An Accurate Tool for Detecting Stator Inter-Turn Fault in LSPMSM

Abstract

This paper proposes an accurate diagnosing tool that can predict the stator inter-turn size in line start permanent magnetic synchronous motor (LSPMSM). The proposed diagnosing approach is developed based on an experimentally validated mathematical model of the motor under inter-turn fault. The developed model has been tested using MATLAB® under different loading and fault size conditions. Since the stator currents and voltages are easily accessible, it is decided to use them as the key signatures for developing the diagnostic tool. Several time and frequency-based features have been extracted using motor current and voltage waveforms under different loading and fault size conditions. The developed tool has been designed to correlate the extracted features with its corresponding size of stator inter-turn fault. Finally, testing of the developed diagnosis tool shows a high accuracy of 96% in detecting the size. Moreover, the proposed diagnostic tool is examined against motor parameter variations. The results confirm the robustness of the proposed approach where the accuracy is slightly affected under a wide range of motor parameter variations.

Keywords

• Motor Faults
• LSPMSM
• Mathematical Modelling
• Neural Network