将聚类PSO优化粗糙集-神经网络算法引入水电机组故障诊断中,利用聚类算法和PSO算法的优点改进粗糙集属性约简方法并对水电机组故障的检测信息进行约简,提取对故障分类起主要作用的信息,并用BP神经网络对粗糙集处理后的故障信息进行诊断。该方法降低了神经网络的输入信息空间维数,简化神经网络结构,有效提高故障诊断的准确性。通过对水电机组振动故障数据进行诊断分析, The clustering PSO optimization rough set - neural network algorithm is introduced into fault diagnosis of hydropower units. By using the advantages of clustering algorithm and PSO algorithm, the method of attribute reduction of rough set is improved and the detection information of hydropower unit faults is reduced. The fault classification Play a major role in the information, and use BP neural network to deal with the rough set of fault diagnosis information. This method reduces the dimension of the input information space of neural network, simplifies the structure of neural network and effectively improves the accuracy of fault diagnosis. Through the diagnosis and analysis of hydroelectric generating set vibration fault data,