基于CBR的故障诊断系统案例检索策略  

Research on case retrieval strategy in fault diagnosis system based on case-based reasoning

作　　者：刘景宁[1] 刘涛[1] 贺晓[2] 

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

出　　处：《华中科技大学学报（自然科学版）》2008年第3期16-19,共4页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家高技术研究发展计划资助项目(2003AA430190)

摘　　要：研究了面向对象的案例表示方法和层次结构的案例组织模式.在对故障案例合理分类的基础上,引入案例集故障状态向量作为案例的分类标志,引入案例发生频率以实现相似案例的最终择优选择,将权重向量放在各个案例集中分别计算.提出了一种类选、粗选、精选、择优"四步走"的案例检索策略,将灰色关联理论和欧几里德距离相结合,采用了一种新的相似度计算方法,以此作为"精选"的主要算法.以装载机的故障智能诊断系统为实例,完成了系统设计与实现.In the representation and organization of cases, the paper dived into the object-oriented case representation approach and hierarchical cases organization pattern. Based on sorting fault cases reasonably, fault state vector of case set was introduced and regarded as a symbol. And case frequency was presented to carry out how to pick up the best from the similar cases, later the weight vector of a case was calculated in different case sets. A retrieval strategy which consisted of four steps had been presented, and the four steps were sort choosing, fuzzy choosing, accurate choosing and choosing the best. It combined grey-relational theory and Euclidean distance, and it＇s also main arithmetic that adopted a new method in calculating similarity degree to make a best choice. Finally, a fault intelligent diagnosis system for a loader was developed and the system could improve the accuracy and efficiency of loader fault diagnosis.

关 键 词：故障诊断 案例推理 案例检索 故障状态向量 

分 类 号：TP306.3[自动化与计算机技术—计算机系统结构]