Harmonic projection methods for large non-symmetric eigenvalue problems

From MaRDI portal





Given an \(n\times n\) non-symmetric eigenvalue problem \(Bz=\lambda z\), interior eigenvalues near a chosen value \(\sigma \) are to be extracted along with relevant eigenvectors. Applying the Rayleigh-Ritz and the Arnoldi procedure to the operator \((B-\sigma I)^{-1}\), the authors propose the interior (also called harmonic) Rayleigh-Ritz and the interior/harmonic Arnoldi method, respectively. They also present further modifications inclusive of preconditioning. All algorithms are illustrated and tested by numerical examples which, together with a fairly detailed discussion, cover one third of the paper and deal with interior eigenvalues extraction and spectrum approximation problems. Besides the proposed methods the reader may also appreciate the list of references comprising 54 items.



Cites work


Cited in
(57)


Describes a project that uses

Uses Software






This page was built for publication: Harmonic projection methods for large non-symmetric eigenvalue problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4940813)