Targeted random sampling for time-invariant reliability analysis Shields, Michael D.; Sundar, V. S.
The Targeted Random Sampling (TRS) method is extended to moderate-to-high dimensional time invariant reliability analysis. TRS is an adaptive sampling method rooted in stratified sampling variance reduction technique and represents a specific algorithm for Refined Stratified Sampling. In this work, a Markov Chain is used to sample the space densely in the vicinity of the limit stat and multi-dimensional stratum division is proposed along with a neural network based approximation of the performance function. The method is shown to perform efficiently and accurately for estimation of probability of failure for a 20 dimensional problem.
Item Citations and Data
Attribution-NonCommercial-NoDerivs 2.5 Canada