计算物理 ›› 1998, Vol. 15 ›› Issue (1): 53-57.

• 论文 • 上一篇    下一篇

无界区域D-N区域分解算法的松弛因子选取与收敛速率

余德浩   

  1. 中国科学院计算数学与科学工程计算研究所, 科学与工程计算国家重点实验室, 北京 100080
  • 收稿日期:1997-03-12 修回日期:1997-05-07 出版日期:1998-01-25 发布日期:1998-01-25
  • 作者简介:余德浩,男,52,研究员,博士,北京2719信箱中科院计算数学研究所
  • 基金资助:
    国家自然科学基金资助项目19331021,19472069

SELECTION OF RELAXATION FACTOR AND CONVERGENCE RATE FOR D-N DOMAIN DECOMPOSITION METHOD OVER UNBOUNDED DOMAIN

Yu Dehao   

  1. State Key Laboratory of Scientific and Engineering ComputingInstitute of Computational Mathematics and Scientific/Engineering Computing Chinese Academy of Sciences, Beijing 100080
  • Received:1997-03-12 Revised:1997-05-07 Online:1998-01-25 Published:1998-01-25

摘要: 在D-N区域分解算法中,松弛因子的选取起着关键作用。应用自然边界归化理论,讨论了无界区域上椭圆型方程边值问题的D-N区域分解算法,给出了其松弛因子的简便的选取方法,保证了算法很好的收敛性。

关键词: D-N区域分解算法, 无界区域, 松弛因子, 自然边界归化

Abstract: The selection of relaxation factor plays a key role in D-N domain decomposition methods (DDM). Based on the natural boundary reduction, the D-N DDM for elliptic problems over unbounded domain is discussed. A simple and convenient approach to selection of relaxation factor is given, and an optimal convergence is obtained.

Key words: Dirichlet-Neumann (D-N) DDM, unbounded domain, relaxation factor, natural boundary reduction

中图分类号: