XIII. Solving Y = <b>K</b>Y + f.

## Linear Methods of Applied Mathematics Evans M. Harrell II and James V. Herod*

version of 1 June 1996

## XIII. Solving Y = KY + f

In this chapter we shall learn how to solve integral equations in three situations:

• K has a separable kernel,

• K has norm less than one,

• K is approximated by K's with separable kernels.

These terms will be explained as they are encountered.

First we suppose that there are an integer n and functions such that, for each p , ap and bp are in L2[0,1]. Then K has a separable kernel if its kernel is given by With the supposition the K is separable, it is not hard to find y such that y = Ky + f, for this equation can be re-written as or, using the notation of inner products, One can guess that, if the sequence of functions on [0,1] is a linearly independent sequence, then y will have this special form:

there is a sequence {cp} of numbers such that In fact, supposing there is such a sequence, we determine what it should be.

Suppose Substitute this in the equation to be solved: and we see that This now reduces to a matrx problem: Define K and f to be the matrix and vector so defined that the last equation is rewritten as

c = K c + f.

We now employ ideas from linear algebra. The equation c = K c + f has exactly one solution provided

det( 1 - K ) != 0.

The Fredholm Alternative Theorems for matrices address these ideas. (Review the alternative theorems for matrices.) If the sequence is found then we have a formula for y(x).

EXAMPLE: In Exercises 1.2, it should have been established that if

K(x,t) = 1 + sin([[pi]]x) cos([[pi]]t),

then

K*(x,t) = 1 + sin([[pi]]t) cos([[pi]]x).

Also,

y = Ky has solution y(x) = 1

and

y = K*y has solution y(x) = [[pi]] + 2 cos([[pi]]x).

It is the promise of the Fredholm Alternative theorems that

y = Ky + f

has a solution provided that Let us try to solve y = Ky + f and watch to see where the requirement that f should be perpendicular to the function [[pi]] +2 cos([[pi]]t) appears.

To solve y = Ky + f is to solve We guess that the solution is of the form y(x) = a + b sin([[pi]]x) + f(x) and substitute this for y: From this, we get the algebraic equations Hence, in our guess for y, we find that a can be anything and that b must be and also must be The naive pupil might think this means there are two (possibily contradictory) requirements on b. The third of the Fredholm Alternative theorems assures the student that there is only one requirement!

If the kernel is not separable, an alternative hypothesis that will enable one to solve the equation y = Ky + f is to suppose that the kernel for K is small. Of course this does not mean that K is of the form K(x,t) = .007 x t . Rather, we ask that K should be small in a sense developed below. The technique for getting a solution in this case is to iterate.

Take [[phi]]0(x) to be f(x) and [[phi]]1 to be defined by It is reasonable to ask: does this generated sequence converge to a limit and in what sense does it converge? The answer to both questions can be found under appropriate hypothesis on K.

THEOREM If K satisfies the condition that then limp [[phi]]p(x) exists and the convergence is uniform on [0,1] - in the sense that if u = limp[[phi]]p then

limp maxx | u(x) - [[phi]]p(x) | = 0.

SUGGESTION FOR PROOF: Note that Furthermore, if p is a positive integer, the distance between successive iterates can be computed: Inductively, this does not exceed Thus, if and n > m then Hence, the sequence {[[phi]]p} of functions converges uniformly on [0,1] to a limit function and this limit provides a solution to the equation COROLLARY. If and

u = limp [[phi]]p

then There is different, independent, concept of K being small which leads to convergence of the iteration process in the norm of L2[0,1] This hypothesis asks that THEOREM If K satisfies the condition that then limp [[phi]]p(x) exists and the convergence is in norm- meaning that if u = limp[[phi]]p then

limp || u(x) - [[phi]]p(x) || = 0.

INDICATION OF PROOF. The analysis of the nature of the convergence will go like this:

|| [[phi]]1 - [[phi]]0 || 2

is defined to be As before,  and

u = limp [[phi]]p

then

||u - [[phi]]m || < F(rm+1,1 - r) ||f||.

Before addressing the final case - where

K does not have a separable kernel, we generate "resolvents" for the integral equations.

Re-examining the iteration process:

[[phi]]0(x) = f(x),

[[phi]]1(x) = K[[phi]]0(x) + f(x)

[[phi]]2(x) = K(K([[phi]]0))x + K(f)(x) + f(x)

One writes [[phi]]0=f, [[phi]]1=Kf+f, [[phi]]2 = K[Kf+f] + f = K2f+Kf+f, .....

In fact, with   Hence, the kernel K2 associated with K2 is Inductively, and We have, in this section, conditions which imply that

[[Sigma]]p=1 Kpf

converges and that its limit y satisfies y = Ky + f. Many authors call this series of operators the "resolvent" and denote

R = [[Sigma]]p=1 Kp.

Note that R is a function which operates on elements of L2[0,1]. One writes that

y = Ky + f

has solution

y(x) = [( 1 + R ) f](x) = f(x) + I(0,1, ) R(x,t) f(t) dt.

Suggestive algebra can be made by identifying (1 + R ) as

(1 - K ) -1 = 1 + K( 1 - K ) -1, so that R = K ( 1 - K ) -1.

Please refer to the accompanying Mathematica notebook for the solution by iteration of a typical integral equation, including error estimates.