arXiv:1705.03038 [math.NA]AbstractReferencesReviewsResources
Energy Error Estimates of Subspace Method and Multigrid Algorithm for Eigenvalue Problems
Yunhui He, Qichen Hong, Hehu Xie, Meiling Yue, Chunguang You
Published 2017-05-08Version 1
In this paper, we give a new error estimate of the subspace projection method for eigenvalue problem in energy norm. The relationship between the error estimates in $L^2$-norm and energy norm is also derived. Based on these new results, the convergence analysis is provided for the inverse power method on the subspace. Then we present the analysis for the geometric and algebraic multigrid methods for the eigenvalue problem based on the convergence result of the inverse power method on the subspace.
Comments: 16 pages, 0 figure
Categories: math.NA
Related articles: Most relevant | Search more
arXiv:2410.13358 [math.NA] (Published 2024-10-17)
Subspace method based on neural networks for eigenvalue problems
arXiv:1409.7944 [math.NA] (Published 2014-09-28)
A Full Multigrid Method for Eigenvalue Problems
arXiv:1401.4969 [math.NA] (Published 2014-01-20)
Local and Parallel Finite Element Algorithm Based On Multilevel Discretization for Eigenvalue Problem