基于D-S证据理论的感应电动机转子故障诊断方法研究

提出运用D-S(Dempster-Shafer)证据理论对感应电动机转子断条故障进行识别的故障诊断方法.基于小波包变换的频率划分特性,对定子电流信号进行小波包分解,建立了转子断条故障的特征矢量,提取转子断条故障的特征信息,利用BP神经网络对其识别的结果形成彼此独立的证据,并根据D-S证据融合规则进行融合处理,以实现对电动机转子断条故障的准确识别.实验结果表明,该方法克服了传统基于FFT分析方法难以提取故障特征频率分量的难点,提高了故障诊断的判决精度,可实现转子断条故障的可靠诊断.

Rotor Fault Diagnosis Method of Induction Motor Based on D-S Evidential Theory

A method based on the D-S (Dempster-Shafer)evidential theory is presented.It is applied to diagnoses the broken rotor bars of the induction motors. Firstly, the wavelet packets decomposition of the stator current is carried out based on the frequency partition characteristics of the wavelet packets transformation. The faults characteristic vector of the broken rotor bars is established, and the characteristic information of the faults is picked up. Then, independent evidence is gained based on the identification result from the BP neural network, and is fused according to the D-S evidential fusion algorithm in order to identify accurately the faults of the induction motors. Finally, the experimental results show that the presented method is reliable, which overcomes the difficulty of the traditional FFT analysis in the fault characteristic frequency extraction and improves the identification precision in fault diagnosis.