感应电机常见故障的模糊诊断（之一）

轴承故障、定子绝缘故障与绕组股线断股、鼠笼转子断条等故障约占鼠笼异步电机故障的８０％。在本文和其姊妹篇中，详细地讨了用模糊诊断方法对上述各类故障进行诊断的问题。本文将监测所得振动信号的时域方差和峭度，以及其振动信号的包络的幅频谱所得的几个特征频率处的幅度转化成隶属度向量，作为模糊诊断度。本文还介绍了用最小二乘原理，对实验室一台７．５ｋＷ的鼠笼异步电机滚动轴承故障的隶属度函数、模诊断关系矩阵进行自学

FAILURE DIAGNOSIS OF INDUCTION MOTOR WITH FUZZY LOGIC

The bearing failure, insulation failure and strand broken of stator winding, cage failure of rotor are the dominance part of cage induction motor failure. The paper and its companion paper discuss the diagnosis of this kind of failure mentioned above using fuzzy logic approach. The paper presents the forming of input vector of fuzzy inference which elements are the membership functions corresponding to the statistic characteristics (variance and kurtosis) obtained from the time domain wave of its vibration signal, and the vibration amplitude of bearing parts at their fault feature frequencies obtained by the envelop analysis of vibration signal measured which is based on the resonant demodulation principle. The the malfunction degree of several typical failures in the bearing can be learned. Additionally, by Principle of least square, the methods of computer self learning for membership functions and fuzzy relation matrixes are developed. The diagnosis software based on the fuzzy logic approach has been used for bearing fault diagnosis of a 7.5kW inductor motor, the results show that the fuzzy logic failure diagnosis is an effective approach.