Peano existence theorem explained

In mathematics, specifically in the study of ordinary differential equations, the Peano existence theorem, Peano theorem or Cauchy–Peano theorem, named after Giuseppe Peano and Augustin-Louis Cauchy, is a fundamental theorem which guarantees the existence of solutions to certain initial value problems.

History

Peano first published the theorem in 1886 with an incorrect proof.[1] In 1890 he published a new correct proof using successive approximations.[2]

Theorem

Let

D

be an open subset of

R x R

with

f\colonD\toR

a continuous function and

y'(x)=f\left(x,y(x)\right)

a continuous, explicit first-order differential equation defined on D, then every initial value problem

y\left(x0\right)=y0

for f with

(x0,y0)\inD

has a local solution

z\colonI\toR

where

I

is a neighbourhood of

x0

in

R

,such that

z'(x)=f\left(x,z(x)\right)

for all

x\inI

.

The solution need not be unique: one and the same initial value

(x0,y0)

may give rise to many different solutions

z

.

Proof

By replacing

y

with

y-y0

,

x

with

x-x0

, we may assume

x0=y0=0

. As

D

is open there is a rectangle

R=[-x1,x1] x [-y1,y1]\subsetD

.

Because

R

is compact and

f

is continuous, we have

style\supR|f|\leC<infty

and by the Stone–Weierstrass theorem there exists a sequence of Lipschitz functions

fk:R\toR

converging uniformly to

f

in

R

. Without loss of generality, we assume

style\supR|fk|\le2C

for all

k

.

yk,n:I=[-x2,x2]\toR

as follows, where

x2=min\{x1,y1/(2C)\}

.

yk,0(x)\equiv0

, and

styleyk,n+1

x
(x)=\int
0

fk(x',yk,n(x'))dx'

. They are well-defined by induction: as

\begin{aligned}|yk,n+1

x|f
(x)|&\lestyle\left|\int
k(x',y

k,n(x'))|dx'\right|\\&\lestyle|x|\supR|fk|\\&\lex2 ⋅ 2C\ley1,\end{aligned}

(x',yk,n+1(x'))

is within the domain of

fk

.

We have

\begin{aligned}|yk,n+1(x)-yk,n

x|f
(x)|&\lestyle\left|\int
k(x',y

k,n(x'))-fk(x',yk,n-1(x'))|dx'\right|\\&\lestyleLk\left|\int

x|y
k,n

(x')-yk,n-1(x')|dx'\right|,\end{aligned}

where

Lk

is the Lipschitz constant of

fk

. Thus for maximal difference

styleMk,n(x)=\supx'\in[0,x]|yk,n+1(x')-yk,n(x')|

, we have a bound

styleMk,n(x)\leLk\left|\int

x
0

Mk,n-1(x')dx'\right|

, and

\begin{aligned}Mk,0

x|f
(x)&\lestyle\left|\int
k(x',0)|dx'\right|\\&\le

|x|style\supR|fk|\le2C|x|.\end{aligned}

By induction, this implies the bound

Mk,n(x)\le

n|x|
2CL
k

n+1/(n+1)!

which tends to zero as

n\toinfty

for all

x\inI

.

The functions

yk,n

are equicontinuous as for

-x2\lex<x'\lex2

we have

\begin{aligned}|yk,n+1(x')-yk,n+1

x'
(x)|&\lestyle\int
x

|fk(x'',yk,n(x''))|dx''\\&style\le|x'-x|\supR|fk|\le2C|x'-x|,\end{aligned}

so by the Arzelà–Ascoli theorem they are relatively compact. In particular, for each

k

there is a subsequence
(y
k,\varphik(n)

)n\inN

converging uniformly to a continuous function

yk:I\toR

. Taking limit

n\toinfty

in
\begin{aligned}style\left|y
k,\varphik(n)
xf
(x)-\int
k(x',y
k,\varphik(n)
(x'))dx'\right|&=|y
k,\varphik(n)
(x)-y
k,\varphik(n)+1

(x)|\\&\le

M
k,\varphik(n)

(x2)\end{aligned}

we conclude that

styleyk(x)=\int

xf
k(x',y

k(x'))dx'

. The functions

yk

are in the closure of a relatively compact set, so they are themselves relatively compact. Thus there is a subsequence

y\psi(k)

converging uniformly to a continuous function

z:I\toR

. Taking limit

k\toinfty

in

styley\psi(k)

xf
(x)=\int
\psi(k)

(x',y\psi(k)(x'))dx'

we conclude that
xf(x',z(x'))dx'
stylez(x)=\int
0
, using the fact that

f\psi(k)

are equicontinuous by the Arzelà–Ascoli theorem. By the fundamental theorem of calculus,

z'(x)=f(x,z(x))

in

I

.

Related theorems

The Peano theorem can be compared with another existence result in the same context, the Picard–Lindelöf theorem. The Picard–Lindelöf theorem both assumes more and concludes more. It requires Lipschitz continuity, while the Peano theorem requires only continuity; but it proves both existence and uniqueness where the Peano theorem proves only the existence of solutions. To illustrate, consider the ordinary differential equation

y'=\left\vert

1
2
y\right\vert
on the domain

\left[0,1\right].

According to the Peano theorem, this equation has solutions, but the Picard–Lindelöf theorem does not apply since the right hand side is not Lipschitz continuous in any neighbourhood containing 0. Thus we can conclude existence but not uniqueness. It turns out that this ordinary differential equation has two kinds of solutions when starting at

y(0)=0

, either

y(x)=0

or

y(x)=x2/4

. The transition between

y=0

and

y=(x-C)2/4

can happen at any

C

.

The Carathéodory existence theorem is a generalization of the Peano existence theorem with weaker conditions than continuity.

l{H}

: for an open subset

D

of

R x l{H}

, the continuity of

f\colonD\toR

alone is insufficient for guaranteeing the existence of solutions for the associated initial value problem.[3]

References

Notes and References

  1. G. . Peano . Sull'integrabilità delle equazioni differenziali del primo ordine . Atti Accad. Sci. Torino . 21 . 1886 . 437–445 .
  2. G. . Peano . Demonstration de l'intégrabilité des équations différentielles ordinaires . . 37 . 2 . 1890 . 182–228 . 10.1007/BF01200235 . 120698124 .
  3. Yorke . J. A. . James A. Yorke . 1970 . A continuous differential equation in Hilbert space without existence . Funkcjalaj Ekvacioj . 13 . 19–21 . 0264196.