Improvement of the Hilbert Method via ESPRIT for Detecting Rotor Fault in Induction Motors at Low Slip

The traditional Hilbert method to detect broken rotor bar fault in induction motors is reviewed and its major drawbacks are clearly revealed, namely, deteriorating or even completely failing when a motor operating at low slip due to the fixed constraints of fast Fourier transform (FFT) is used in this method. To overcome this, the estimation of signal parameters via rotational invariance technique (ESPRIT) is then introduced to replace FFT, and an improved Hilbert method is thus presented by conjugating the Hilbert transform and ESPRIT together. Experimental results of a small motor in a laboratory and a large motor operating on an industrial site are reported to demonstrate the effectiveness of the improved Hilbert method.