Pryse, S.E., Adhikari, S. and Kundu, Abhishek ![]() |
Preview |
PDF
- Accepted Post-Print Version
Download (1MB) | Preview |
Abstract
A comparative study of two new Galerkin projection schemes to com- pute the response of discretised stochastic partial di�erential equations is presented for discretised structures subjected to static and dynamic loads. By applying an eigen-decomposition of a discretised system, the response of a discretised system can be expressed with a reduced basis of eigen-components. Computational reduction is subsequently achieved by approximating the random eigensolutions, and by only including dominant terms. Two novel error minimisation techniques have been proposed in order to lower the error introduced by the approximations and the trunca- tions: a) Sample-based Galerkin projection scheme, b) Sample-aggregated based Galerkin projection scheme. These have been applied through intro- ducing unknown multiplicative scalars into the expressions of the response. The proposed methods are applied to analyse the bending of a cantilever beam with stochastic parameters undergoing both a static and a dynamic load. For the static case the response is real, however the response for the case of a dynamic loading is complex and frequency-dependent. The re- sults obtained through the proposed approaches are compared with those obtained by utilising a direct Monte Carlo approach.
Item Type: | Article |
---|---|
Date Type: | Publication |
Status: | Published |
Schools: | Engineering |
Publisher: | Elsevier |
ISSN: | 0266-8920 |
Date of First Compliant Deposit: | 12 October 2017 |
Date of Acceptance: | 20 September 2017 |
Last Modified: | 18 Nov 2023 21:16 |
URI: | https://orca.cardiff.ac.uk/id/eprint/105476 |
Citation Data
Cited 5 times in Scopus. View in Scopus. Powered By Scopus® Data
Actions (repository staff only)
![]() |
Edit Item |