Fault diagnosis of induction motor rotor based on BP neural network and D-S evidence theory

Directing to the shortage of single method of BP neural network or D-S evidence theory in rotor fault diagnosis, a fault diagnostic method for induction motor rotor was proposed, which was based on BP neural network and D-S evidence theory. The BP neural network method was applied to the fault diagnosis firstly, and then, the partial diagnostic results of BP neural network were taken as the basic probability assignment, finally, the D-S evidence theory was applied to fuse different results from all the neural networks and got the finally diagnostic results. The experiment simulation results of fault diagnostic example show that the method is available for the induction motor rotor fault diagnosis and has better classified diagnosis ability than single fault diagnostic method.