Reliability assessment of multistate wind turbine gear train system based on T-S fuzzy fault tree and Bayesian network

In recent years, the number of wind farms and the power of wind turbines have been greatly improved, and the gearing system, as a key structure in doubly-fed wind turbines, is of great significance to the safe and stable operation of wind turbines. Therefore, this paper uses a combination of the T-S fuzzy fault tree and Bayesian network to analyze the reliability of wind turbine gear transmission systems. According to the type of gearbox faults, the fault tree models of the lubrication system, cooling system, monitoring and protection system, and mechanical components are established, respectively. Then, the Bayesian network model is determined by the method of transforming the T-S fuzzy fault tree to the Bayesian network. Finally, the basic events and gate events in the fault tree are determined. These are then fuzzified using T-S fuzzy logic and combined with expert natural language descriptions of the different faults to derive the fuzzy probability of the actual fault occurrence in the system. Finally, the reliability indexes of the gearbox components are calculated by combining the T-S fuzzy fault tree and the Bayesian network. The findings indicate that this approach can reliably assess the reliability of wind turbine gearing systems, which is of significant importance in enhancing the reliability of wind turbines.