Fault diagnosis of mechanical unbalance for permanent magnet synchronous motor drive system under nonstationary condition

Motor current signature analysis based on Fast Fourier Transform (FFT) has been successfully used in a permanent magnet synchronous motor (PMSM) drive system for fault diagnosis. However, the method does not always achieve good results according to two aspects. Firstly, the amplitudes of fault harmonic components are quiet small compared with the fundamental amplitude if the fault severity is small. Then fault characters are likely to be hidden and difficult to be distinguished in the stator current. Secondly, as the speed is not constant and this causes variation on the fault characteristic frequencies where FFT is no longer suitable. This paper presents a novel method to diagnose the mechanical unbalance for a PMSM drive system under nonstationary condition, which is based on Park´s vector and discrete wavelet transform. Simulations are carried out in the PMSM drive system based on MATLAB/Simulation and fault indicators are successfully extracted from the square of the stator current Park´s vector modulus. The simulation results show that the proposed method can effectively detect the mechanical unbalance fault in the PMSM drive system under nonstationary condition, and the method is suitable for different control strategies.