A Fault Diagnosis Approach for Rolling Bearings Based on Enhanced Blind Equalization Theory

Fault diagnosis of rolling bearings remains a very important and difficult research task in engineering and technique. After analyzing the shortcoming of current bearings fault diagnosis technologies, a novel enhanced blind equalization (BE) technology based on wavelet packet (WP) analysis and eigenvector algorithm (EVA) was proposed to extract directly impacting features and diagnose bearings´ faults in this paper. First, the blind equalization model and algorithm of impacting signal processing of rolling bearings were established based on the BE theory and EVA algorithm. Then, the WP theory and method are applied to the model and algorithm. After these, the enhanced signal processing and fault diagnosis algorithm based on WP and EVA is presented, and the shortcomings are fixed. Finally, the built model and algorithm were applied to two impact experiments and two real engineering data for verification. The results show that the method is very effective in extracting the impacting features and intelligent fault diagnosis for rolling bearings.