ABOUT THE CONDITIONING OF MATRICES IN THE P-VERSION OF THE FINITE-ELEMENT METHOD FOR 2ND-ORDER ELLIPTIC PROBLEMS

被引：4
|
作者：
MAITRE, JF [1 ]
POURQUIER, O [1 ]
机构：
[1] ECOLE CENT LYON,DEPT MATH,CNRS,URA 0740,EQUIPE ANALYSE NUMER LYON ST ETIENNE,F-69131 ECULLY,FRANCE
来源：
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS | 1995年 / 63卷 / 1-3期
关键词：
FINITE ELEMENT METHOD; CONDITION NUMBER; DIAGONAL PRECONDITIONING;
D O I：
10.1016/0377-0427(95)00062-3
中图分类号：
O29 [应用数学];
学科分类号：
070104 ;
摘要：
In the framework of adaptive methods, bases of hierarchical type are used in the p-version of the finite element method. We have studied the matrices corresponding to the most used basis, introduced by Babuska and Szabo, in the case of d-dimensional rectangular elements. For the internal nodes, we show that the condition number is O(p(4(d-1))) (resp, O(p(4d))) for the stiffness (resp. mass) matrix. Moreover, we show that the usual diagonal preconditioning divides the exponents by two.
引用
收藏
页码：341 / 348
页数：8
相关论文
共 50 条
12345