The effective fault diagnosis of bearing can guarantee the safety of rotating machinery and is very important for its stable operation. The information fusion of multi-sensor data has been a feasible method to enhance the performance of fault diagnosis. However, how to fuse the joint information from different channels or even different kinds of sensors is still an important challenge. The present study proposes a novel multi-sensor information coupling network (MICN) for bearing fault diagnosis, which handles the signals from the same or different types of sensors, and the deeper features can be extracted from multi-sensors independently and simultaneously fused layer by layer. Especially, during the multi-layer feature fusion process, a novel feature-level information coupling model is developed based on the mutual attention mechanism. Finally, to validate the efficiency of the proposed method, several different experiments are designed, and the results show validity and superiority.

Bearing preload has a great impact on the service performance of high-speed spindle, such as the temperature rise and stiffness. Because of the flatness error of spacer, the axial parallelism error of the spindle, the nonuniform heat distribution of ball bearing and the external loads, the outer ring of the bearing is often subjected to nonuniform distributed loads, which will result in the misalignment between outer ring and spindle. In such conditions, traditional uniform preload method may not meet the requirements for bearing temperature rise and spindle rotation accuracy. In this paper, a nonuniform preload is presented which not only includes the preload magnitude in traditional way, but also involves the preload distribution nonuniformly. Firstly, an equivalent transformation for nonuniform preload is built to simplify the nonuniform preload acting on the outer ring. Then based on the Hertz contact theory and ball bearing geometry theory, a five degrees of freedom quasi-static model of ball bearing under nonuniform preload is established. On the basis of this model, internal load and other contact parameters in bearing are calculated and the heat mechanism of ball bearing is analyzed. Finally, according to the heat generation model under nonuniform preload which include both preload magnitude and distribution, effects of both uniform and nonuniform preload on bearing heat and its distribution are investigated and compared, and the contact forces in bearing are then analyzed. The results show that compared with uniform preload, proper nonuniform preload could effectively improve the contact status between balls and rings, decrease the heat generation rate, and reduce the excessive local heat in ball bearing under practical working conditions.