Using vibration signals for bearing fault diagnosis can generally achieve good diagnostic results. However, it is not suitable for practical industrial applications due to the restricted installation and high cost of vibration sensors. Therefore, the easily obtainable motor current signal (MCS) has received widespread attention in recent years. Meanwhile, traditional fault diagnosis methods cannot meet the diagnostic accuracy requirements because of the low signal-to-noise ratio (SNR) of the MCS. Committed to achieving bearing fault diagnosis through MCS, a rolling bearing fault diagnosis method, ISCV-ViT, based on the MCS and the Vision Transformer (ViT) model, is proposed. In particular, a signal processing method based on the instantaneous square current value (ISCV) is proposed to process the MCS directly obtained through a frequency converter into time-domain images. Then, the ViT model is applied for bearing fault diagnosis. Finally, experimental verification is carried out based on the public bearing dataset of Paderborn University (PU) and the bearing dataset of Shenzhen Technology University (SZTU). The analysis of the experimental results demonstrates that the average accuracy of the ISCV-ViT for the two datasets is up to 96.60% and 94.87%, respectively.

Existing methods cannot satisfy the reliability allocation demands of the early design phase for the modern complex system of NC turrets. Motivated by the need of practical application, this paper proposes a new and practical reliability allocation method in the early design stage for NC turrets considering failure mode and system complexity, information inaccessibility, and expert knowledge limitation. First, the fault tree of a NC turret is quickly built to clear the relationship between the system’s compositions and failure modes up. Second, the happening probability of each failure event in the fault tree is firstly calculated by fuzzy expert evaluation to provide the reliability allocation with complete information. Third, by discussing the practical meaning of every layer in the fault tree, the proposed allocation strategy is within the experts’ knowledge scope for evaluating accurately. Eventually, the application result of the AK36100 A turret is presented and compared with some existing allocation methods, illustrating the rationality of the proposed allocation method.

To realize the accurate performance test under high‐speed operation and the long‐term stable reliability test of the motorized spindle, a reliability test rig (RTR) which can simulate the cutting force in the actual machining process is presented. Firstly, a reasonable prototype integrating dynamic force loading devices and torque loading devices is designed and established based on the load analysis of the spindle, and a complete and explicit control strategy of the reliability loading test is designed. Secondly, the effects of misalignment of the diaphragm coupling caused by assembling are analyzed, and experiments are conducted to test the axis orbit of the motorized spindle. The experimental results illustrate that the axis orbit can identify the occurrence of misalignment, which ensures timely adjustment of misalignment and the accurate performance test under high‐speed operation. Lastly, a damper is added in the mechanical structure of the electrohydraulic servo loading system (EHSLS), and the comparison of Bode maps before and after optimization is analyzed by the Nyquist criterion. After the optimization, the gain and phase margin of the Bode diagram are 12.9 dB and 57.2°, respectively, which are both within the stable range and validate the improvements on the ability of long‐term reliability tests of the motorized spindle. The presented RTR is able to simulate the actual cutting force and provides an efficient loading approach to guarantee the accuracy and stability of motorized spindle tests.

The lack of labeled data and variable working conditions brings challenges to the application of intelligent fault diagnosis. Given this, extracting labeled information and learning distribution-invariant representation provides a feasible and promising way. Enlightened by metric learning and semi-supervised architecture, a triplet-guided path-interaction ladder network (Tri-CLAN) is proposed based on the aspects of algorithm structure and feature space. An encoder–decoder structure with path interaction is built to utilize the unlabeled data with fewer parameters, and the network structure is simplified by CNN and an element additive combination activation function. Metric learning is introduced to the feature space of the established algorithm structure, which enables the mining of hard samples from extremely limited labeled data and the learning of working condition-independent representations. The generalization and applicability of Tri-CLAN are proved by experiments, and the contribution of the algorithm structure and the metric learning in the feature space are discussed.