Fault Diagnosis of Motor Broken Bar Using Current and Vibration Fusion Signal

Abstract

Motor broken bar is a common fault for the asynchronous motor. An intelligent diagnosis method for motor broken bar fault is presented. The intelligent diagnosis method combining Ensemble Empirical Mode Decomposition (EEMD) and Support Vector Machine (SVM) is used to identify the fault type of the motor broken bar. EEMD is used to extract the frequency character from motor vibration signal and current signal. By comparing the stability of the frequency band energy characteristics between the vibration signal and the current signal under different working conditions, it is concluded that the results of vibration signal is better than the current signal. At the same time, the quad-classifier kernel function parameters are optimized using the grid selection method. A multi-information fusion method based on current and vibration signal is designed. It is effective to identify broken bar fault from motor multi-faults. It can resolve the difficulty of multi-component fault feature extraction from multi-faults with broken bar fault.