粗糙集理论在滚动轴承故障诊断中的应用

Application of Rough Set Theory in Fault Diagnosis of Roller Bearing

粗糙集是一种软计算方法,它不需要任何先验知识和理论推导,仅依赖于原始数据,从数据中发现潜在规律和隐含规则,其核心任务是数据约简,得到最简决策规则。文章基于粗糙集理论的可辨识矩阵算法,对滚动轴承故障信息表进行属性约简和属性值约简,得到了精简的诊断规则,从而方便了信息的机器处理,也为基于规则的滚动轴承智能故障诊断系统提供了一种更为简单的知识库自动构造方法。

Rough set is a soft computing method which needn＇t any validated knowledge and theoretics deducing,just depends on original data,and it aims at discovering latent laws and rules.Its core task is reducing data in order to get the simplest decision rules.Based on the discernibility matrix of rough set theory,the attribute and attribute value of fault information table of roller bearing were reduced and the simpler diagnosis rules of fault diagnosis were achieved,which was good for computer dealing.This method provides a simpler method of automatic construction of knowledge base for intelligent fault diagnosis system of roller bearings based on rules.