A RAYLEIGH–RITZ SUBSTRUCTURE SYNTHESIS METHOD IN PHYSICAL CO-ORDINATES FOR DYNAMIC ANALYSIS OF STRUCTURES
✍ Scribed by M. Scheble; C.N. Strizzolo; J. Converti
- Publisher
- Elsevier Science
- Year
- 1998
- Tongue
- English
- Weight
- 227 KB
- Volume
- 213
- Category
- Article
- ISSN
- 0022-460X
No coin nor oath required. For personal study only.
✦ Synopsis
1.
A new method is presented that retains all the advantages of the more elaborate existing Rayleigh-Ritz methods such as the revised approach presented by Jen et al. [1], but with two additional important advantages.
The first advantage is related to simplicity of order reduction by using standard ''assembling'' techniques similar to those used in FEM. Thus, assembling and order reduction is simpler and numerically more efficient than singular value decomposition. This results from the use of Boolean instead of full real matrices and shows up as improved precision in large complex structures. The second advantage is related to computational cost. The Rayleigh-Ritz type methods lead to full matrices K and M to be diagonalized. The method presented yields sparse matrices that can be solved at much lower cost by sparse matrix techniques developed for FEM. The characteristic feature of this approach is to work in physical instead of generalized co-ordinates. The general steps of the procedure will be developed and commented upon in section 2. The main features are described below.