Study on Rolling Bearing Performance Degradation Method Based on Sequential Extension

Aiming at the problem of rolling bearing performance degradation method research, the rolling bearing performance degradation method research based on time series extension is presented in this paper. The characteristics of vibration signals are firstly extracted by using the autoregressive (AR) models; and then the most-value normalization treatment of the obtained features are conducted; then, the normalized features are scored and dimensionalized by Fisher comparison. Finally, the dimensionalized feature vectors are input into the extenics model for qualitative and quantitative evaluation of bearing performance. Accuracy of the conclusions is verified through experiments and the envelope spectrum analysis; the experiments show that the proposed method can effectively detect the early faults.