phys. 38 (1960), 295-3 1 1. OEUZTORELI, N. N. [l] “Time-lag Control System.” Academic Press, New York, 1966. 121 On the existence and uniqueness of solutions of an ordinary differential equation in the case of Banach space, Bull. Acad. Polon. Sci. Sir. Sci. Math. Astronom. Phys. 8 (1960), 667-673.
306
BIBLIOGRAPHY
ONUCHIC, N. [I] On the asymptotic behavior of the solutions of functional-differential equations, Proc. Irttertz. Symp. Dfferential Eqs. and Dynamical S y s t e m , Puerto Rim, 1965, pp. 223-233. [2] On the uniform stability of a perturbed linear functional-differential equation, Proc. Amer. Math. Sac. 19, (1968), 528-532. PALCZEWSKI, B. [I] On the uniqueness of solutions and the convergence of successive approximations in the Darboux problem under the conditions of the Krasnosielski and Krein type, Ann. Polon. Math. 14 ( I 964), 183-1 90. PALCZEWSKI, B., AND PAWELSKI, W. [I] Some remarks on the uniqueness of solutions of the Darboux problem with conditions of the Krasnosielski-Krein type,” Ann. Polon. Math. 14 (1964), 97-100. PELCZAR, A. [I] On the existence and uniqueness of solutions of the Darboux problem for the equation z,, = f ( x , y , z , z , , zv),Bull. Acad. Polon. Sci. Sir. Sci. Math. Astronom Phvs. 12 (1964), 703-707. [2] On the existence and the uniqueness of solutions of certain initial-boundary problems for the equation z,, = f ( x , y , z , z , , zy),Bull. Acad. Polon. Sci. S ~ Y , Sci. Math. Astronom. Phys. 12 (1964), 709-714. [3] On a modification of the method of Euler polygons for the ordinary differential equation, Ann. Polon. Math. 15 (1964), 195-202. [4] O n the convergence to zero of oscillating solutions of some partial differential equations, Zeszyty Nazdk. Uniw. Jagiello. Prace M a t . I1 (1966), 49-51. [5] O n the method of successive approximations for some operator equations with applications to partial differential hyperbolic equations, Zeszyty Nauk. Uniw. Ja+llo. Prnce M a t . 1 I ( I 966), 59-68. PEHELLO, C. (see Hale, J. K.) [ I ] A note o r 1 periodic solutions of nonlinear differential equations with time lags, Proc. Intern. Symp. Differential Eqs. and Dynamical Systems, Puerto Rico, 1965, pp. 185-187. PHILIPS, K. S. [I1 Dissipative operators and hyperbolic systems of partial differential equations, Trans. Amrr. Math. Sac. 90 (l95Y), 193-254.
M. [ I ] “Appunti di Analisi Superiore.” Napoli, 1941.
PICONE,
PINI,B. [l] Un problema di valori a1 contorno, generalizzato per I’equazione a derivate parziali lineare parabolice del second0 ordine, Riv. M a t . Parma 3 (1952), 153-187. [2] Maggioranti c: minoranti delle solzioni delle equazioni paraboliche, Ann. M a t . Pura Appl. 31 (1954), 249-264.
BIBLIOGRAPHY
307
[3] Sur primo problema di valori a1 contorno per l’equazione parabolica non lineare del second0 ordine, Rend. Sem. Mat. Univ. Padova 27 (1957), 149-161.
PINNEY, E. [I] “Differential-Difference Equations.” Univ. of California Press, Berkeley, California, 1958.
PLIS,A. [l] The problem of uniqueness for the solution of a system of partial differential equations, Bull. Acad. Polon. Sci. Cl. III 2 (1954), 55-57. [2] On the uniqueness of the non-negative solution of the homogeneous Cauchy problem for a system of partial differential equations, Ann. Polon. Math. 2 (1955), 314-318. [3] On the estimation of the existence domain for solutions of a nonlinear partial differential equation of the first order, Bull. Acad. Polon. Sci. Cl. III 4 (1956), 3, 125. [4] Generalization of the Cauchyproblem for a system of partial differential equations, Bull. Acad. Polon. Sci. Cf.III 4 (1956), 741-744. [S] On the uniqueness of the non-negative solution of the homogeneous mixed problem for a system of partial differential equation, Ann. Polon. Math. 7 (1960), 255-258. [6] Remark on a partial differential inequality of the first order, Ann. Polon. Math. 16 (1965), 371-375.
