An Adaptive Importance Sampling Strategy

by Yaacob Ibrahim, Natl Univ of Singapore, Singapore,



Document Type: Proceeding Paper

Part of: Mechanics Computing in 1990's and Beyond

Abstract: An adaptive importance sampling scheme is presented. The scheme updates both the estimate of system failure probability and the importance sampling density. The sampling density used is the stratified density in which each component is associated with a failure mode. The initial choice of the sampling density is based upon component reliability information and the density is updated based upon the failed points for each mode uncovered. Results obtained for a brittle system compares well with the Monte Carlo estimate.

Subject Headings: Adaptive systems | Failure analysis | Failure modes | Structural analysis | Density stratification | Probability | Brittleness | Information management | Europe | Monaco | Monte Carlo

Services: Buy this book/Buy this article

 

Return to search