一般化共役勾配法(GCG法)の有限要素法への適用に関する研究 : その4 特異性判定修正式と平面・立体骨組の悪条件下数値実験

概要

論文の詳細を見る
The generalized conjugate gradient method (the GCG method) is a powerful method to solve large sparse symmetric positive definite linear systems than the direct methods for the solution of linear equations with banded or skylined matrices and is able to check the singularity of matrices. In this paper, modified equations to find the singularity of coefficient matrices, the ordering effect of the GCG method, improvement of diagonal matrix in the incomplete modified Cholesky decomposition and numerical simulations of plane and space frames in ill-condition are given. The results of numerical simulations show that the GCG method is useful in the solution of large sparse systems of equations by the accuracy and stability of this method.
社団法人日本建築学会の論文
1999-07-30