POGORZELSKI, W. [l] fitude de la solution fondamentale de I’kquation parabolique, Ricerche Mat. 5 (1956), 25-57. [2] Propriktks des intkgrales gknkralisCes de Poisson Weierstrass et problkme de Cauchy pour un systeme parabolique, Ann. Ec. Norm. 76 (1959), 125-149. PRODI,G. [l] Questioni di stabilita per equazioni non lineari alle derivate parziali di tipo parabolico, Atti Accad. Naz. Lincei. Rend. Cl. Sci. Fis. Mat. Natur. 10 (1951), 365-370. [2] Teoremi di esistenza per equazioni alle derivate parziali non lineari di tipo parabolico. I. 11, Rend. Inst. Lombard0 (1953), 1-47. [3] Soluzioni periodiche di equazioni alle derivate parziali di tipo iperbolico nonlineari, Ann. Mat. Pura Appl. 42 (1 956), 2549.
PROTTER, M. H. (see Lees, M.) [ I ] Asymptotic behavior and uniqueness theorems for hyperbolic equations and inequalities, Univ. of California, Berkeley, Tech. Rep. No. 9, March, 1960. [2] Properties of solutions of parabolic equations and inequalities, Canad. J . Math. 13 (1961), 331-345.
RAO,M., AND TSOKOS,C. P. RAMAMOHAN [l] On the stability and boundedness of differential systems in Banach spaces. Proc. Cambridge Philos. SOC.65 (1969), 507-512.
308
BIBLIOGRAPHY
RAMAMOHAN RAO,M., AND VISWANATHAM, B. [ I ] On the asymptotic behaviour of solutions of partial differential equations of hyperbolic type. Proc. Nut. Acad. Sci. (India), 33 (1963), 279-284. RAZUMIKHIN, B. S. [ I ] Stability of motion, Prikl. Mat. Meh. 20 (1956), 266-270. [2] Stability of systems with lag, Prikl. Mat. Meh. 20 (1956), 503-512. [3] Estimates of solutions of a system of differential equations of the perturbed motion with variable coefficients, Prikl. Mat. Meh. 21 (1957), 119-120. [4] Stability in the first approximation of systems with lag, Prikl. Mat. Meh. 22 (1958), 155-1 66. [ 5 ] O n the application of Liapunov’s method to stability problems, Prikl. Mat. Meh. 22 (1958), 338-349. [6] Application of Lyapunov’s method to problems in the stability of systems with a delay, Automat. i Telemeh. 21 (1960), 740-748.
Z . I. REKLISHKII, [l] Stability of solutions of certain linear differential equations with a lagging argument in Banach space, Dokl. Akad. Nauk SSSR 11 1 (1956), 29-32. [2] Necessary and suficient conditions for stability of solutions of certain systems of linear differential equations with lagging argument in Banach space, Proc. 3rd All Union Math. Congr. 2 (1956), 128-129. [3] Tests for boundedness of solutions of linear differential equations with variable lag of the argument, Dokl. Akad. Nauk SSSR 118 (1958), 447-449. [4] Tests for boundedness of solutions of linear differential equations with several lags of the argument, Dokl. Akad. Nauk SSSR 125 (1959), 46-47. [ 5 ] Tests for boundedness of solutions of differential equations with continuous lag of the argument in a Banach space, Dokl. Akad. Nauk SSSR 127 (1959), 971-974. P. SANTORO, [I] Sul problema di Darboux per l’equazione s = f ( x , y , z , p , q) e il fenomeno di Peano, Rend. Accad. Naz. X L 10 (1959), 3-17.
N. V. SARKOVA, [I] Approximate solution of differential equations with lagging argument, (Coll. of papers on oscillating functions.) U h . Zap. Perm. Univ. 13 (1959), 63-73. SATO,
‘r.
[l] S u r Yequation aux dCrivCes partielles LIZ = f ( x , t , z , p , q), Compositio. Math. 12 (1954). 157-177; and II,Ibid. 14 (1959), 151-171.
SEIFERT, G. [ I ] A condition for almost periodicity with some applications to functional-differentiai equations, /. Differential Equations 1 ( 1 965), 393-408.
J. SERRIN, [I] Local behavior of solutions of quasilinear equations, Acta Math. 11 1 (1964), 247-302.
BIBLIOGRAPHY
SHANAHAN, J. P.
[I] O n uniqueness questions for hyperbolic differential equations, Pacific 10 (1960), 677-688.
