Motor bearings are prone to different degrees of performance degradation, fatigue damage and failure undergoing complex and harsh environments. Vibration signal analysis is a mature method for diagnosing motor bearing faults, while it is not applicable for installing additional vibration sensors on many occasions. Practically, the fault of motor bearings changes the air gap flux between the rotor and stator, which leads to harmonic fluctuations in the stator current. The current signals can be used to diagnose the motor bearing faults without additional sensors. Inevitably the harmonics caused by the motor bearing faults will be coupled with the original signals. This paper combines bi-spectrum and Convolution Neural Network (CNN) to analyze the current signals of motor bearing faults. The CNN diagnosis model is trained based on the local bi-spectrum of current, and the CNN parameters are optimized. Diagnose and analyze motor bearing faults with different fault implantation methods, working conditions, fault degrees and fault locations. The diagnostic accuracy reaches more than 80%.

Bearing is the critical basic component of rotating machinery and its remaining life prediction is very important for mechanical equipment’s smooth and healthy operation. However, fast and accurate bearing life prediction has always been a difficult point in industry and academia. This paper proposes a new strategy for bearing health assessment based on a model-driven dynamic interval prediction model. Firstly, the mapping proportion algorithm is used to determine whether the measured data are in the degradation stage. After finding the starting point of prediction, the improved annealing algorithm is used to determine the shortest data interval that can be used for accurate prediction. Then, based on the bearing degradation curve and the information fusion inverse health index, the health index is obtained from 36 general indexes in the time domain and frequency domain through screening, fusion, and inversion. Finally, the state space equation is constructed based on the Paris-DSSM formula and the particle filter is used to iterate the state space equation parameters with the minimum interval data to construct the life prediction model. The proposed method is verified by XJTU-SY rolling bearing life data. The results show that the prediction accuracy of the proposed strategy for the remaining life of the bearing can reach more than 90%. It is verified that the improved simulated annealing algorithm selects limited interval data, reconstructs health indicators based on bearing degradation curve and information fusion, and updates the Paris-DSSM state space equation through the particle filter algorithm. The bearing life prediction model constructed on this basis is accurate and effective.