Induced Dimension Reduction Method to Solve the Quadratic Eigenvalue Problem
Conference Paper
(2017)
Author(s)
R. Astudillo (TU Delft - Electrical Engineering, Mathematics and Computer Science)
M.B. van Gijzen (TU Delft - Electrical Engineering, Mathematics and Computer Science)
Research Group
Numerical Analysis
DOI related publication
https://doi.org/10.1007/978-3-319-57099-0_20
Final published version
To reference this document use
https://resolver.tudelft.nl/uuid:d754e43c-8933-4755-8a7c-b2b6fc956002
More Info
expand_more
expand_more
Publication Year
2017
Language
English
Research Group
Numerical Analysis
Pages (from-to)
203-211
Publisher
Springer
ISBN (print)
9783319570983
Event
NAA 2016 (2016-06-15 - 2016-06-22), Lozenetz, Bulgaria
Downloads counter
190
Abstract
In this work we are interested in the numerical solution of the Quadratic Eigenvalue Problem (QEP) (λ2M + λD + K)x = 0, where M, D, and K are given matrices of order N. Particularly, we study the applicability of the IDR(s) for eigenvalues to solve QEP. We present an IDR(s) algorithm that exploits the special block structure of the lin-ealized QEP to compute its eigenpairs. To this end we incorporate ideas from Second Order Arnoldi method proposed in [3].