• 検索結果がありません。

We establish the existence and uniqueness of a local solution for the system of differential equations ∂ ∂tu(x, t) =u“ v“Z t 0 u(x, s)ds, t

N/A
N/A
Protected

Academic year: 2022

シェア "We establish the existence and uniqueness of a local solution for the system of differential equations ∂ ∂tu(x, t) =u“ v“Z t 0 u(x, s)ds, t"

Copied!
7
0
0

読み込み中.... (全文を見る)

全文

(1)

ISSN: 1072-6691. URL: http://ejde.math.txstate.edu or http://ejde.math.unt.edu ftp ejde.math.txstate.edu (login: ftp)

EXISTENCE OF SOLUTIONS TO A SELF-REFERRED AND HEREDITARY SYSTEM OF DIFFERENTIAL EQUATIONS

EDUARDO PASCALI

Abstract. We establish the existence and uniqueness of a local solution for the system of differential equations

∂tu(x, t) =u vZ t

0

u(x, s)ds, t , t

∂tv(x, t) =v

u

Zt

0

v(x, s)ds, t

, t

. with given initial conditionsu(x,0) =u0(x) andv(x,0) =v0(x).

1. Introduction

Equations representing self-reference phenomena have been written of the form

Au(x, t) =u(Bu(x, t), t), (1.1)

whereA,B are functionals on a real function space. The existence and uniqueness of solutions to this equation have been studied by several authors. The particular case when the variable xdoes not appear explicitly was studied in [1, 2, 3]. More general cases have been studien in [4, 5, 6]. In [4],

Au(x, t) = ∂

∂tu(x, t) and Bu(x, t) = Z t

0

u(x, s)ds,

where B can be interpreted as a “memory” functional. In [6], we have considered the equation

2

∂t2u(x, t) =k1u∂2

∂t2u(x, t) +k2u(x, t), t

whereki are nonnegative real numbers, or bounded regular real functions. In this paper we establish the existence and uniqueness of local solutions for the system of functional differential equations

∂tu(x, t) =u vZ t

0

u(x, s)ds, t , t

∂tv(x, t) =v uZ t

0

v(x, s)ds, t , t

.

2000Mathematics Subject Classification. 47J35, 45G10.

Key words and phrases. Non-linear evolution systems; Hereditary systems.

c

2006 Texas State University - San Marcos.

Submitted June 13, 2005. Published January 16, 2006.

1

(2)

This system can be considered a model for the evolution of two reasonings, as follows: Ifxis an event,t is the time, andu(x, t), v(x, t) are two reasonings about xat time t, then the termv(Rt

0u(x, s)ds, t) can be considered as a “criticism” ofv over all previous reasonings ofuonx, up to timet.

2. The main result In this section we prove the following theorem.

Theorem 2.1. Let u0, v0 : R → R be bounded and Lipschitz continuous. Then, there existT0>0and two real bounded and Lipschitz continuous functionsu, v: R×[0, T0]→Rsuch that

∂tu(x, t) =u

vZ t 0

u(x, τ)dτ, t , t

∂tv(x, t) =v

u

Z t

0

v(x, τ)dτ, t , t u(x,0) =u0(x), v(x,0) =v0(x)

for allx∈Rand all t∈[0, T0]. Moreover the functionsu, v are unique.

Proof. Letu0, v0be given, and let L0, M0>0 be such that

|u0(x)−u0(y)| ≤L0|x−y|, |v0(x)−v0(y)| ≤M0|x−y|.

for all x, y∈R. Define the sequences of functions (un)n,(vn)n, for allx∈R and t >0,as follows:

u1(x, t) =u0(x) + Z t

0

u0 v0(u0(x)τ) dτ, v1(x, t) =v0(x) +

Z t

0

v0 u0(v0(x)τ) dτ, un+1(x, t) =u0(x) +

Z t

0

un vnZ τ

0

un(x, s)ds, τ , τ

dτ, vn+1(x, t) =v0(x) +

Z t

0

vn unZ τ

0

vn(x, s)ds, τ , τ

dτ.

