Wind turbine reliability analysis with fault-related weighted Bayesian networks

Due to each component has an unbalanced influence on the reliability of the whole machine during the failure process of wind turbines as well as the multi-fault correlation is complicated, a method for reliability analysis of multi-fault mode correlation failure based on weighted Bayesian networks is proposed. By analyzing the wind turbine fault data, the weight of the wind turbine subsystem is calculated based on the combined weight method of fuzzy analytic hierarchy process, entropy weight method and historical fault maintenance data method, and the reliability of the subsystem is obtained by using the weighted reliability calculation formula; Considering the correlation of subsystem faults and the characteristics of Bayesian graphics and probabilistic reasoning, combined with Copula function model and Bayesian network, a wind turbine reliability model considering fault-related Bayesian network is established. Research shows that the reliability of the wind turbine system is between completely independent and completely correlated faults when weighting is considered, which can provide a reference for the reliability design and optimization of wind turbines.