Study On Fault Diagnosis Of Induction Motor With Broken Rotor Bars

With simple structure, low price and good environmental adaption, induction motor has been widely used in production. The good running status of induction motor is a guarantee of the normal production, once the serious fault occurs to motor, the entire production process will be interrupted. Broken rotor bar is one of the common faults, therefore, the research on the fault of induction motor with broken rotor bar has great theoretical significance and economic benefit.This paper firstly expounds the basic principle of broken bars detection of induction motor, and also introduces the theory knowledge of adaptive filter. According to the analysis of the fault current, the method based on the traditional LMS adaptive notch and FFT is adopted to extract the fault feature and has been verified by simulation and experimental results. In order to solve the contradiction between convergence rate, steady-state error and tracking of the traditional LMS algorithm, on the basis of the traditional LMS algorithm, two new algorithms are proposed: variable step size VS-LMS algorithm and LMS optimization algorithm, and this article has given experimental analysis under the two new algorithms. Above all in the under-frequency power supply, in order to improve the accuracy and reliability of fault diagnosis, the article also analyzes the stator current in the under-converter supply, by using the continuous refinement Fourier transform and band-pass filter to accomplish the fault detection of the rotor bars breaking, and experiment results has proved that the method above is feasible and effective, at the same time, the paper has summed up the change ruler of fault feature under different frequencies and loads of the motor.