Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4640071 | Journal of Computational and Applied Mathematics | 2012 | 11 Pages |
Abstract
The problem of finding the least change adjustment to a stiffness matrix modeled by finite element method is considered in this paper. Desired stiffness matrix properties such as symmetry, sparsity, positive semidefiniteness, and satisfaction of the characteristic equation are imposed as side constraints of the constructed optimal matrix approximation for updating the stiffness matrix, which matches measured data better. The dual problems of the original constrained minimization are presented and solved by subgradient algorithms with different line search strategies. Some numerical results are included to illustrate the performance and application of the proposed methods.
Related Topics
Physical Sciences and Engineering
Mathematics
Applied Mathematics
Authors
Quan Yuan,