Rotor Fault Detection of Squirrel Cage Induction Motor Using Spectrum Analysis of Dynamic Simulation and Experimental Validation

Abstract

This study investigates the equations of characteristic frequencies in the torque and vibration spectrum to determine rotor faults based on the dynamic simulation of a squirrel cage induction motor with a broken rotor bar. The dynamic simulation was developed using the winding function approach and characteristic frequencies were analyzed using Fast Fourier Transform. Using the equations of characteristic frequencies to detect broken rotor bars, we found not only that sideband slip frequencies formed around the harmonics of rotation, but also sideband slip frequencies formed around the harmonics of a rotor bar's frequency. Experimental results have validated characteristic frequencies in the torque spectrum to determine rotor faults of the dynamic simulation, and show that in order to determine that a rotor bar is broken, the characteristic frequencies in vibration analysis are the same as those in torque analysis. Therefore, either a vibration data acquisition device or a torque data acquisition device can be used. Moreover, the current V line analysis is used to validate characteristic frequencies of broken rotor faults in the spectrums. The equations of characteristic frequencies have been applied to detect rotor faults in squirrel cage induction motors at the early stage.