Detection of the broken rotor bars of squirrel-cage induction motors based on normalized least mean square filter and Hilbert envelope analysis

Abstract

This paper presents a combination of normalized least mean square (NLMS) filter and Hilbert envelope analysis method to detect the broken rotor bars of a squirrel-cage induction motor. Broken rotor bars are one of the most common fault types that may be seen in induction motors. Rotor faults are reflected in the stator current of the motor as side-band harmonics. Most of the previous works utilize motor current signature analysis using spectral methods to determine required features for detecting motor faults. It may be difficult to detect the faults from stator current when the motor is lightly loaded under noisy conditions. In this study, Hilbert envelope analysis method is used along with an NLMS filter to process the stator current of an induction motor to detect the broken rotor bars. The proposed method is verified experimentally (under 25, 50, and 100 % loading conditions). The results of the recommended method are compared with the results of Fast Fourier Transform (FFT) analysis performed on the stator current of the same motor under the same conditions. The results show that the recommended method gives a more accurate solution compared with the results of FFT method.