Online mechanical fault diagnosis of induction motor by wavelet artificial neural network using stator current

A novel online fault diagnostic method for the mechanical faults of induction motors is proposed. The method is based on artificial neural networks and Wavelet Packet Decomposition. New feature for mechanical fault detection is defined through the signature analysis of the wavelet packet decomposition coefficients of induction motor stater current. The theoretical background and description of the detection algorithm using artificial neural network is presented. Simulation results prove that the proposed method accurately detects the faults for a wide range of load conditions.