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1 理論およびアルゴリズムの概要
1994 年 59 巻 456 号 p. 87-95
詳細
1994 年 59 巻 456 号 p. 87-95
There has been remarkable progress in the computational engineering science by the large scale linear and non-linear finite element analyses. This study proposes an analytical procedure to find the solution of sparse systems arising from the finite element structural analyses by a generalized conjugate gradient method with symmetric matrices. The fundamental equations are derived retaining the advantages of the conjugate gradient method which includes the PCG method (e.g. the ICCG method), and can find the singularity of matrices. In this paper, theory of the generalized conjugate gradient method and the outline of algorithm are presented.
