Restarted GMRES preconditioned by deflation

被引：139

作者：

Erhel, J

Burrage, K

Pohl, B

机构：

[1] INST NATL RECH INFORMAT & AUTOMAT,F-35042 RENNES,FRANCE

[2] UNIV QUEENSLAND,DEPT MATH,BRISBANE,QLD 4072,AUSTRALIA

[3] ETH ZURICH,SEMINAR ANGEW MATH,CH-8092 ZURICH,SWITZERLAND

来源：

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS | 1996年 / 69卷 / 02期

关键词：

GMRES; preconditioning; invariant subspace; deflation;

D O I：

10.1016/0377-0427(95)00047-X

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

This paper presents a new preconditioning technique for the restarted GMRES algorithm. It is based on an invariant subspace approximation which is updated at each cycle. Numerical examples show that this deflation technique gives a more robust scheme than the restarted algorithm, at a low cost of operations and memory.

引用

页码：303 / 318

页数：16

共 19 条

[1]

ASHBY SF, 1987, THESIS U ILLINOIS UR

[2]

Barret R., 1993, TEMPLATES SOLUTION L

[3]

HUANG Y, 1989, 8909 FAC TECHN MATHI

[4]

JARAUSCH H, 1987, NUMERICAL METHODS BI

[5] A ROBUST GMRES-BASED ADAPTIVE POLYNOMIAL PRECONDITIONING ALGORITHM FOR NONSYMMETRIC LINEAR-SYSTEMS [J].

JOUBERT, W .

SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1994, 15 (02) :427-439

[6]

JOUBERT W, 1994, J NUMER LINEAR ALGEB, V1, P427

[7]

MICHELLI CA, 1983, SIAM J NUMER ANAL, V20, P362

[8]

POHL B, 1994, 9402 ETH

[9]

REICHEL L, 1994, NUMER MATH, V67, P21

[10] PRACTICAL USE OF POLYNOMIAL PRECONDITIONINGS FOR THE CONJUGATE-GRADIENT METHOD [J].

SAAD, Y .

SIAM JOURNAL ON SCIENTIFIC AND STATISTICAL COMPUTING, 1985, 6 (04) :865-881

← 1 2 →