Power Generation System Reliability Evaluation Base On Cross Entropy Method

Due to the low calculation efficiency of Monte Carlo simulation in power system reliability analysis, a stochastic simulation algorithm based on the combination of multi- layer cross entropy and antithetic variable sampling technology for reliability analysis is proposed in this paper. First, the multilayer cross-entropy method is used to construct an approximate function of the zero variance probability density function to improve the extraction probability of the small probability failure events. Then, based on the constructed approximate probability function, the antithetic variable sampling technology is used for sampling, which further improves the convergence speed of sampling. Finally, the proposed stochastic simulation algorithm is used for the stochastic production simulation of a wind power generation system, and the reliability analysis of the power generation system is performed. The results of case studies verify the effectiveness of the proposed improved cross entropy method.