This preview shows page 1. Sign up to view the full content.
Unformatted text preview: able 9.1.2.
Following the logic introduced in (9.1.30), Flaherty and Mathon selected their collocation points to re ect the boundary layer in the Green's function. Speci cally, they
1=x 12; h
2=x 12+ h
; i i; = i i i; = i symmetrically disposed with respect to the center of each subinterval so that Z1
; 1 e ; (1 ) 2
2( ; 4 2 )P ( )d = 0
13 for polynomials P ( ) of as high a degree as possible. 9.2 Collocation for First-Order Systems
Let us extend the collocation methods to rst-order vector BVPs of the usual form y = f (x y) a<x<b (9.2.1a) g (y(a)) = g (y(b)) = 0: (9.2.1b) 0 L R We'll follow a di erent approach than the one used in Section 9.1 and integrate (9.2.1a)
on a subinterval (x 1 x ) to obtain
i; i y(x) = y(x 1 ) + Z x Z y (x)dx = y(x 1 ) +
0 i; xi ;1 x i; xi ;1 f (x y)dx: (9.2.2) As with initial value methods, we'll construct a numerical method by approximating y
(or f ) by a polynomial and integrating the result. Any polynomial basis may be used,
but let us concentrate on the Lagrange form of the interpolating polynomial
J y (x) =
0 k where Y
J L ( )=...
View Full Document
This document was uploaded on 03/16/2014 for the course CSCI 6820 at Rensselaer Polytechnic Institute.
- Spring '14
- The Land