基于粗集神经网络的故障诊断专家系统研究  

Research of Fault Diagnosis Expert System Based on Rough Set and Neural Network

作　　者：朱万富[1] 徐宪明[2] 

机构地区：[1]安徽理工大学机械工程学院,安徽淮南232001 [2]中国石油天然气集团公司,北京100007

出　　处：《广州化工》2013年第15期18-20,共3页GuangZhou Chemical Industry

基　　金：安徽理工大学引进人才基金项目资助(11137)

摘　　要：对于化工、石化行业使用最广的一类机器-往复式压缩机的故障诊断,提出了一种粗集集成神经网络的故障诊断专家系统模型,模型中粗集约简算法首先被用来对诊断输入数据进行约简,然后被用来对神经网络进行剪枝优化;测试结果表明该系统诊断时不仅速度快,而且正确率高,明显优于其它系统。For the chemical industry and petrochemical industry, reciprocating compressor was one of the most widely used classes of machine. In order to diagnose the fault of the compressor, a new model for expert system based on rough sets was put forward, which was combined tightly between rough sets and neural network, rough sets was used to not only reduce and optimize the fault diagnosis data, but also prolong and optimize the structure of neural network. The results showed that the expert system had better efficiency and diagnosis accuracy than the other system, so it was estimated that the expert system would be applied in fault diagnosis.

关 键 词：往复式压缩机 粗集 决策表 神经网络 故障诊断 专家系统 

分 类 号：TQ440.5[化学工程—化学肥料工业]