Variance Reduction Monte Carlo methods for wind turbines

Development of Variance Reduction Monte Carlo (VRMC) methods has proposed the possibility of estimation of rare events in structural dynamics. Efficiency of these methods in reducing variance of the failure estimations is a key parameter which allows efficient risk analysis, reliability assessment and rare event simulation of structural systems. Different methods have been proposed within the last ten years with the aim of estimating low failure probabilities especially for high dimensional problems. In this paper applicability of four of these methods i.e. Importance Sampling (IS), Distance Controlled Monte Carlo (DCMC), Asymptotic Sampling (AS) and Subset Simulation (SS) are compared to each other on a common problem. The aim of the study is to determine the most appropriate method for application on realistic systems, e.g. a wind turbine, which incorporate high dimensions and highly nonlinear structures.