Approximate sensitivity analysis of structure system based on failure probability under epistemic and aleatory uncertainties

Pan Wang, Zhenzhou Lu, Qi Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

For analyzing effect of epistemic and aleatory uncertainties on failure probability of structure system involving uncertainties, two sensitivity indices, i.e. Correlation Coefficient and Correlation Ratio on the failure probability are defined, the universal method and approximate method for solving two indices are investigated and the corresponding precision and efficiency are discussed. By analyzing the two indices on the failure probability, the importance ranking of the uncertainty impacting on the failure probability can be quantified, on which the information for controlling the failure probability can be obtained effectively. After the detailing implementations for two sensitivity indices, several numerical and engineering samples are used to validate the precision and efficiency of two methods, and the law of the impact of the uncertainty parameters on the sensitivity result is obtained as well.

Original languageEnglish
Title of host publicationAdvances in Civil Engineering and Architecture
Pages5674-5679
Number of pages6
DOIs
StatePublished - 2011
Event1st International Conference on Civil Engineering, Architecture and Building Materials, CEABM 2011 - Haikou, China
Duration: 18 Jun 201120 Jun 2011

Publication series

NameAdvanced Materials Research
Volume243-249
ISSN (Print)1022-6680

Conference

Conference1st International Conference on Civil Engineering, Architecture and Building Materials, CEABM 2011
Country/TerritoryChina
CityHaikou
Period18/06/1120/06/11

Keywords

  • Epistemic and aleatory uncertainties
  • Failure probability
  • Sensitivity
  • Structure system

Fingerprint

Dive into the research topics of 'Approximate sensitivity analysis of structure system based on failure probability under epistemic and aleatory uncertainties'. Together they form a unique fingerprint.

Cite this