1: \begin{abstract}

2: This paper has proposed the GMRES that augments Krylov subspaces with a set of approximate right singular vectors.

3: %GMRES that augments Krylov subspaces with a set of approximate singular vectors has proposed in this paper. 

4: The proposed method suppresses the error norms 

5: %in addition to residual norms 

6: of a linear system of equations. Numerical experiments comparing the proposed method with the Standard GMRES and GMRES with eigenvectors methods\cite{aug}  have been reported for benchmark matrices.

7: %It is well-known that at each iteration the GMRES method reduces residual norm associated with the linear system of equations. In this paper, we propose an augmentation of a search subspace in the GMRES method with singular vector

8: %approximations. The proposed

9: %method has an advantage that  apart from reducing residual norms it also reduces corresponding error norms.  Numerical

10: %experiments have been carried out on some benchmark matrices and results have been compared with the standard GMRES and the GMRES with

11: %eigenvectors proposed in \cite{aug}. Numerical results have been demonstrated that the error

12: %reduction feature of the proposed method accelerates the convergence of approximate solutions in the GMRES.

13: \end{abstract}

14: