Motor Fault Diagnosis Based on Wavelet Energy and Immune Neural Network

Motor fault diagnosis methods are crucial in acquiring safe and reliable operation in motor drive systems. In this paper, a new method for the motor fault diagnosis is proposed based on wavelet packet transform (WPT) and artificial neural network (ANN). The energy of the vibration signals of motor can be obtained by the multi-decomposition of WPT and used as feature values of ANN inputs for fault diagnosis system. The artificial immune algorithm (AIA) for data clustering is employed to adaptively choose the centers and widths of the hidden layer centers of the radial basis function neural network (RBFNN). The simulation experiment results show the applicability and effectiveness of the proposed method to motor fault diagnosis.