Fault diagnosis and location of brushless DC motor system based on Wavelet Transform and Artificial Neural Network

The reliability of Electro-mechanical Actuator (EMA) is extremely important in industrial, commercial, aerospace, and military applications. Fault diagnosis and location of the brushless DC motor (BLDCM) system used in the EMA offer a means of improving reliability and security of the EMA. In this paper normal model as well as three fault models of the BLDCM system, which are stator winding inter-turn short circuit fault model, open-switch fault model and open-winding fault model, are developed. Performance characteristics under the faulty conditions are studied through simulation. Using Wavelet Transform (WT) and Artificial Neural Network (ANN), fault diagnosis and location method of BLDCM system is developed. Simulation results demonstrate the validity of the proposed method.