It remains to determine the collocation points that

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: d is a cubic polynomial it can be evaluated exactly by the two-point Gauss-Legendre quadrature formula (cf. 5], Chapter 7) Z1 1 ; 1 1 f ( )d = f (; p ) + f ( p ) + E 3 (9.1.31a) 3 where the discretization error E is given by 1 E = 135 f ( ) 2 (;1 1): iv (9.1.31b) Thus, applying (9.1.31) to (9.1.30), we have 1 1 1 1 1 1 (; p + 2 1 )(; p ; 2 2)P (; p ) + ( p + 2 1)( p ; 2 2)P ( p ) = 0: 3 3 3 3 3 3 Once again, the integrand in (9.1.30) is a cubic polynomial so E = 0. We see that we can satisfy the above condition by choosing 1 1 = 2= p : 23 (9.1.32a) Expressed in terms of the original variables through (9.1.29b), the collocation points are i 1=x 12; i; = h p i 23 i 2=x 12+ i; = h: p i 23 (9.1.32b) Regardless of how the result is expressed, the key is to perform collocation at the GaussLegendre points mapped to the appropriate interval. 11 N Midpoint Rule Collocation 2 0:20 10 3 5 0:64 10 5 10 0:10 10 4 0:46 10 6 20 0:26 10 5 0:33 10 7 40 0:65 10 6 0:23 10 8 80 0:16 10 6 0:16 10 9 ; ; ; ; ; ; ; ; ; ; Table 9.1.1: Maximum pointwise errors in the s...
View Full Document

This document was uploaded on 03/16/2014 for the course CSCI 6820 at Rensselaer Polytechnic Institute.

Ask a homework question - tutors are online