In the field of fault diagnosis, a large number of sensors are used to reflect the working status of equipment with the increasing complexity of mechanical systems. However, collecting enough fault monitoring data, which leads to the widespread problem of data imbalance and continues effect on the improvement of fault diagnosis accuracy, is difficult for some devices in the actual work process. To improve the performance of fault diagnosis under imbalance, firstly, this paper proposes a generative model based on generative adversarial networks (GAN), the first stage of which uses Time series GAN to overcome the problem wherein traditional GAN cannot effectively extract the time–series information of sequence data, and the second stage incorporates the characteristics of supervised learning of Auxiliary Classifier GAN to solve the problem wherein it cannot generate multi-mode samples. Secondly, this paper also studies the fault diagnosis method based on two stage GAN (2S GAN) and applies the classification model combining continuous wavelet transform and ResNet18 to the field of mechanical fault diagnosis. The method includes the data pre-processing stage, data augmentation and model training stages. Finally, this method is validated by two datasets, and the results show that the absolute increments of the accuracy of this method are improved by 5.70%, 6.34%, 9.08% and 16.35% in case 1 under four imbalance ratios and by 3.36%, 4.65%, 15.93% and 26.34% in case 2. Thus, it provides an effective solution to the fault diagnosis problem in the data imbalance state.

The functional parts of a machine tool determine its reliability level to a great extent. The failure prediction of the functional part is helpful to prepare the maintenance scheme in time, in order to ensure a stable manufacturing process and the required production quality. Due to the rise of digital twin (DT), which has the characteristics of virtual reality interaction and real-time mapping, a DT-based real-time prediction method of the remaining useful life (RUL) and preventive maintenance scheme is proposed in this study. In this method, a DT model of the manufacturing workshop is established based on real-time perceptual information obtained by the proposed acquisition method. Subsequently, the real-time RUL of the functional part is predicted by establishing an RUL prediction model based on the nonlinear-drifted Brownian motion, which takes the working conditions and measurement errors into consideration. On this basis, the optimal preventive maintenance scheme can be determined and fed back to the manufacturing workshop, in order to guide the maintenance of relevant parts. Finally, an example case study is presented to illustrate the feasibility and effectiveness of the proposed method.

Owing to the competitive advantages of fast response speed, large pushing force, high reliability, and high precision, the permanent magnet linear synchronous motor (PMLSM) has played an increasingly vital role in various high-speed and high-precision control systems. However, PMLSM exhibits nonlinear behavior in actual operation, and position tracking precision is negatively affected by friction, load changes, and other external disturbances. To meet the growing demand and solve the position tracking control problem for the PMLSM, the control system is critical. Sliding-mode control (SMC) has been used extensively in nonlinear control systems due to its superior performance characterized by simplicity, good dynamic response and insensitivity to parameter perturbation and external disturbances, and has been implemented in PMLSMs to track practical position. The objective of this article is to classify, scrutinize and review the major sliding-mode control approaches for position control of PMLSM. The three different conventional SMC methods, namely the boundary layer approach, the reaching law approach and the disturbance observer-based SMC, are discussed in detail. The four advanced forms of SMC, namely terminal SMC, super-twisting SMC, adaptive SMC and intelligent SMC, are also presented. A comparison of these approaches is given, in which the advantages and disadvantages of each approach are presented; additionally, they are presented in table form in order to facilitate reading. It is anticipated that this work will serve as a reference and provide important insight into position control of PMLSM systems.

Remaining useful life (RUL) prediction of rolling bearings plays a crucial role in ensuring safe operation and maintenance decisions for equipment. However, due to the influence of monitoring location and working conditions, traditional deep learning methods are challenging to extract multi-dimensional and multiscale degradation features, decreasing the accuracy of RUL prediction. At the same time, there are uncertainties, such as noise and model parameters, which makes it difficult for RUL’s point prediction to meet maintenance requirements. A framework for bearing RUL interval estimation based on a cascaded multi-scale convolutional neural network (CMS-CNN) module is proposed. Firstly, depthwise separable convolution (DSC) and dilated causal convolution (DCC) constitute the main framework of the CMS-CNN module in the form of a cascade to realize multi-dimensional degenerate feature extraction in space and time. The convolution operation with different dilation rates is introduced into the module to achieve multi-scale feature extraction, and the convolutional block attention module (CBAM) is embedded to adaptively assign the importance of features. In addition, a staged-optimized mean-variance two-branched interval estimation output network layer is constructed to quantify the uncertainty of bearing RUL prediction results. Finally, the method is verified with two rolling bearing datasets. Experimental results show that the proposed method not only has high RUL prediction accuracy, but also accurately gives the uncertainty interval of the prediction results, which is better than some advanced prediction methods.

As the key functional component of the machine tool, the reliability test of the spindle is necessary to verify the reliability of the machine tool. In the reliability test, the cutting load model is the guideline of workload simulation and is the prerequisite to ensure the accuracy and effectiveness of the long-term experiment. However, the existing load models usually aim at the specific cutting force at the tool-tip, thereby ignoring the versatility, maneuverability, and accuracy of the load model when applied in the spindle reliability test. In this study, a general cutting load model for the machine tool spindle is established in a form of radial-axial-torque decomposition, and the radial force is simplified as non-rotating status for the maneuverability of conducting a load simulation. The difference between rotating and non-rotating radial force on the reliability calculation is also discussed and corrected using bearing fatigue analysis. A spindle reliability test platform with radial force, axial force, and torque simulation is developed according to the cutting load model, while the loading spectrum is compiled for conducting the spindle reliability test. This research is of great engineering value for the designing of the spindle reliability test.

The rotating performance of motorized spindles is critical to the dimensional accuracy and surface quality of the parts to be machined. This paper proposes a novel measurement strategy and an evaluation method of spindle’s rotating performance, considering the loads imposed on the spindle. To be specific, cutting loads are simplified as a unified function to guide the load simulation, and the load-induced error is eliminated using a position adjustment procedure. In a certain speed-load combination, the spindle’s rotating performance under the unified load function is measured using a data envelope method. To extend the assessment to all typical work conditions, an evaluation scheme for different speed and load combinations is proposed based on a visual ideal of Multi-Zones Error Map (MZE Map). The necessity of considering the load-dependent rotating performance is demonstrated via several tests conducted on proposed load simulation platform. Besides, the procedures for developing the MZE Map are performed, and the results express an overview relationship between rotating performance and different speed-load conditions. The proposed method provides a new approach to evaluate the rotating performance considering the actual load conditions. This load-dependent oriented evaluating strategy contributes to a more precise health evaluation and management method for motorized spindles in actual machining process.