Notice that

|u1(x, t)−u0(x)| ≤ ku0kt≡A1(t) (2.1)

|v1(x, t)−v0(x)| ≤ kv0kt≡B1(t), (2.2)

(3)

for allx∈R, t >0. Moreover, using (2.1), (2.2) we have

|u2(x, t)−u1(x, t)|

Z t

0

u1(v1( Z τ

0

u1(x, s)ds, τ)τ)dτ − Z t

0

u0(v0(u0(x)τ))dτ

Z t

0

u1(v1( Z τ

0

u1(x, s)ds, τ)τ)dτ − Z t

0

u0(v1( Z τ

0

u1(x, s)ds, τ))dτ

+

Z t

0

u0(v1( Z τ

0

u1(x, s)ds, τ))dτ − Z t

0

u0(v0(u0(x)τ))dτ

≤ Z t

0

ku0kτ dτ+ Z t

0

L0

v1(

Z τ

0

u1(x, s)ds, τ)−v0(u0(x)τ) dτ

≤ Z t

0

ku0kτ dτ+ Z t

0

L0h v1(

Z τ

0

u1(x, s)ds, τ)−v0( Z τ

0

u1(x, s)ds)

+ v0(

Z τ

0

u1(x, s)ds)−v0(u0(x)τ) i

≤ Z t

0

ku0kτ+L0

hkv0kτ+M0

Z τ

0

ku0ksdsi dτ

= Z t

0

A1(τ) +L0h

B1(τ) +M0 Z τ

0

A1(s)dsi dτ for allx∈R, and allt >0. In a similar way we prove

|v2(x, t)−v1(x, t)| ≤ Z t

0

B1(τ) +M0[A1(τ) +L0 Z τ

0

B1(s)ds]

dτ for allx∈R, and allt >0. We have also

|u1(x, t)−u1(y, t)| ≤L0|x−y|+ Z t

0

L20M0|x−y|τ dτ

= L0+

Z t

0

L20M0τ dτ

|x−y| ≡L1(t)|x−y|;

|v1(x, t)−v1(y, t)| ≤M0|x−y|+ Z t

0

L0M02|x−y|τ dτ

= M0+

Z t

0

L0M02τ dτ

|x−y| ≡M1(t)|x−y|.

It is easy to prove the inequality

|u2(x, t)−u2(y, t)| ≤h L0+

Z t

0

M1(τ)1 2

d dτ(

Z τ

0

L1(s)ds)2dτi

|x−y|. Set now

L2(t)≡L0+ Z t

0

M1(τ)1 2

d dτ

Z τ

0

L1(s)ds2

dτ.

Moreover, we remark that

|v2(x, t)−v2(y, t)| ≤h M0+

Z t

0

L1(τ)1 2

d dτ

Z τ

0

M1(s)ds2 dτi

|x−y|,

(4)

and set

M2(t)≡M0+ Z t

0

L1(τ)1 2

d dτ

Z τ

0

M1(s)ds2 dτ.

We define for allnand t >0:

An+1(t) = Z t

0

An(τ) +Ln−1(τ)[Bn(τ) +Mn−1(τ) Z τ

0

An(s)ds]

dτ; Bn+1(t) =

Z t

0

Bn(τ) +Mn−1(τ)[An(τ) +Ln−1(τ) Z τ

0

Bn(s)ds]

dτ; Ln+1(t) =L0+

Z t

0

Mn(τ)1 2

d dτ(

Z τ

0

Ln(s)ds)2dτ; Mn+1(t) =M0+

Z t

0

Ln(τ)1 2

d dτ(

Z τ

0

Mn(s)ds)2dτ.

