Fault Diagnosis of Hydraulic Pump Based on Rough Set and PCA Algorithm

In order to analyze the fuzzy and low signal-noise ratio on fault characteristics of hydraulic pump, a new rough set (RS) fault diagnosis algorithm based on principal component analysis (PCA) as guide is proposed. Firstly, the algorithm uses wavelet analysis to process noise abatement so as to extract the effective fault characteristic. Secondly, it utilizes PCA method to make dimension reduction and decoupling relativity between these characteristics. Thirdly, it establishes diagnosis rules knowledge base through the rough set theory. Lastly, the validity of this method is verified through the simulated experiment on doffing-shoe fault diagnosis of hydraulic pump.