309
I. Math.
SHIMANOV, S. N. (see El’sgol’ts, L. E., and Myshkis, A. D.) [I] Almost periodic solutions of non-homogeneous linear differential equations with lag, Izv. Vyss. Utebn. Zaved. Mat. 4 (1958), 270-214. [2] Almost periodic oscillations in nonlinear systems with lag. Uokl. Akad. Nauk S S S R 125 (1959), 1203-1206. [3] O n the theory of oscillations of quasilinear systems with delay, Prikl. Mat. Meh. 23 (195Y), 836-844. [4] O n the instability of the motion of systems with retardation, Prikl. Mat. Meh. 24 (1960), 55-63. [5] O n stability in the critical case of a zero root with a time lag, Prikl. Mat. Meh. 24 (1960), 447-457. [6] Almost periodic oscillations of quasilinear systems with time lag in degenerate case, Dokl. Akad. Nauk SSSR 133 (IY60), 36-39. [7] Oscillations of quasilinear autonomous systems with time lag, Izv. VysS. UEebn. Zaved. Radiofizika, 3 (1960), 456-466. [8] O n stability of quasiharmonic systems with retardation, Prikl. Mat. Meh. 25 (1961), 992-1002. [9] O n the theory of linear differential equations with periodic coefficients and time lag, Prikl. Mat. Meh. 27 (1963), 450-458. [lo] On the theory of linear differential equations with retardations, Differencial’nye Uravnenija 1 (1965), 102-116.
.
SLOBODETSKI, L. N. [I]O n the fundamental solution and the Cauchy problem for parabolic systems, Math. Sb. 46 (1958), 229-258. SMIRNOVA, G. N.
[I] O n uniqueness classes for the solution of Cauchy’s problem for parabolic equations, Dokl. Akad. Nauk SSSR 153 ( I Y63), 1269-1 272. SOBOLEVSKI, P. E. (see Krasnoselskii, M. A., and Krein, S. G.) [I] Approximate methods of solving differential equations in Banach space, Dokl. Akad. Nauk S S S R I I5 ( 1 957), 240-243. [2] O n equations of parabolic type in Banach space, Trudy Moscov. Mat. ObSt. 10 (1961), 298-350. SOLOV’EV, P. V. [I] Some remarks on periodic solutions of the nonlinear equation of hyperbolic type, Izzv. Akad. Nauk S S S R Ser. Mat. 2 (1939), 15&164. STOKES, A. P. [l] A Floquet theory for functional-differential equations, Proc. Nut. Acad. Sci. U . S . A .48 (1962), 1330-1334. [2] O n the stability of a limit cycle of an autonomous functional-differential equations, Contr. D%ferentiaIEquations 3 (1964), 121-140.
310
BIBLIOGRAPHY
SUGIYAMA, S.
[I] On the existence and uniqueness theorems of difference-differential equations, Kcidai Math. Sen. Rep. 12 (1960), 179-190. 121 A note on difference-differential equations with a parameter, Mem. School Sci. Engrg. Wuseda Univ. Tokyo 25 (1961), 23-36.
[3] Existence theorems on difference-differential equations, Proc. Japan Acad. 38 (1962), 145-149. [4] Dependence properties of solutions on retardation and initial values in the theory of difference-differentia1 equations, Ksdai Math. Sem. Rep. 15 (1963), 67-78. [5] On the theory of difference-differential equations, I, Bull. Sci. Engrg. Res. Lab. Waseda Univ. 26 (1964), 97-111. [6] On the theory of difference-differential equations, 11, Bull. Sci. Engrg. Res. Lab. Wuseda Univ. 27 (1964), 74-84. [7] On the theory of difference-differential equations, 111. Applications to Lyapunov functions, Mem. School Sci. Engrg. Waseda Univ. 28 (1964), 73-85. [8] On a certain functional differential inequality, Kcidai Math. Sem. Rep. 17 (1965), 273-280. [9] On a certain functional differential equation, Proc. Japan Acad. 41 (1965), 360-362.
SZARSKI, J. [l] Sur certains systhnes d’inCgalitCs diffkrentielles aux dkrivkes partielles du premier ordre, Ann. SOC.Pol. Math. 21 (1948), 7-26. [2] Sur certaines inCgalitts entre les intCgrales des equations diffkrentielles aux dCrivkes partielles du premier ordre, Ann. SOC.Pol. Math. 22 (1949), 1-34. [3] Systttmes d’inkgalitks diffkrentielles aux dkrivCes partielles du premier ordre, et leurs applications, Ann. Polon. Math. 1 (1954), 149-165. 141 Sur la limitation et 1’unicitC des solutions d’un systttmi: non-linkaire d’kquations paraboliques aux dkrivkes partielles du second ordre, Ann. Pol. Math. 2 (1955), 237-249. [5] S u r la limitation et I’unicitC des solutions des problttmes de Fourier pour un systttme non-linbaire d’kquations paraboliques, Ann. Pol. Math. 6 (1959), 21 1-216. 161 “Characteristics and Cauchy Problem for Non-linear Partial Differential Equations of the First Order.” Univ. of Kansas, Lawrence, Kansas, 1959. [7] Sur un systttme non-linkaire d’inkgalitks diffkrentielles paraboliques, Ann. Pol. Math. 15 (1964), 15-22. [8] “Differential Inequalities.” PWN, Polish Sci. Publ., Warsaw, 1965.