By induction, it is easily to prove that for allx∈R, t >0,

|un+1(x, t)−un(x, t)| ≤An+1(t) (2.3)

|vn+1(x, t)−vn(x, t)| ≤Bn+1(t) (2.4) and, for allx, y∈R, t >0,

|un+1(x, t)−un+1(y, t)| ≤Ln+1(t)|x−y| (2.5)

|vn+1(x, t)−vn+1(y, t)| ≤Mn+1(t)|x−y| (2.6) In a very simple way we can prove also that for allx∈R, t >0,

|un+1(x, t)| ≤etku0k (2.7)

|vn+1(x, t)| ≤etkv0k (2.8) Since

0≤L1(t) =L0+M0L20t2/2 0≤M1(t) =M0+L0M02t2/2,

we can chooseT0>0 andh >0 such that 2h <1 and for allt∈[0, T0]:

L20t2 2 ≤1, M02t2

2 ≤1, (M0+L0)3t2

2 ≤M0∧L0, 0≤(M0+L0+ 1)t+ (M0+L0)2t2

2 ≤h.

Then 0≤L1(t),M1(t)≤M0+L0≡K0for allt∈[0, T0].

From the previous definitions we deduce:

0≤L2(t)≤L0+ Z t

0

M1(τ)1 2

d dτ

Z τ

0

L1(s)ds2

dτ ≤L0+K03t2 2, 0≤M2(t)≤M0+

Z t

0

L1(τ)1 2

d dτ(

Z τ

0

M1(s)ds)2dτ ≤M0+K03t2 2.

(5)

Then we have

0≤L2(t), M2(t)≤K0 ∀t∈[0, T0], and hence, by induction,

0≤Ln(t), Mn(t)≤M0+L0≡K0 ∀t∈[0, T0]. (2.9) From the definitions ofAn eBn, we deduce

0≤An+1(t)≤ Z t

0

An(τ) +K0Bn(τ) +K02 Z τ

0

An(s)ds dτ; 0≤Bn+1(t)≤

Z t

0

An(τ) +K0An(τ) +K02 Z τ

0

Bn(s)ds dτ.

For the continuity ofAn and Bn in [0, T0],we deduce:

0≤An+1(t)≤ kAnk

t+K02t2 2

+K0tkBnk; 0≤Bn+1(t)≤ kBnk

t+K02t2 2

+K0tkAnk. Now, for allt∈[0, T0],

0≤An+1(t); Bn+1(t)≤h(kAnk+kBnk)

Hence, taking the supremum overtand adding the inequalities, we deduce that the series

X(kAnk+kBnk) is convergent; then the same holds for both the seriesP

kAnkandP

kBnk. We remember that L(R×[0, T0];R) is a complete metric space with respect to lagrangian metric; then from the inequalities (2.3), (2.4), applying the Banach- Caccioppoli theorem, we have that (un)n and (vn)n are Cauchy sequences. Hence there exist two real functionsu andv, defined inR×[0, T0] such that: (un)n is uniformly convergent tou and (vn)n is uniformly convergent tov in R×[0, T0];

moreover, from (2.7),(2.8) and (2.9), u and v are Lipschitz continuous in all the variables.

We remark that, for alln∈N,x∈R,t∈[0, T0]:

un

vn

Z t

0

un(x, τ)dτ, t , t

−u vZ t

0

u(x, τ)dτ, t , t

≤ kun−uk+K0kvn−vk+K02tkun−uk. Thenuandv verify that for allx∈Randt∈[0, T0]:

u(x, t) =u0(x) + Z t

0

u v(

Z τ

0

u(x, s)ds, τ), τ dτ, v(x, t) =v0(x) +

Z t

0

v u(

Z τ

0

v(x, s)ds, τ), τ dτ,

(6)

respectively. Let us now prove the uniqueness. Let (u, v) be another pair of solutions and remark that:

u

v( Z τ

0

u(x, s)ds, τ), τ

−u

v(

Z τ

0

u(x, s)ds, τ), τ

≤K0 v(

Z τ

0

u(x, s)ds, τ)−v( Z τ

0

u(x, s)ds, τ)

+ u

v( Z τ

0

u(x, s)ds, τ), τ

−u v(

Z τ

0

u(x, s)ds, τ), τ

≤K0

K0

Z τ

0

u(x, s)ds− Z τ

0

u(x, s)ds

+ v(

Z τ

0

u(x, s)ds, τ)−v( Z τ

0

u(x, s)ds, τ))

