摘要
论述了建立规则型模糊神经网络的理论和方法，针对大型旋转机械提出了一种采用多层规则库结构及智能推理机的故障诊断技术。该技术以 Ｒｕｌｅ 型模糊联想记忆器作为诊断系统的分类和综合算法，把基于知识的符号处理方法与模糊神经网络有机地结合在一起。讨论了模糊神经网络输入和输出模糊化的问题。为电厂汽轮发电机组故障诊断专家系统提供了新的思路。
Discussed in this paper are the theory and method of constructing a rule based fuzzy neural network (FNN). With respect to a large sized rotating machine the authors have come up with a fault diagnostic technique, which employs a multi layer rule base and intelligent reasoning approach. With the rule type fuzzy association memory device serving as a classification and synthesis algorithm of the diagnostic system the knowledge based symbol processing method is organically integrated with the FNN. Discussed are the issues of FNN input and output vectors. This paper has provided a new approach for setting up a fault diagnosis expert system suitable for power plant turbogenerators.
出处
《热能动力工程》
CAS
CSCD
北大核心
1999年第5期390-392,共3页
Journal of Engineering for Thermal Energy and Power
关键词
旋转机械
故障诊断
模糊神经网络
专家系统
rotating machinery, fault diagnosis, fuzzy neural network, expert system