SZMYDT, 2. [l] S u r un nouveau type de problttmes pour un systttme d’kquations diffkrentielles hyperboliques du second ordre a deux variables indkpendantes, Bull. Acad. Polon. Sci. Cl. IIZ 4 (1956), 67-72. [2] Sur une gknkralisation des problttmes classiques concernant un sysdme d’Cquations diffkrentielles hyperboliques du second ordre a deux variables indCpendantes, Bull. Acad. Polon. Sci. CI. ZIZ 4 (1956), 579-584.
BIBLIOGRAPHY
31 1
[3] Sur l’existence de solutions de certains nouveaux problbmes pour un systbme d’kquations diffkrentielles hyperboliques du second ordre a deux variables indtpendantes, Ann. Polon. Math. 4 (1957), 40-60. [4] Sur I’existence d’une solution unique de certains problbmes pour un systbme d’kquations diffkrentielles hyperboliques du second ordre a deux variables indkpendantes, Ann. Polon. Math. 4 (1958), 165-182. [5] Sur l’existence de solutions de certains problbmes aux limites relatifs 2 un systbme d’tquations diffkrentielles hyperboliques, Bull. Acad. Polon. Sci. Sdr. Sci. Math. Astronom. Phys. 6 (1958), 31-36.
TAAM, C. T., AND WELCH,J. N. [l] Compact solutions of non-linear differential equations in Banach spaces, Michigan Math. J. 13 (1966), 271-284.
TANABE, H. (see Kato, T.) [I] On the equations of evolution in a Banach space, Math. J. Osaka 12 (1960), 363-376. [2] Convergence to a stationary state of the solution of some kind of differential equations in a Banach space, Proc. Japan Acad. 37 (1961), 127-130.
TICHONOV, A. N. [I] Functional equations of Volterra type and their applications to some problems of mathematical physics, Byull. Moskov. Univ., Sect. A 1 ( 8 ) (1938), 1-25. TURSKI, S. [l] On certain generalization of theorems on the existence and uniqueness of integrals of hyperbolic equations of type aaz/ax ay = f (x, y , z, azji3.x , az/ay), Dodatek do Roczn. Pol. Tow. Mat. Krakow. 1935. VEJVODA, 0. [l] The stability of a system of differential equations in the complex domain, Czechoslovak Math. J. 7 (1957), 137-159. VISHIK,M. I. [ 11 Solvability of boundary problems for quasi-linear parabolic equations of higher order, Math. Sb. 59 (1962), 289-335. VOLKOV, D. M. [I] An analogue of the second method of Liapunov for nonlinear boundary value problems of hyperbolic equations, Leningrad Gos. Univ. UEen. Zap. Ser. Math. Nauk 33 (1958), 90-96. WALTER, W. [l] Uber die Euler-Poisson-Darboux-Gleichung, Math. Z. 67 (1967), 361-376. [2] Uber die Differentialgleichung uxy= f ( x , y , u, u., u,). I, Math. Z . 71 (1959), 308-324. [3] Uber die Differentialgleichung uxy= f (x, y , u, u, , u,). 11, Math. Z . 71 (1959), 436-453.
312
BIBLIOGRAPHY
[4] Uber die Differentialgleichung uZy= f ( x , y , u , u, , uv). 111, Math. Z . 73 (1960), 268-279. [ 5 ] Eindeutigkeitssatze f u r gewohnliche, parabolische und hyperbolische Differentialgleichungen, Math. Z . 74 (1960), 191-208. [6] Fehlerabschatzungen bei hyperbolischen Differentialgleichungen, Arch. Rational Mech. Anal. 7 (1961), 249-272. [7] Fehlerabschatzungen und Eindeutigkeitssatze f u r gewohnliche und partielle Differentialgleichungen, Z. Angew. Math. Mech. 42 (1962), T49-T62.
