Structural Reliability Analysis Based On Stochastic Response Surface Method

There are so many uncertainties in the civil engineering that make it an important research field of the influences to structures. For the engineering structure, reliability stands for the ability of supportment for load and environment. Hence, it is significant to study how to compute the engineering structural reliability effectively.In this paper, we utilize Stochastic Response Surface Method to simulate the complicated relation between the input uncertainties and the system response, and transform the outputs into standard random variables through the Hermite polynomial chaos expansions. While obtain the stochastic response surface function, we build the limit-state function and combine with the geometric method to calculate the structural reliability. We carry on the study based on random variables and random fields respectively, and make comparison with Monte-Carlo simulation and Stochastic Finite Element Method based on perturbation techniques.The key advantage of SRSM is that the collocation points in the proposed method are selected for minimizing the mean square error, and from high probability regions, thus leading to fewer function evaluations for high accuracy. Comparison studies with the Monte-Carlo simulation suggest that the proposed method gives results of comparable accuracy at a lower computational cost. It has been shown that unlike the Perturbation Stochastic Finite Element Method the fairly good accuracy can be achieved even when the coefficients of variation of the input random variables are increased.