油液故障诊断系统知识发现的包含度方法  

INCLUSION DEGREE METHOD OFKNOWLEDGE DISCOVERY ON OILFAULT DIAGNOSIS SYSTEM

作　　者：王金涛[1] 谢友柏[1] 

机构地区：[1]西安交通大学润滑理论及轴承研究所,西安710049

出　　处：《机械工程学报》2003年第9期58-61,65,共5页Journal of Mechanical Engineering

基　　金：国家自然科学基金(59990470;59990472)

摘　　要：油液故障诊断系统的知识发现本质上是一个模式分类与识别的问题,结合Rough set理论和摩擦学系统的特点,建立了油液故障诊断系统知识发现模型。通过将包含度方法引入到Rough set理论中,得到基于包含度的粗糙集知识约简和发现方法,并对包含度与粗糙集数据分析中的度量之间的关系进行了分析。最后通过一个9属性柴油机油液诊断系统的实例来说明这种方法的有效性。Knowledge discovery on oil fault diagnosis system (OFDS) is essentially a problem of pattern classifier and recognition. Based on Rough set theory and tribology system, knowledge discovery model of OFDS is proposed. By applying the concept of inclusion degree to Rough set theory, method of knowledge reduction and discovery based on inclusion degree is deduced, and the relation between the inclusion degree and measures of rough set data analysis is also discussed. In the end an application example of diesel engine OFDS with 9 attributes is given to show the effect of this method.

关 键 词：油液分析 故障诊断 ROUGH set 包含度 知识发现 摩擦学系统 

分 类 号：TH117[机械工程—机械设计及理论]