Multidomain Line Sampling for Reliability and Sensitivity Analysis in Stochastic Linear Dynamics
摘要
This paper presents a framework for evaluating the first excursion probability and its sensitivity in linear dynamical structural systems subjected to Gaussian stochastic loading. The proposed approach involves multidomain Line Sampling, which is a technique that involves sampling along lines containing an infinite set of realizations of the stochastic loading. The numerical results demonstrate that this framework not only maintains accuracy but also exhibits superior computational efficiency compared to similar methodologies, making it a promising tool for reliability analysis.