arrow
返回

An adaptive fast multipole boundary element method for three-dimensional potential problems

delete2006-03-15
delete73
PRE
AI
L
Liang Shen
Y
Yijun Liu *
DOI:10.1007/s00466-006-0046-9delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
An adaptive fast multipole boundary element method (FMBEM) for general three-dimensional (3-D) potential problems is presented in this paper. This adaptive FMBEM uses an adaptive tree structure that can balance the multipole to local translations (M2L) and the direct evaluations of the near-field integrals, and thus can reduce the number of the more costly direct evaluations. Furthermore, the coefficients used in the preconditioner for the iterative solver (GMRES) are stored and used repeatedly in the direct evaluations of the near-field contributions. In this way, the computational efficiency of the adaptive FMBEM is improved significantly. The adaptive FMBEM can be applied to both the original FMBEM formulation and the new FMBEM with diagonal translations. Several numerical examples are presented to demonstrate the efficiency and accuracy of the adaptive FMBEM for studying large-scale 3-D potential problems. The adaptive FMBEM is found to be about 50% faster than the non-adaptive version of the new FMBEM in solving the model (with 558,000 elements) for porous materials studied in this paper. The computational efficiencies and accuracies of the FMBEM as compared with the finite element method (FEM) are also studied using a heat-sink model. It is found that the adaptive FMBEM is especially advantageous in modeling problems with complicated domains for which free meshes with much more finite elements would be needed with the FEM.
Keyword:
fast multipole method
boundary element method
three-dimensional potential problems

期刊

Computational Mechanics 封面图
Computational Mechanics
IF:
3.8
论文数:
3.2K
被引数:
9.0K

机构

暂无机构信息
引用论文

引用论文

err分享
err收藏
Congenital Anomalies Induced in Hamster Embryos with Ribavirin
err1977-01-28
err0
PREAI
errLawrence Kilham; Virgil H. Ferm
err分享
err收藏
err分享
err收藏
err分享
err收藏
学者 查看更多内容