Fault diagnosis of mine hoist based on fuzzy fault tree and Bayesian network

In order to solve problems of low efficiency and poor accuracy of existing mine hoist fault diagnosis methods, a fault diagnosis method of mine hoist based on fuzzy fault tree and Bayesian network was proposed. Firstly, denoising preprocessing and multi-source information fusion are carried out for hoist running parameters collected by sensors in real time, which can ensure accuracy of the data. Then the processed data is input into fault tree of mine hoist, and triangular fuzzy number is used to represent occurrence probability of bottom even to obtain fuzzy probability of bottom event. Finally, the fuzzy fault tree is mapped to Bayesian network for reliability analysis, and the fuzzy probability of bottom event is taken as priori probability to calculate probability of leaf node occurrence, thus posterior probability, probability importance and key importance of root node are obtained, so as to quickly determine fault type and fault location. The example analysis results verify feasibility of the method.