摘要
针对电磁场分析中的大型稀疏对称线性方程组 ,给出一种新的预处理解法———带有松弛因子的修改型不完全因子分解共轭梯度法 (RMICCG法 )。适当选取松弛因子 ,RMICCG法具有较快的渐进收敛速度、较好的计算稳定性和较广泛的适用范围。数值例子表明 ,RMICCG法比常规ICCG法减少 30 %以上的迭代次数和计算时间。
A new preconditioning solution for large real sparse symmetric linear equations in finite element analysis of electromagnetic field problem is presented in this paper. With the proper choice of the relaxation factor, RMICCG method have not only rapid rate of convergence, but also better stability of computational process and wide applications. Numerial examples show that RMICCG method can reduce iterative number and the computation time by 30% less than the ICCG method.
出处
《电工电能新技术》
CSCD
2002年第1期69-72,共4页
Advanced Technology of Electrical Engineering and Energy
