Application of Rough Set Theory in Fault Diagnosis of Roller Bearing

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.