Strong noise in practical engineering environments interferes with the signal of a rolling bearing, which leads to the decline of the diagnosis accuracy of intelligent diagnosis models. This paper proposes a novel hybrid model (a convolutional denoising auto-encoder (CDAE)-BLCNN) to address this problem. First, the rolling bearing vibration signal containing noise was input into the CDAE, which denoises the signal through unsupervised learning and then outputs the reconstructed data. Secondly, a hybrid neural network (BLCNN), composed of a multi-scale wide convolution neural network and a bidirectional long-short-term memory network, was used to extract intrinsic fault features from the reconstructed signal and diagnose fault types. The analysis results demonstrate that the proposed hybrid deep-learning model achieves higher detection accuracy, even under different noise levels and various rotating speeds. Compared with other models, there is a high fault recognition rate, robustness, and generalization ability, which may be favorable to practical applications.

The effects of the uncertainty of angular misalignment on the wear of angular contact ball bearings in the spindle system are investigated. The wear model of bearing raceways is derived based on the Archard model. The angular misalignment caused by mounting error is assumed to comply with normal distribution, Monte Carlo method is used to extract the angular misalignment sample. Based on the spindle system model, the spinning speed, pitch angle, and contact force of ball elements in each angular contact ball bearing are calculated. The wear depth of bearing raceways is analyzed. The results show that the uncertainty of the initial angular misalignment leads to significant dispersion of wear depth of ball bearings, but the dispersion degree is related to spindle speed, angular misalignment cases, and bearing mounting position. Increasing the spindle speed increases the wear depth but decreases the wear depth dispersion.

The application feedback on existing silicon nitride ceramic bearings and RCF experimental research all indicate that the primary failure mode of silicon nitride ceramic bearings is material spalling on the contact surface. Spalling failure occurs due to the initiation and propagation of cracks under rolling contact. However, silicon nitride ceramic bearings, owing to their unique manufacturing method, inevitably exhibit defects and cracks. Therefore, as silicon nitride ceramic bearings are increasingly prevalent, reducing the probability of spalling failure is crucial for extending their service life. This can only be achieved by gaining a clear understanding of the crack initiation and expansion mechanisms in silicon nitride ceramic bearings. This paper is based on silicon nitride rolling friction experiments. It involves the joint simulation of Franc3D-V8.4 and ABAQUS2020, wherein the crack front SIFs are calculated for each load contact position of the surface crack on the silicon nitride ceramic bearing ring during cyclic movement. The study also delves into the determination of the maximum effective stress intensity factors and explores the influence of the initial crack depth on the cycle life and direction of crack propagation. The research yields several valuable conclusions. The findings of this research offer theoretical guidance for formulating grinding technologies for silicon nitride rings and adjusting and controlling working parameters of silicon nitride ceramic ball bearings. These insights are crucial for enhancing the reliability and longevity of silicon nitride ceramic bearings in practical applications.

Based on the difficult machining characteristics of silicon nitride materials, this manuscript focuses on optimizing the precision machining process of silicon nitride bearing components and improving the machining efficiency and quality of silicon nitride bearing components. Firstly, the mechanism of crack formation and propagation in hard and brittle materials under the action of abrasive particles is discussed in this paper, and based on this, a kinematic model of single abrasive grain cutting on hard and brittle materials is established. The effect of workpiece linear velocity, grinding wheel linear velocity, grinding wheel oscillation velocity and feed velocity on inner surface roughness of Si₃N₄ ring was discussed through grinding test. The experimental results show that the surface roughness can reach about Ra0.20 ∼Ra0.33 μm through a large number of grinding experiments by adjusting the combination of process parameters μm. On this basis, use the optimized process parameters calculated by the surface quality prediction model constructed in this manuscript to conduct grinding test again, and the surface roughness value reaches about Ra0.19 ∼Ra0.23 μm. The purpose of optimizing the process parameters is realized. Finally, the surface roughness can be further reduced and maintained at Ra0.05 ∼Ra0.06 μm by further superfinishing the surface after the process optimization with an oilstone. The research work of this manuscript realized the optimization of precision machining process of silicon nitride bearing ring and the rapid optimization of machining process through the established prediction model, which improved the precision machining efficiency of ceramic bearing components. Through the study of this manuscript, a process route of controllable processing of inner surface quality of Si₃N₄ ring is formed, from preliminary selection of process parameters by optimization model to precision grinding and then to ultra-finishing of whetstone, which provides theoretical reference for efficient and precise manufacturing of practical bearing components.