+ku−uk

≤(1 +K02t)ku−uk+K0kv−vk. Therefore,

|u(x, τ)−u(x, τ)| ≤(t+K02t2

2)ku−uk+K0tkv−vk. In a similar way we can prove the estimates:

|u(x, τ)−u(x, τ)| ≤(t+K02t2

2)ku−uk+K0tkv−vk,

|v(x, τ)−v(x, τ)| ≤(t+K02t2

2)kv−vk+K0tku−uk. Then

|u(x, τ)−u(x, τ)| ≤(t+K02t2

2 +K0t) max(ku−uk;kv−vk), and

|v(x, τ)−v(x, τ)| ≤(t+K02t2

2 +K0t) max(kv−vk;ku−uk).

From (t+K02t22+K0t)≤h <1,we have

max(ku−uk;kv−vk)< hmax(ku−uk;kv−vk).

Then the uniqueness follows and the proof is complete.

3. Some Open Problems

The previous results and the proposed type of systems can be investigated and generalized in many different directions. In what follows, we give some of the problems whose investigation seems to be interesting.

(A) The first problem is to investigate the existence of global solutions, also for Lipschitzian and bounded initial data.

(B) It could be more difficult to establish existence and uniqueness for data u0, v0bounded and uniformly continuous (or simply continuous). Moreover, when the global existence is guaranteed, an interesting problem can be to give particular condition on data u0, v0 such that there exists T >0 for whichu(x, t) =v(x, t) for allx∈Randt≥T.

(7)

References

[1] E. Eder: The functional-differential equationx0(t) =x(x(t)), J. Diff. Equat. 54 (1984), no.

3, 399-400.

[2] Jan-Guo Si and Sui Sun Cheng: Analytic solutions of a functional differential equation with state dependent argument, Taiwanese J. Math., 1 (1997), no. 4, 471-480.

[3] Jan-Guo Si, Xin-Ping Wang and Sui Sun Cheng: Analytic solutions of a functional- differential equation with a stese derivative dependent delay, Aequ. Mathem., 57 (1999) no.

1, 75-86.

[4] M. Miranda (Jr), E. Pascali: On a type of evolution of self-referred and hereditary phenom- ena, Aequ. Mathem. (in press).

[5] M. Miranda (Jr), E. Pascali: Other classes of self-referred equations, (submitted).

[6] M. Miranda (Jr), E. Pascali: On a class of differential equations with self-reference, Rend.

Mat., serie VII, vol.25, Roma(2005), 155-164.

[7] V. Volterra: Opere Matematiche: Memorie e note, Vol. V, 1926-1940, Accademia Nazionale dei Lincei, Rome 1962.

Eduardo Pascali

Department of Mathematics “Ennio De Giorgi”, University of Lecce, C. P. 193, 73100, Lecce, Italy

E-mail address:[email protected]

参照

関連したドキュメント

Yang, Existence of solutions to the third-order nonlinear differential equations arising in boundary layer theory, Appl.. Yang, Positive solutions of singular Dirichlet boundary

The main purpose of this paper is to establish the existence, uniqueness and continuous dependence of classical solutions on initial data for a class of initial- boundary value

In this article, we are concerned with the existence of weak solutions for nonlocal (Kirchhoff type) parabolic variational inequality involving variable exponentc. Especially, in

The problem of solving the basic boundary value problems and Cauchy’s problem has been thoroughly investigated for a wide class of nonlinear parabolic equations of second order..

Braverman; On oscillation and asymptotic properties of a logistic equation, Proceedings of the International Conference: Biomathematics Bioinformatics and Applications of

In Section 4, we prove an uniform estimate for the attractor and finally, in Section 5, after obtaining some estimates for the flow of (1.1), we prove the upper semicontinuity

By transforming the studied equa- tion into a perturbed non-autonomous ordinary equation and using a bifur- cation result and the Poincar´ e procedure for this last equation, we

The Green’s function of the boundary-value problem (1.3) and the relevant prop- erties are to be presented later, and because of the nonlinear terms involving the higher-derivative