广义逐次超松弛加速Hermitian和skew-Hermitian分裂迭代法.pdfVIP

On generalized successive-overrelaxation acceleration of the Hermitian and skew-Hermitian splitting iterations Ling ling Meng ∗ Yu jiang Wu School of Mathematics and statistics, Lanzhou University, Lanzhou 730000, P.R.China Abstract For solving large sparse non-Hermitian positive definite linear equations, Bai, Golub and NG studied an Hermitian and skew-Hermitian splitting methods(HSS). Bai, Golub and Michael K. Ng recently further generalize this technique to the normal and skew-Hermitian splitting methods (NSS) . In this paper, we present a generalized successive-overrelaxation (GSOR) acceleration scheme which involve two iteration parameters for the NSS iteration, specifically results in more generalized acceleration scheme for the NSS iteration. Theoretical determine the convergence domain of the GSOR acceleration scheme under the assumption that all the eigenvalues of the corresponding block Jacobi iteration matrix are real. A numerical example is used to show that the GSOR technique can significantly accelerate the convergence rate of the NSS or the HSS iteration method. Key words: non-Hermitian matrix; normal matrix; skew-Hermitian matrix; splitting ; gener- alized successive-overrelaxation; acceleration technique 1 Introduction In scientific computing, many problems give rise to a system of linear equations Ax = b, (1.1) where A ∈ Cn ×n is a nonsingular large sparse non-Hermitian and positive definite matrix, and x, b ∈ Cn . Iterative methods for the system of linear e