Block subspace projection preconditioned conjugate gradient method for structural modal analysis
Authors:
- Sergiy Fialko,
- Viktor Karpilovskyi
Abstract
The method for extracting natural vibration frequencies and modes of design models arising when the finite element method is applied to the problems of structural and solid mechanics is proposed. This approach is intended to be used on multicore SMP computers and is an alternative to the conventional block Lanczos and subspace iteration methods widely used in modern FEA software. We present the main idea of the method as well as the parallel fast block incomplete factorization approach for creating efficient preconditioning, the shift technique and other details accelerating the solution and improving the numerical stability. Real-life examples are taken from the computational practice of SCAD Soft IT company and approve the efficiency of the proposed method.
- Record ID
- CUT23e6a7e3ca0847ebb9d6761d3fbaa0b6
- Publication categories
- ; ;
- Author
- Pages
- 497-506
- Other elements of collation
- il. (w tym kolor.); Bibliografia (na s.) - 506; Bibliografia (liczba pozycji) - 26; Oznaczenie streszczenia - Abstr.
- Substantive notes
- Tyt. źródła wg okł.
- Book
- Ganzha Maria, Maria Ganzha Maciaszek Leszek, Leszek Maciaszek Paprzycki Marcin Marcin Paprzycki (eds.): Proceedings of the 2017 Federated Conference on Computer Science and Information Systems, September 3-6, 2017, Prague, Czech Republic, Annals of Computer Science and Information Systems, no. 11, 2017, New York City, Warszawa, Institute of Electrical and Electronics Engineers, Institute of Electrical and Electronics Engineers, Polskie Towarzystwo Informatyczne, ISBN 978-83-946253-7-5 (Web)
- Keywords in English
- block preconditioned conjugate gradient method, finite element method, natural vibration analysis, multicore computers, multithreading, reorthogonalization procedure
- DOI
- DOI:10.15439/2017F64 Opening in a new tab
- URL
- https://annals-csis.org/Volume_11/drp/64.html Opening in a new tab
- Language
- eng (en) English
- License
- Score (nominal)
- 20
- Additional fields
- Indeksowana w: Web of Science, Scopus
- Uniform Resource Identifier
- https://cris.pk.edu.pl/info/article/CUT23e6a7e3ca0847ebb9d6761d3fbaa0b6/
- URN
urn:pkr-prod:CUT23e6a7e3ca0847ebb9d6761d3fbaa0b6
* presented citation count is obtained through Internet information analysis, and it is close to the number calculated by the Publish or PerishOpening in a new tab system.