Recurrence Plot Quantitative Analysis-Based Fault Recognition Method of Rolling Bearing

Abstract

Aiming at the problem that rolling bearings are widely used but have high failure rate. A fault diagnosis method combined with local projective noise reduction method and recurrence plots quantification analysis for rolling bearings is presented. The fault vibration signals of rolling bearings are taken as the analysis objects. Firstly, the vibration signal is denoised by local projective noise reduction method. Then, the recurrence plots of denoised vibration signals are drawn. The system dynamic behavior reflected in the recurrence plots is extracted by recurrence quantification analysis. The determinism (DET) and entropy (ENTR) are selected to form the characteristic vectors. Finally, the characteristic vectors of the training samples are clustered by kernel fuzzy C-means (KFCM) clustering method. The minimum Euclidean distance principle is used to identify the test samples. Comparing the recurrence plot quantification analysis method with the quantitative feature method of phase space complex network. The results show that the recurrence plots quantification analysis-based fault diagnosis method of rolling bearings has a higher diagnosis rate.