[8] “Differential und Integral Ungleichungen.” Springer, Berlin, 1964. WAZEWSKI, T [I] Sur l’apprkciation d u domaine d‘existence des integrales de l’bquation aux dkrivkes partielles d u premier ordre, Ann. SOC.Polon. Math. 14 (19361, 149-177. [2] S u r le problkme de Cauchy relatif B un systkme d’bquations aux derivbes partielles, Ann. SOC. Polon. Math. 15 (1936), 101. [3] Ueber die Bedingungen des Existenz der lntegrale partieller Differentialgleichungen erster Ordnung, Math. 2. 43 (1948), 522-532. [4] Une genbralisation des thborkmes sur les accroissements finis au cas des espaces de Banach et applications B la genkralisation d u thCorkme de l’Hospita1, Ann. Sac. Polon. Math. 24 (1953), 132-147.
WEND,I). V. V [I] On the zeros solutions of some linear complex differential equations, Pacific J.
Math. 10 (l960), 71 3-722. [2] Singularity-free regions for solutions of non-linear complex differential equations, Proc. Amer. Math. Sac. 18 (1967), 592-597.
WESTFKAI., H. [ 11 Zur Abschatzung der Losungen nichtlinearer parabolischer Differentialgleichungen, Math. Z. 51 (1949), 69e695.
[ 11 On the convergence of successive approximation in the Darboux problem, Ann. Polon. Math. 17 (1966), 329-336.
[2] Remarks on the uniqueness theorem of solutions of the Darboux problem, Cunad. Math. BILZI.,8 (1965), 791-796.
WmGwr, E. M.
[ I ] The linear difference-differential equation with asymptotically constant coefficients, Amer. J . Math. 70 (1948), 221-238. [2] T h e linear difference-differentia1 equation with constant coefficients, Proc. Roy. Soc. Edinbzrrgh Sect. A 62 (1949). 387-393. [3] T h e stability of solutions of nonlinear difference-differential equations, Proc. Roy. Soc. Edinbnrgh Sect. A 63 ( I 950). 18-26.
[4] A nonlinear difference-differential equation, J . Reine Angew. Math. 194 (1955), 66-87.
BIBLIOGRAPHY
313
YOSHIZAWA, T.
[I] Extreme stability and almost periodic solutions of functional-differential equa-
tions, Arch. Rational, Mech. Anal. 17 (1964), 148-170. [2] Asymptotic stability of solution of an almost periodic system of functionaldifferential equations, Rend. Circ. Mat. Palermo [2] 13 (1964), 1-13. [3] The stability theory by Liapunov’s second method, Math. SOC.Japan, Tokyo (1 967).
ZERAGIA, P. K. [I] Solution of a basic boundary problem for a certain system of differential equations of parabolic type, Dokl. Akad. Nauk G S S R 26 (1951), 257-264. [2] Boundary problems for certain nonlinear equations of parabolic type, Trudy Tbil. Math. Inst. Akad. Nauk G S S R 24 (1957), 195-222. ZVERKIN, A. M. (see Els’gol’ts, L. E., Kamenskii, G. A. and Norkin, S.B.) [I] Dependence of stability of solutions of linear differential equations with lagging argument upon the choice of initial moment, Vestnik Moscov. Univ. Ser. Mat. Meh. Astronom. Fiz. Him. 5 (1959), 15-20. [2] On the theory of linear equations with a lagging argument and periodic coefficients, Dokl. Akad. Nauk S S S R 128 (1959), 882-885. [3] General solution of a linear differential equation with perturbed argument, Naucn. Dokl. Vyss. Skoly, Fis. Mat. Nauik 1 (1960), 30-37.
