Design and emulate on motor fault diagnosis system
Xiuli Zeng, Zheng Qiang Yao, Xuemei Song, Qingxin Zhao · 2010
In the traditional motor fault diagnosis, only a certain type of motor fault diagnosis was diagnosed. The less amount of information leads to diagnostic conclusions unreliable. In this article, a new fault diagnosis method was put forward. Information fusion technology, stator current and rotor vibration signals as a diagnostic characteristics input signal were introduced into the motor fault diagnosis. Neural network method was applied to the fault identification. In order to improve the diagnostic precision, the input signs were divided into the stator current signal related and the rotor vibration signal related. They separately adopt a diagnosis sub-network to complete different aspects of fault diagnosis. Finally, each sub-network diagnostic results information fusion were carried out and the final diagnosis results were got. The simulation of the diagnostic method showed that it is feasible that the neural network data fusion applied to the motor fault diagnosis.