Research on Fault Diagnosis Model of Shearer Cutting Motor Based on Fuzzy Neural Network and Expert System

Insufficient fault data or label data has always been one of the reasons for the low accuracy of fault diagnosis models. In this paper, we propose a novel fault diagnosis method named Fuzzy Neural Network with Expert System (ES-FNN), which combines fuzzy neural networks with expert systems to improve diagnosis accuracy. Specifically, ES-FNN introduces expert knowledge to participate in training set construction, thus avoids the impact of small data set on model training. Furthermore, the combination of fuzzy logic system and neural network can fuzzifies the input information and strengthen inference. Finally, we compared the performance of the proposed model with other three classical neural network algorithms: BP, GABP, RBF and ES-FNN. Results under the same settings demonstrate that the proposed model effectively learns fuzzy rules and achieves the highest diagnosis accuracy among the compared models.