基于粗糙集和AFSA-BP的水电机组故障诊断

将粗糙集和人工鱼群优化-神经网络算法引入水电机组故障诊断中,利用人工鱼群优化算法中聚类的特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。实验结果表明:该方法降低了神经网络的输入信息空间维数,简化了神经网络结构,有效提高了故障诊断的准确性。

Fault diagnosis of hydropower unit based on AFSA-BP neural network and rough set

The neural network combined with the rough set optimized by artificial fish swarm is ap?plied to the fault diagnosis of hydroelectric units. We use the clustering characteristics of the artificialfish swarm algorithm to improve the attribute reduction of rough set and reduce the fault information ofhydropower units to obtain the key fault features. And, the BP neural network is used to diagnose thefault information processed by the modified rough set. The experimental result shows that the methodcan reduce the dimension of the neural network inputs and simplifying the structure of the neural net?work, and effectively improve the accuracy of the fault diagnosis.