LAUSR

Permanent magnet demagnetization and rotor eccentricity are the main faults in permanents magnet synchronous motors (PMSMs). To detect these faults, motor current signature analysis (MCSA) has become the most common method in the field due to its simplicity, low computation time, and remote monitoring capability. Unfortunately, this method has a major drawback relating to the frequency signatures of demagnetization and eccentricity faults, which appear in the same frequency locations. To avoid this drawback while retaining the main advantages of the MCSA method, this paper proposes a new approach for discrimination between demagnetization and eccentricity faults in PMSMs. The proposed approach is based on the simultaneous monitoring of the real and imaginary components of the characteristic harmonics of these faults, which are obtained from stator current spectral analysis. This monitoring is based on the evolution of the characteristic harmonic signs in these two components. The positive or negative signs, make it possible to accurately discriminate between the effects produced by eccentricity faults from the effects produced by demagnetization faults. Finally, several experiments have been presented in this paper to demonstrate the capability and effectiveness of the proposed approach.