Reliability analysis of Bayesian Monte Carlo Adaptive Importance Sampling method for structural safety

According to the high nonlinear and strong correlation characteristics of limit state function for structural reliability analysis, and the defects of low efficiency and long computational time of classical Monte-Carlo simulation sampling Bayesian Monte Carlo Adaptive Importance Sampling method (BMCAIS) is introduced in this paper. Considering MCAIS needs initial important sampling density function, the function is inferred by the posterior distribution with Bayes´ theorem and Maximum Entropy. Then MCAIS is used to improve efficiency of realization in the simulation. This study suggests speeding up the simulation process by considering the logical dependence of neighboring points as prior information. Also the different priors can be integrated into one model by using BMCAIS to further reduce cost of simulations. Finally, BMCAIS is used in reliability analysis of structure dynamics simulation for some smart antenna structure, and the results are compared with classical MC in order to verify the applicability of this method.