Wavelet-based switching faults detection in direct torque control induction motor drives

Abstract

This study describes a method of detection and identification of IGBT-based drive open-circuit faults in direct torque control (DTC) induction motor drives. The detection mechanism is based on wavelet decomposition. The Symlet2 wavelet was selected as the wavelet base to perform stator current analysis during transients. In this method, the stator currents will be used as an input to the system. The MATLAB program was used to process discreet wavelet transform (DWT) of the signals. The stator current was used for the detection of the fault. When an open-circuit fault appears in an inverter IGBT, the signal fault information is included in each frequency region. There are spikes in the sixth level detail for incipient faults. The time of the spikes in the DWT is correlated with the time of fault. As a result of time domain studies, a faulty system can be easily discriminated from a healthy one. In this study the analysis of electrical transients arising during switch open-circuit faults in a three-phase power inverter feeding DTC induction motor drives was decomposed using the wavelet transform. The results demonstrate that the proposed fault detection and diagnosis system has very good capabilities.