Aiming at the problem that rolling bearings are prone to failure in large mechanical equipment, an e- valuation method for rolling bearings' performance degradation is proposed, which combines the complete empir- ical mode decomposition (CEEMDAN) of adaptive noise with the grey correlation analysis. Firstly, by using CEEMDAN to decompose the vibration signal of the whole life time of bearing, the energy entropy feature was obtained. Then, by taking the characteristic of the normal vector as reference variables of gray correlation analy- sis, the correlation degree was calculated between the feature vector and the normal vector of the bearing’s whole life time as a quantitative evaluation index in the process of performance degradation. The results show that the method can figure out early failure and describe the bearing degradation at each stage. Finally, CEEM- DAN and Hilbert envelope demodulation were used to verify the validity of the evaluation results.