The application of time domain and frequency domain statistical factors on rolling bearing performance degradation assessment

Yujing Wang^{1, 2}, Yicheng Jiang¹, Shouqiang Kang²

¹School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001

 ²School of Electrical and Electronic Engineering, Harbin University of Science and Technology, Harbin 150080

Rolling bearing performance degradation assessment is a predict and prevent technology. In order to assess the performance degradation degree of the rolling bearing, and make the time domain and frequency domain statistical factors be applied more effectively in rolling bearing performance degradation assessment, a comprehensive analysis method is proposed based on time domain and frequency domain statistical factors. Time domain and frequency domain statistical factors are calculated and analysed for the life cycle data of the rolling bearing. Outer raceway moderate fault and severe fault of the rolling bearing can be distinguished well by peak-to-peak level, the root-mean-square (RMS) value, and kurtosis value of time domain factors; normal state and mild fault can be distinguished better by frequency centroid, F3, F4 and F5 of frequency domain factors than each time domain factor. The outer raceway performance degradation condition of the rolling bearing can be monitored well by using the proposed comprehensive analysis method, which uses partly frequency domain factors to analyse mild fault and partly time domain factors to analyse moderate fault and severe fault.