An Improved DGMRES Algorithm for Computing the Drazin-inverse Solution of Singular Linear Systems

Abstract

Krylov subspace methods have been considered to solve singular linear systems Ax = b. One of these methods is the DGMRES method. DGMRES is an algorithm to solve the Drazin– inverse solution of the large scale and sparse consistent or inconsistent singular linear systems with with arbitrary index. In this paper, we present an improved version of this algorithm. Numerical experiments show that computation time is significantly less than that of computation time obtained by the DGMRES algorithm.