Aiming at the non- station of vibration signals,this paper presents a motor fault diagnosis method based on the wavelet packet decomposition and the neural network for the sake of quick and accurate detection to the reasons of rotor unbalance. This method strikes the kurtosis of frequency band through the coefficient of wavelet packet,and gains the feature vector from various changes in the kurtosis of each frequency band,and identifies fault with RBF neural network,and uses the MATLAB software to realize i...