Fault diagnosis is an important means to ensure the safe and reliable operation of mechanical equipment. In machinery fault diagnosis, collecting and mining the potential fault information of the vibration signal is the most commonly used method to reflect the operating status of the equipment. In engineering scenarios, in the face of rotating machinery with variable speed, simple time domain analysis or frequency domain analysis is difficult to solve the problem. The time-frequency analysis technology that combines time-frequency transformation and data analysis can solve practical engineering problems by capturing the transient information of the signal. At present, a large number of related literatures have been published in academic journals. This paper hopes to provide convenience for relevant researchers and motivate researchers to further explore by summarizing the published literature. First, this paper briefly explains the concept of time-frequency analysis and its development. Then, the time-frequency transformation method proposed for the characteristics of rotating machinery fault vibration signal and related works of literature are reviewed, and the key issues of the application of time-frequency transformation method in rotating machinery fault diagnosis are discussed. Next, this paper summarizes the relevant literature on the combination of data analysis technology and time-frequency transformation and sorts out its development route and prospects. The study reveals that time-frequency analysis technology is able to detect the rotating machinery fault effectively. The time-frequency analysis technology has made abundant achievements in the field of rotating machinery fault diagnosis. It is expected that this review would inspire researchers to explore the potential of time-frequency analysis as well as to develop advanced research in this field.

A turnout switch machine is key equipment in a railway, and its fault condition has an enormous impact on the safety of train operation. Electrohydraulic switch machines are increasingly used in high-speed railways, and how to extract effective fault features from their working condition monitoring signal is a difficult problem. This paper focuses on the sectionalized feature extraction method of the oil pressure signal of the electrohydraulic switch machine and realizes the fault detection of the switch machine based on this method. First, the oil pressure signal is divided into three stages according to the working principle and action process of the switch machine, and multiple features of each stage are extracted. Then the max-relevance and min-redundancy (mRMR) algorithm is applied to select the effective features. Finally, the mini batch k-means method is used to achieve unsupervised fault diagnosis. Through experimental verification, this method can not only derive the best sectionalization mode and feature types of the oil pressure signal, but also achieve the fault diagnosis and the prediction of the status of the electrohydraulic switch machine.

Deep learning has the characteristics of high efficiency, high accuracy and low knowledge dependence, and has become a hot method in the research field of rotating machinery fault diagnosis. The time–frequency transform can show both the time and frequency characteristics of the vibration signal, so the time–frequency image is often used as the input of deep learning networks. At present, there are many time–frequency transform methods, and how to choose one is a problem worth discussing. This paper proposed a time–frequency image quality evaluation method based on improved local interpretable model-agnostic explanations (LIME). With the input of deep learning networks as the application background, this method evaluates the time–frequency image quality of rotating machinery vibration signals from two aspects: the accuracy of the diagnosis results and the consistency of the interpretation results with prior knowledge. The feasibility of the proposed evaluation method is verified by experiments on the measured data set, and engineers’ trust in the deep learning model is improved.

With the great leaps forward in the development of the railway, the importance of safe operation and maintenance has become increasingly prominent. Varistor is the key component insurge protective devices for railway communication and signaling equipment, it is necessary to study a description model of the varistor degradation process and predict its lifetime through condition monitoring. Among the monitoring parameters, nonlinear coefficient is an important index to measure the health of varistors. Considering that the degradation process of varistors is the cumulative effect of surge history, and its nonlinear coefficient has a time–series relationship, a life prediction model for varistors based on BiLSTM is proposed. The model innovatively uses nonlinear coefficient as the characterization of the deterioration degree of varistors and uses a small-scale network to predict the development trend of nonlinear coefficient automatically and accurately. Verified by surge impact experiments, the model can accurately predict the state of nonlinear coefficient according to historical data and has the potential for engineering applications in predicting the life of varistors.

This paper proposes a fault diagnosis method for rotating machinery based on evolutionary convolutional neural network (ECNN). With the time-frequency images as the network input, with the help of the global optimization ability of the genetic algorithm, the structure of the convolutional neural network can evolve autonomously, and the adaptive configuration of the structural hyperparameters for the target task is realized. In this paper, the proposed method is verified by the measured signal of the planetary gearbox. The results show that the proposed method is helpful to obtain a convolutional neural network structure with better performance and achieve higher fault diagnosis accuracy.

Rotating machinery is an important part of mechanical system. Effective fault diagnosis of rotating machinery is of great significance to the whole operation and maintenance of mechanical system. The effectiveness of fault diagnosis for rotating machinery is the comprehensive embodiment of the accuracy of diagnosis results and the reliability of diagnosis methods. Since the 21 ^st century, in the field of rotating machinery fault diagnosis, a large number of studies have made gratifying achievements in improving the accuracy of diagnosis results with the help of deep learning. However, there are few studies on the reliability of diagnostic methods. In this paper, a deep convolutional network based on visual optimization is designed for fault diagnosis of rotating machinery. Class activation mapping (CAM) technology is used to observe the area of concern of the neural network and provide feedback with loss function to ensure that the neural network can correctly focus on the effective information in the image. With this optimization method, the reliability of diagnosis is improved. Through repeated test verification of the measured data, it is found that the proposed method improves the accuracy and stability of the diagnostic results without changing the network structure, which verifies the effectiveness of the proposed method.