ScholarLens Back to search
2003•Journal of Xidian UniversityRequires access

Iterative solution for dense linear systems arising in computational electromagnetics

Changhong Liang

Open publisher page 2 citations

Abstract

This paper discusses some preconditions with Krylov methods for the solutoin of large dense complex symmetric nonHermitian systems arising in computational elecetromagnetics. Many tests show that BICGSTAB or GMRES Krylov subspace methods preconditioned with ILUT are quite efficient for this class of application. They can deliver a good rate of convergence but their construction and storage are not expensive.

About this research paper

What this paper is about

This paper discusses some preconditions with Krylov methods for the solutoin of large dense complex symmetric nonHermitian systems arising in computational elecetromagnetics. Many tests show that BICGSTAB or GMRES Krylov subspace methods preconditioned with ILUT are quite efficient for this class of application. They can deliver a good rate of convergence but their construction and storage are not expensive.

Why it matters

OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper discusses some preconditions with Krylov methods for the solutoin of large dense complex symmetric nonHermitian systems arising in computational elecetromagnetics. Many tests show that BICGSTAB or GMRES Krylov subspace methods preconditioned with ILUT are quite efficient for this class of application. They can deliver a good rate of convergence but their construction and storage are not expensive.

Key concepts: Biconjugate gradient stabilized method, Krylov subspace, Generalized minimal residual method, Linear system, Computer science, Applied mathematics, Convergence (economics), Iterative method

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Iterative solution for dense linear systems arising in computational electromagnetics — Research Paper | ScholarLens