Fault detection in induction motors using Hilbert and Wavelet transforms

In this work, a new on-line method for detecting incipient failures in electrical motors is proposed. The method is based on monitoring certain statistical parameters estimated from the analysis of the steady state stator current (for broken bars, saturation, eccentricities, and bearing failures) or the axial flux signal (for coil short-circuits in the stator windings). The approach is based on the extraction of the envelop of the signal by Hilbert transformation, pre-multiplied by a Tukey window to avoid transient distortion. Then a wavelet analysis (multi-resolution analysis) is performed, which makes the fault diagnosis easier. Finally, based on a statistical analysis, the failure thresholds are determined. Thus, by monitoring the mean value estimate it is possible to detect an incipient failure condition on the machine.