A subspace iteration for eigenvector derivatives (Q3989494)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 34616
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | A subspace iteration for eigenvector derivatives |
scientific article; zbMATH DE number 34616 |
Statements
A subspace iteration for eigenvector derivatives (English)
0 references
28 June 1992
0 references
numerical example
0 references
overrelaxation
0 references
eigenvector derivatives
0 references
rate of convergence
0 references
subspace iterations
0 references
large-scale systems
0 references
0.8629767894744873
0 references
0.846035897731781
0 references
0.8089555501937866
0 references
0.8010775446891785
0 references
The paper exploits a new direction in the form of an iterative process to simultaneously calculate eigenvector derivatives of many eigenvectors with respect to multi-variables. The method fully uses all the available information from the preceding eigenvalue solution and thus economizes computational efforts. It iterates through two equations derived from the first variation of the two fundamental equations used in a subspace iteration method. It is interesting to notice that, usually, the method does not require good starting trial vectors for improving the overall rate of convergence. An overrelaxed scheme is given as well as an illustrative example. Since the subspace iterations do not require the decomposition of a system-size matrix, the approach seems to be useful for large-scale systems.
0 references