This page intentionally left blank
Author Index
A Agmon, S., 272,289 Alexiewicz, A., 233, 289 Antosiewicz, H. A., 272, 289 Aronson, D. G., 219, 289 Aziz, A. K., 233, 289, 290
D Danskin, J. M., 42, 290 Das, K. M., 288, 293 Deo, S. G., 288, 294 Diaz, J. B., 233, 289, 294 Driver, R. D., 42, 80, 110, 294 E
B Barrar, R. B., 219, 290 Bellman, R., 42, 219, 290 Besala, P., 219, 290 Bhatia, N. P., 291 Bielecki, A., 233, 291 Bogdanowicz, W. M., 291 Browder, F. E., 272, 291 Brzychaczy, S., 219, 291 Bullock, R. M., 42, 291 C Calderon, A. P., 292 Cameron, R. H., 219, 292 Cesari, L., 233, 292 Chandra, J., 272, 292 Chang, H., 292 Chu, S. C., 233, 292 Ciliberto, C., 219, 233, 292 Coffman, C. V., 272, 292 Cohen, P., 272, 293 Conlan, J., 233, 293 Conti, R., 233, 293 Cooke, K. L., 42, 290, 293 Corduneanu, C., 80, 291
Edmunds, D. E., 272, 294 Eidelman, S. D., 219, 294 El’sgol’ts, L. E., 42, 110, 294, 295
F Fleishman, B. A., 272, 292 Fodcuk, V. I., 42, 295 Foias, C., 219, 272, 295 Franklin, J., 42, 295 Friedman, A,, 219, 295
G Giuliano, L., 296 Glicksberg, I., 290 Guglielmino, F., 233, 296 Gussi, G., 295
H Hahn, W., 42, 297 Halanay, A., 42, 80, 110, 297 Hale, J. K., 42, 80, 233, 297 Hartman, P., 272, 297 Hastings, S. P., 42, 297 315
316
AUTHOR INDEX
I Ilin, A. M., 219, 298 Ito, S., 219, 299
J Jones, G. S., 42, 299
K Kakutani, S., 42, 299 Kalashnikov, A. S., 219, 298 Kamenskii, G. A., 42, 295, 299 Kaplan, S., 219, 299 Karasik, G. J., 42, 299 Kato, J., 80, 299 Kato, T., 272, 300 Kayande, A. A,, 288, 300 Kisynski, J., 233, 272, 300 Komatsu, H., 219, 300 Krasnosel’skii, M. A., 272, 300 Krasovskii, N. N., 42, 80, 110, 300 Krein, S. G., 272, 300, 301 Krzyzanski, M., 219, 301 L Ladyzhenskaja, 0. A , , 219, 301 Lakshmikantham, V., 42, 80, 110, 219, 233, 272, 288, 294, 300, 301, 302 Lax, P. D., 219, 302 Leela, S., 80, 110, 219, 302 Lees, M., 219, 272, 302 Levin, J. J., 302 Liberman, L. H., 80, 302 Lions, J. L., 219, 272, 302 Lyubic, Yu. I., 272, 302 M McNabb, A., 219, 303 Malgrangc, B., 219, 302 Maloney, J. P., 233, 290 Mamedov, Ja. D., 272, 303 Maple, C. G., 219, 303 Markus, L., 42, 299 Massera, J. L., 272, 303 Milgram, A,, 219, 302
Milicer-Gruzewska, H., 219, 303 Miller, R. K., 80, 303 Minorsky, N., 303 Minty, G. J., 272, 303 Mitryakov, A. P., 303 Mizohata, S., 219, 303 Mlak, W., 219, 272, 304 Murakami, H., 219, 272, 304 Myshkis, A. D., 42, 295, 305
N Nagumo, M., 219, 305 Narasimhan, R., 219, 305 Nickel, K., 219, 305 Nirenberg, L., 219, 272, 289, 305 Nohel, J. A., 302, 305 Norkin, S. G., 42, 295
0 Ogustoreli, N. N., 42, 110, 305 Olenik, 0. A., 219, 298 Onuchic, N., 42, 306 Orlicz, W., 233, 289 P Palczewski, B., 233, 306 Pawelski, W., 233, 306 Pelczar, A., 233, 306 Perello, C., 42, 298, 306 Peterson, L. G., 219, 303 Phillips, R. S., 233, 306 Picone, M., 219, 306 Pini, B., 219, 306 Pinney, E., 42, 307 Plis, A., 148, 307 Poenaru, V., 295 Pogorzelski, W., 219, 307 Prodi, G., 219, 307 Protter, M. H., 219, 233, 302, 307 Prozorovskaya, D. I., 272, 301 R Ramamohan Rao, M., 272, 307, 308 Razumikhin, B. S., 80, 110, 308 Reklishii, 2. I., 80, 308
317
AUTHOR INDEX
S Santoro, P., 233, 308 Sarkova, N. V., 308 Sato, T., 308 Schaffer, J. J., 272, 303 Seifert, G., 80, 308 Serrin, J., 219, 308 Shanahan, J. P., 233, 309 Shendge, G. R., 42, 302 Shimanov, S. N., 42, 295, 309 Sirnoda, S., 219 Slobodetski, L. N., 219, 309 Smirnova, G. N., 219, 309 Sobolevski, P. E., 272, 300 Solov’ev, P. V., 309 Stokes, A. P., 42, 309 Sugiyama, S., 42, 80, 310 Szarski, J., 148, 219, 233, 272, 310 Szego, G. P., 291 Szmydt, Z., 233, 310
T Taam, C. T., 272, 311 Tanabe, H., 272, 300, 311 Tichonov, A. N., 311 Tsokos, C. P., 272, 307
Turski, S., 31 1 V
Vejvoda, O., 311 Vishik, M. I., 219, 311 Viswanatham, B., 308 Volkov, D. M., 233, 311 W Walter, W., 219, 233, 311 Wazewski, T., 312 Welch, J. N., 272, 291, 311 Wend, D. V. V., 288, 312 Westphal, H., 219, 312 Wong, J. S. W., 233, 312 Wright, E. M., 42, 312
Y Yoshizawa, T., 42, 80, 313 Z Zeragia, P. K., 219, 313 Zverkin, A. M., 42, 313
Subject Index
D
A Almost periodic solutions, 77 Approximate solutions, 9, 38, 134, 170, 231, 255, 275 Asymptotic behaviour, 24, 105 Asymptotic equivalence, 25 Asymptotic in the sense of Wintner, 32 Autonomous systems, 58
Demi-continuity, 240 Differential inequalities functional, 34 hyperbolic, 221 parabolic, 149, 181, 205 partial (of the first order), 113, 136 Diffusion equation, 160 Directional derivative, 151
B
E
Boundary value problem Dirichlet type, 164 exterior, 213 first, 163 first Fourier, 164 mixed, 164 Newmann type, 164 second, 164 Boundedness, 60, 190, 196, 200, 269 Bounds, 13, 127, 163, 229, 284.
