文献类型：期刊文献

中文题名：模糊神经网络推理的实时故障诊断专家系统

英文题名：Real-time Fault Diagnosis Expert System Based on Fuzzy Neural Network

作者：郑小霞[1];钱锋[1]

机构：[1]华东理工大学自动化研究所,上海200237

年份：2006

卷号：42

期号：3

起止页码：226

中文期刊名：计算机工程与应用

外文期刊名：Computer Engineering and Applications

收录：CSTPCD;;北大核心:【北大核心2004】；CSCD:【CSCD2011_2012】；

基金：国家自然科学基金资助项目(编号:60074027);国家863高技术研究发展计划资助项目(编号:2003AA412010);国家973重点基础研究发展规划资助项目(编号:2002CB3122000)

语种：中文

中文关键词：模糊神经网络;故障诊断;专家系统;PTA

外文关键词：fuzzy neural network,fault diagnosis,expert system,PTA

摘要：将模糊神经网络推理引入专家系统,采用修正的RLS算法训练网络的权系数,以此开发了模糊神经网络实时故障诊断专家系统,并将其成功应用于某化工厂大型PTA装置。文章介绍了系统的总体结构和主要特点,并以溶剂脱水塔釜水浓度高事件为例阐述了系统的具体实现。现场运行表明:该系统预报准确,界面友好,能满足工厂的实际需求,具有良好的易维护性和可扩充性。
A real-time fault diagnosis expert system based on fuzzy neural network has been developed by applying fuzzy neural network to expert system and using a modified learning algorithm based on the RLS method.It has been applied to the large scale 1WA plant.The total structure and main features are introduced.At the end,the realization has been introduced by taking example for high water on the bottom of PTA solvent dehydration tower.The practical application shows that it can make accurate forecast and has many specialties such as friendly interface,good maintainability and high extensibility.