Rotor Cage Assessment in Soft-Started Induction Motors Applying DWT and Hilbert Transform to Transient Stray Flux Signals

The identification of the fault-related signatures, caused by the presence of fault harmonics in the electrical signals, is quite more difficult in soft-started motors than in line-started ones. This is not only because of the higher harmonic content introduced by soft-starters, but also because of the amplification of certain components already present in line-fed motors. This work proposes a methodology that allows to enhance the identification of the rotor-fault-related components in soft-started induction motors. This methodology relies on the obtention of the envelope of the start-up stray-flux signals through the Hilbert transform. Then, the Discrete Wavelet Transform (DWT) is applied to the resulting envelope signals. This proposed method, unlike other methodologies, allows to clearly see the rotor-fault-related signatures when using different models of soft-starter and in different operating conditions. Additionally, an indicator to quantify the severity level of the failure is introduced. The results shown in this paper confirm the potential of this approach, even when different topologies of soft-starter and operating conditions are used.