Egress points, 29 Ellipticity, 152 Error estimates, 38, 170, 229, 231 Estimate of Time lag, 39, 100 Existence theorems for abstract differential equations, 237, 24 1 for evolution equations, 249 for functional differential equations, 5 G
C Chaplygin’s method, 259 Charactcristic exponent, 264 Comparison theorems, 37, 81, 118, 155, 267, 287 Continuous dependence, 18, 247 Converse theorems, 49 for extreme stability, 67 for generalized exponential stability, 49, 50
Gateaux derivative, 264 Growth damping factor, 207 H Homotopic, 241
I Instability, 61 Invariant sets, 58 318
319
SUBJECT INDEX
L 3’-condition, 205 Lyapunov functions, 81, 267 Lyapunov functionals, 43 Lyapunov-like functions, 144, 186, 287
M Maximal and minimal solutions, 36 existence, 36 Maximum and minimum principles, 159 Mayer’s transformation, 143 Method of averaging, 247 Mild solution, 250 Monotonicity condition, 240
N Nonlinear boundary conditions, 154
P ParaboIicity, 152 Perturbed systems, 23, 62, 97, 197
R Retarded arguments, 3
S Several Lyapunov functions, 110 Singularity-free regions, 279
Stability, 21, 43, 87, 146, 175, 190, 267 asymptotic, 22, 45, 89, 177, 192, 271 conditional, 200 eventual, 101 extreme, 66 generalized exponential, 49 perfect, 74 of steady state solutions, 174 strong, 74 Strict solution, 250 Systems with repulsive forces, 32
T Topological principle, 29
U Under and over functions, 35, 163, 222 Uniqueness criteria, for abstract differential equations, 243, 257 for complex differential equations, 278 for functional differential equations, 11, 39, 72 for hyperbolic differential equations, 223, 225 for nonlinear evolution equations, 257 for parabolic differential equations, 174, 210, 214 for partial differential equations of first order, 136
Mathematics in Science and Engineering A Series of Monographs and Textbooks Edited by RICHARD BELLMAN, University of Southern California 1. T. Y. Thomas. Concepts from Tensor Analysis and Differential Geometry. Second Edition. 1965 2. T. Y. Thomas. Plastic Flow and Fracture in Solids. 1961 3. R. Aris. The Optimal Design of Chemical Reactors: A Study in Dynamic Programming. 1961
4. J. LaSalle and S. Lefschetz. Stability by by Liapunov’s Direct Method with Applications. 1961 5. G. Leitmann (ed.). Optimization Techniques: With Applications to Aerospace Systems. 1962
6. R. Bellman and K. L. Cooke. DifferentialDifference Equations. 1963 7. F. A. Haight. Mathematical Theories of Traffic Flow. 1963
8. F. V. Atkinson. Discrete and Continuous Boundary Problems. 1964 9. A. Jeffrey and T. Taniuti. Non-Linear Wave Propagation: With Applications to Physics and Magnetohydrodynamics. 1964
19. J. Aczel. Lectures on Functional Equations and Their Applications. 1966 20. R. E. Murphy. Adaptive Processes in Economic Systems. 1965 21. S. E. Dreyfus. Dynamic Programming and the Calculus of Variations. 1965 22. A. A. Fel’dbaum. Optimal Control Systems. 1965
23. A. Halanay. Differential Equations: Stability, Oscillations, Time Lags. 1966 24. M. N. Oguztoreli. Time-Lag Control Systems. 1966 25. D. Sworder. Optimal Adaptive Control Systems. 1966 26. M. Ash. Optimal Shutdown Control of Nuclear Reactors. 1966
27. D. N. Chorafas. Control System Functions and Programming Approaches (In Two Volumes). 1966
28. N. P.Erugin. Linear Systems of Ordinary Differential Equations. 1966 29. S. Marcus. Algebraic Linguistics; Analytical Models. 1967
