基于小波包与神经网络的往复压缩机故障诊断方法

Wavelet packet and neural network based approach for fault diagnosis of reciprocating compressors

针对用传统方法难以提取往复压缩机故障特征的实际情况 ,提出用小波包分解和单支重构来构造能量特征向量的方法 ,直接利用各频段成分能量的变化来提取往复压缩机的故障特征。用该方法构造的特征向量能突出反映往复压缩机的故障特征 ,通过用径向基 (RBF)神经网络进行故障诊断 ,结果表明 ,该方法可有效地诊断往复压缩机的各种故障。另外 。

It is difficult to extract the fault features of reciprocating compressors by conventional methods, so a new method of fault feature extraction is proposed by decomposition and reconstruction of wavelet packet in this paper. The feature vectors that reflect the energy change of different frequency ranges are constructed by the method and the feature vectors can extract the fault feature of reciprocating compressors effectively. Moreover, RBF neural network is used to diagnose various faults by the feature vectors constructed above. Result shows various faults of reciprocating compressors can be diagnosed correctly. In addition, the approach can also be applied to the fault diagnosis of other equipment.