基于灰色关联理论的案例推理在故障智能诊断系统中的应用  

Applications of Case-based Reasoning Based on Grey-relational Theory in Fault Intelligent Diagnosis System

作　　者：贺晓[1] 刘景宁[2] 李淑霞[1] 

机构地区：[1]华中科技大学机械科学与工程学院,武汉430074 [2]华中科技大学计算机科学与工程学院,武汉430074

出　　处：《中国机械工程》2004年第22期2022-2026,共5页China Mechanical Engineering

基　　金：国家 8 63高技术研究发展计划资助项目( 2 0 0 3AA43 0 190 )

摘　　要：从灰色关联理论出发 ,结合欧几里德距离 ,提出了一种新的相似度计算方法 ,并以装载机故障智能诊断系统为例 ,将该计算方法应用于案例的检索中 ;同时从案例的表示、案例的检索、案例库的更新等方面 ,讨论了基于灰色关联理论的案例推理在整个系统中的实现方法以及案例推理与其他诊断方式的结合策略。该计算方法大大提高了相似度的分辨率 。Combining grey-relational theory and Euclidean distance, a new method for computing degree of similarity in case-based reasoning (CBR) was presented. And the systematic approach of intelligent fault diagnosis based on CBR was put forward, which included representation, search and updates of the cases. Especially, the grey-relational theory based case searching and matching method was described in detail. Furthermore, a case study of fault intelligent diagnosis system for loader is given. The strategic synthesis of CBR and other diagnosis methods was also envisioned. Through this method degree of similarity resolution is greatly improved, and computing problem of degree of compound similarity when multi-index factors are correlative is solved well, too.

关 键 词：案例推理 灰色关联理论 复合相似度 故障智能诊断 

分 类 号：TP306.3[自动化与计算机技术—计算机系统结构]