10. J. T. Tou. Optimum Design of Digital Control Systems. 1963.
30. A. M. Liapunov. Stability of Motion. 1966
11. H. Flanders. Differential Forms: With Applications to the Physical Sciences. 1963
31. G. Leitmann (ed.). Topics in Optimization. 1967
12. S. M. Roberts. Dynamic Programming in Chemical Engineering and Process Control. 1964
32. M. Aoki. Optimization of Stochastic Systems. 1967
13. S. Lefschetz. Stability of Nonlinear Control Systems. 1965 14. D. N. Chorafas. Systems and Simulation. 1965 15. A. A. Petvozvanskii. Random Processes in Nonlinear Control Systems. 1965 16. M. C. Pease, Ill. Methods of Matrix Algebra. 1965 17. V. E. Benes. Mathematical Theory of Connecting Networks and Telephone Traffic. 1965
18. W. F. Ames. Nonlinear Partial Differential Equations in Engineering. 1965
33. H. J. Kushner. Stochastic Stability and control. 1967
34. M. Urabe. Nonlinear Autonomous Oscillations. 1967 35. F. Calogero. Variable Phase Approach to Potential Scattering. 1967 36. A. Kaufmann. Graphs, Dynamic Programming, and Finite Games. 1967 37. A. Kaufmann and R. Cruon. Dynamic Programming: Sequential Scientific Management. 1967
38. J. H. Ahlberg, E. N. Nilson, and J. L. Walsh. The Theory of Splines and Their Applications. 1967
39. Y. Sawaragi. Y. Sunahara. and T. Naka-
In preparation
mizo. Statistical Decision Theory in Adaptive Control Systems. 1967
R. Bellman. Methods of Nonlinear Analysis, Volume 1
40. R. Bellman. Introduction t o the Mathematical Theory of Control Processes Volume I. 1967 (Volumes II and Ill in preparation)
41. E. S. Lee. Quasilinearization and Invariant Imbedding. 1968 42. W. Ames. Nonlinear Ordinary Differential Equations in Transport Processes. 1968 43. W. Miller, Jr. Lie Theory and Special Functions. 1968 44. P. B. Bailey, L. F. Shampine, and P. E. Waltman. Nonlinear Two Point Boundary Value Problems. 1968. 45. Iu. P. Petrov. Variational Methods in Optimum Control Theory. 1968 46. 0. A. Ladyzhenskaya and N. N. Ural'tseva. Linear and Quasilinear Elliptic Equations. 1968
47. A. Kaufmann and R. Faure. Introduction to Operations Research. 1968 48. C. A. Swanson. Comparison and Oscillation Theory of Linear Differential Equations.
1968 49. R. Hermann. Differential Geometry and the Calculus of Variations. 1968 50.
N. K.
Jaiswal. Priority Queues. 1968
51. H. Nikaido. Convex Structures and Economic Theory. 1968 52. K. S. Fu. Sequential Methods in Pattern Recognition and Machine Learning. 1968 53. Y. L. Luke. The Special Functions and Their Approximations (In Two Volumes). 1969 54. R. P. Gilbert. Function Theoretic Methods in Partial Differential Equations. 1969 55.. V. Lakshmikantham and S. Leela. Differential and Integral Inequalities (In Two Volumes.) 1969
56. S. H. Herrnes and J. P. LaSalle. Functional Analysis and Time Optimal Control.
1969. 57. M. Iri. Network Flow, Transportation, and Scheduling: Theory and Algorithms. 1969 58. A. Blaquiere. F. Gerard, and G. Leitmann. Quantitative and Qualitative Games.
1969 59. P. Falb and J. De Jong. Successive Approximation Methods in Control and Oscillation Theory. 1969
R. Bellman, K. L. Cooke, and J. A. Lockett. Algorithms, Graphs, and Computers A. H. Jazwinski. Stochastic Processes and Filtering Theory S. R. McReynolds and P. Dyer. The Computation and Theory of Optimal Control
J. M. Mendel and K. S. Fu. Adaptive, Learning, and Pattern Recognition Systems: Theory and Applications
G. Rosen. Formulations of Classical and Quantum Dynamical Theory
E. J. Beltrami. Methods of Nonlinear Analysis and Optimization H. H. Happ. The Theory of Network Diakoptics
This page intentionally left blank