Rotor fault detection in induction motors operated with different types of soft-starters

Development of advanced techniques for predictive maintenance are essential to extend the useful life of induction motors (IM). This paper aims to validate a diagnosis method for rotor fault detection, which is based on the transitory current analysis (ATCSA) during the motor start-up. Such analysis is carried out by applying two kinds of time-frequency representation (TFR): Short-time Fourier Transform (STFT) and Discrete Wavelet Transform (DWT). Three different types of soft-starters are tested in order to evaluate the method feasibility in the presence of noise produced by power electronics devices. The followed methodology consists in assessing the results provided by the diagnosis method when the motor is under healthy and failure condition. In both cases, measurements are performed for each kind of soft-starters actuating as motor driver.