A fault feature reduction method based on rough set attribute reduction and principal component analysis
Qiang Huang, Jian Wang, Haixia Su, Yang Lu, Zhaoping Ding, Guigang Zhang · 2016
Recently, precise diagnosis of faults is increasingly taken seriously, and the fault feature reduction is one of the key technologies to carry out accurate and reliable diagnosis. In this paper, a feature reduction method based on rough set attribute reduction and principal component analysis is proposed. Firstly the rough set attribute reduction is used to remove the irrelevant features, and then the principal component analysis is adopted to further reduce the features. Finally, the validity of the method is verified by the aero engine rotor fault data. Experimental results show that the proposed method can not only improve the accuracy of fault diagnosis, but also reduce the number of fault features and improve the diagnostic efficiency.