In this paper, Fe-based and Co-based alloy powders were chosen to perform laser cladding on wheel materials through conventional laser cladding (CLC) and ultra-high-speed laser cladding (UHSLC) processes, respectively. The microstructures, element distribution, phase composition and hardness of the Fe-based alloy and Co-based alloy coating layers using the CLC and UHSLC processes were compared and analysed. The results show that the CLC and UHSLC alloy coatings were dense and free of defects such as pores and cracks. Compared with the CLC alloy coating, the grain size of the UHSLC alloy coating was smaller, the coating composition was close to the powder design composition, and the distribution of Cr within and between the grains was more uniform. The Fe-based coating was mainly composed of (Fe, Ni) and Cr₇C₃, and the Co-based coating was mainly composed of γ-Co and Cr₂₃C₆. It was found that the cooling rate of the CLC alloy coating was smaller than that of the USHLC, and the hardness of the CLC alloy coating was less than that of the USHLC. The average hardness of the UHSLC Fe-based and Co-based alloy coatings was 709 HV and 525 HV, respectively. The average hardness of the CLC Fe-based and Co-based alloy coatings was 615 HV and 493 HV, respectively. The rolling friction and wear tests were carried out with the CLC-treated and UHSLC-treated wheel specimens on the GPM-30 rolling contact fatigue testing machine. The results showed that the wear rate of the UHSLC alloy coating on the wheel specimens was significantly lower than that of the CLC alloy coating on the wheel specimens. The wear rates of the UHSLC Fe-based and Co-based alloy coatings on the wheel specimens were reduced by 40.7% and 73.8%, respectively. It was demonstrated that the wear resistance of the USHLC alloy coatings was better than those of the CLC alloy coatings. The CLC alloy coating exhibited more severe fatigue damage with small cracks. Furthermore, the damage of the UHSLC alloy coating was relatively minor, with slight spalling. The Co-based alloy coating exhibited superior wear properties with the same laser cladding process.

Turnouts are the weak spot in high-speed rail systems, and it is simple for the phenomenon of the wheel–rail force and the carbody lateral acceleration over-limit to arise when the train passes through, which affects the service life of the rail and the running stability of the train. In this paper, the turnout with wheel–rail force over-limit and carbody lateral acceleration over-limit is selected for analysis, and the profiles of the wheel and rail are monitored. Then, the vehicle–turnout coupled multi-body dynamics model is simulated. Additionally, the portable vibration analyzer, the comprehensive inspection train, and the wheel–rail contact dynamic stress tester monitors the data and evaluates the impact of rail grinding on high-speed railway. The results of this study demonstrated that the turnout profiles are in good agreement with the standard wheel profiles following grinding, and the wheel–rail contact point and equivalent conicity both improved. When the train passes the ground turnout at high speed with and without the wheel polygonal wear, the wheel–rail force and the carbody acceleration were clearly improved. Using the wheel–rail contact dynamic stress tester, the comprehensive inspection train, and the portable vibration analyzer monitoring the changes in the carbody acceleration, the wheel–rail force and the carbody acceleration are definitely better after grinding. Similar to the pattern in the simulation, the train’s running steadiness increased by grinding.

Traditional methods of gearbox fault diagnosis rely heavily on manual experience. To address this problem, our study proposes a gearbox fault diagnosis method based on multidomain information fusion. An experimental platform consisting of a JZQ250 fixed-axis gearbox was built. An acceleration sensor was used to obtain the vibration signal of the gearbox. Singular value decomposition (SVD) was used to preprocess the signal in order to reduce noise, and the processed vibration signal was subjected to short-time Fourier transform to obtain a two-dimensional time–frequency map. A multidomain information fusion convolutional neural network (CNN) model was constructed. Channel 1 was a one-dimensional convolutional neural network (1DCNN) model that input a one-dimensional vibration signal, and channel 2 was a two-dimensional convolutional neural network (2DCNN) model that input short-time Fourier transform (STFT) time–frequency images. The feature vectors extracted using the two channels were then fused into feature vectors for input into the classification model. Finally, support vector machines (SVM) were used to identify and classify the fault types. The model training performance used multiple methods: training set, verification set, loss curve, accuracy curve and t-SNE visualization (t-SNE). Through experimental verification, the method proposed in this paper was compared with FFT-2DCNN, 1DCNN-SVM and 2DCNN-SVM in terms of gearbox fault recognition performance. The model proposed in this paper had the highest fault recognition accuracy (98.08%).

With the continuous expansion of road networks and the rise in railway transportation capacity, the scale of locomotive crew needs has increased sharply. However, the current training of locomotive crews is inefficient and costly and cannot meet the needs of the industry. This project adopted virtual reality technology to develop the driving and emergency skills of harmonious electric locomotive crews. The trainees can learn the composition and structure of the locomotive, master the working principles of the locomotive and receive training in relevant emergency skills in the virtual environment. After completion of the system development, we carried out a series of research studies and experiments; the results show that the use of this system to train personnel can lead to them quickly mastering the practical training content. This new training mode can effectively solve the outstanding problems with the current training system for locomotive crews.