Improved Direct Importance Sampling Method for System Reliability

摘要

This report presents an importance-sampling method for the first-order problem ofreliability analysis of structural systems, having a failure domain defined by linear or linearized functions. Truncated multimodal simulation is suggested as a new technique, offering an advantage of locating all samples in the failure domain and thus increasing computation efficiency. Variance of estimator is evaluated by an analytically derived upper bound. It is compared with that of the conventional Monte Carlo method by a variance change factore for conservative estimation of increase in accuracy and efficiency. The upper bound of variance can be used for a priori determination of required sample size, given an acceptable maximum error associated with a confidence level. Various application examples of both series and parallel systems are included for illustration