Diagnostic and protection of inverter faults in IPM motor drives using wavelet transform

This paper presents a novel faults diagnostic and protection technique for interior permanent magnet (IPM) motor drives using wavelet transform. The proposed wavelet based diagnostic and protection technique for inverter faults is developed and implemented in real-time for a voltage source inverter fed IPM motor. In the proposed technique, the motor currents of different faulted and unfaulted conditions of an IPM motor drive system are preprocessed by wavelet packet transform. The wavelet packet transformed coefficients of motor currents are used as inputs of a three-layer wavelet neural network. The performances of the proposed diagnostic and protection technique are investigated in simulation and experiments. The proposed technique is experimentally tested on a laboratory 1-hp IPM motor drive using the ds1102 digital signal processor board. The test results showed satisfactory performances of the proposed diagnostic and protection technique in terms of speed, accuracy and reliability.