灰关联分析及其在装备故障诊断中的应用  

Grey Relevancy Analysis and Its Application to Equipment Fault Diagnosis

作　　者：董立立[1] 朱煜[1] 黄道[2] 顾幸生[2] 

机构地区：[1]清华大学精密仪器与机械学系,北京100084 [2]工业自动化国家工程研究中心华东理工大学分部,上海200237

出　　处：《华东理工大学学报（自然科学版）》2008年第4期563-567,597,共6页Journal of East China University of Science and Technology

基　　金：上海市科委重大科技攻关项目(04dz11008);国家自然科学基金(50577037)

摘　　要：灰理论是研究和解决信息不完备系统的一种有效方法,具有"小样本、贫信息、不确定性"问题处理能力强、计算复杂度低等优点。实际工程中变速箱系统的故障具有复杂的灰特性,通过对灰理论中的灰关联原理、灰关联系数与分辨系数的关系、以及分辨系数的性质和取值原则的研究,成功地将灰关联分析应用于变速箱系统的故障诊断。实例结果表明:灰关联故障诊断方法计算简单,诊断结果与定性分析结论一致;能够为装备维修及状态预测提供理论指导;且对样本数量及其分布规律没有特殊要求,便于形成知识库中的规则,实用价值较大。Grey theory is an effective method for studying and resolving the deficient information systems, and has advantages of simplified computation and effectively dealing with the problems with uncertainty, small samples and lack of information. The running gearbox system has complicated grey characteristics. Grey relational principle, the relationship between the grey relational coefficient and the distin guishing coefficient, the property and the selection principle of the distinguishing coefficient are studied, and grey relevancy analysis theory is successfully applied to the fault diagnosis problems of gearbox system, which have complicated grey characteristics. Results of the practical application show that the fault diagnosis method based on grey relevancy analysis theory has the advantages of simplified computation, consistency to the quality analysis result. It provides theoretical guidance for the prediction of the running state or maintenance of the equipment. Furthermore, it has no special requirements for the quantity and distribution of samples, and is convenient for form knowledge base rules, and has great practical value.

关 键 词：灰关联度 分辨系数 灰关联分析 故障诊断 

分 类 号：TQ050.7[化学工程]