Open-switch Fault Diagnosis of Inverters in Linear Induction Machine Drives Based on Normalized Wavelet Packet Entropy

The linear induction machine (LIM) drive is the power source of the medium-low speed Maglev trains, in which the inverter plays the role of regulating the speed of the LIM. Thus, the reliable operation of the inverter is importance. In this paper, a fault diagnosis method for the open-switch faults in the inverter is introduced. It utilizes the phase currents only and requires no additional sensors or hardware. The amplitudes of the currents are normalized to strengthen the performance under different working conditions. Meanwhile, the several kinds of wavelet packet entropies are investigated regarding the diagnosis ability. Finally, an open-switch fault diagnosis method is proposed, and its effectiveness is verified by the simulation results.