A wear prediction model is built to research the wear of the curved switch rail in a high-speed turnout. The Archard wear model is used in the wear prediction model to analyze the profile evolution law. The non-Hertzian contact Kik–Piotrowski method based on virtual penetration is used as the contact algorithm for the Archard wear model. A dynamic model of the vehicle–curved switch rail system based on the predicted profiles of the curved switch rail and the measured wheel profiles with different stages is established. The effect of the wheel and curved switch rail profiles’ wear on vehicle dynamic performance is analyzed. The results show that the wheel completely transitions from the stock rail to the curved switch rail between 35 and 50 mm head widths. As the head width of the curved switch rail increased, the position of the maximum wear depth gradually moved to the gauge shoulder. When the total passing weight of the train is 50 Mt, the 20 mm head width curved switch rail side wear reaches a maximum of 5.3 mm. The position in which the wheel transitions from the stock rail to the curved switch rail will be further away from the tip of the curved switch rail due to wheel–rail wear. Regarding the derailment coefficient, the wheel–rail vertical force and lateral force are both significantly impacted. However, they have little effect on the vertical and lateral acceleration of the vehicle. The wear of the wheels and rails has a higher impact on vehicle driving safety and a lower impact on vehicle driving stability.

Wheel wear is unavoidable, which affects the contact performance of the wheel and rail. This article explores the effects of wheel profile wear on the static contact and dynamic interaction between wheel and standard fixed frog in heavy haul railway. The coupling dynamic models of the vehicle-fixed frog system are established to calculate the change regulation of displacement, contact force, and acceleration when a vehicle passes through the standard fixed frog at a speed of 50 km/h in the facing move in the diverging line. Besides, the finite element models of wheel and standard fixed frog at key positions are developed to simulate the contact patch and distribution of von Mises stress in the regions of the wheel-fixed frog. Compared with the standard profile, the maximum lateral displacement of the worn profile can be reduced by up to 9 mm. The vertical contact force can be reduced from 750 kN to 320 kN, and the decrease is 57.3%. The von Mises stress could decrease up to 34% compared with the standard. And the results show that the wheel profile wear changes the positions of the wheel-rail contact points along the longitudinal direction and affects the dynamic interaction of vehicle and standard fixed frog. For the measured worn wheel profiles in this article, profile wear relieves the dynamic responses and it is good for the nose rail.

To address the challenges of real-time path planning in a dynamic environment and balance motion feasibility and global optimality, this paper proposes a hybrid pre-training algorithm. This method combines the traditional A* heuristic search with deep learning pre-training to develop a three-dimensional path planner that can learn the environmental topology and respond to dynamic obstacles. The key innovations include discretization modeling of the state space and the composite heuristic function based on reed-shepp, which effectively reduce node redundancy and improve path smoothness. The spatio-temporal obstacle path coordination was achieved by using the incremental search mechanism of pre-trained knowledge, and a dynamic response time of 76 milliseconds was realized. The simulation results show that, compared with the traditional A* and LSTM-related algorithms, the paths generated by this algorithm are significantly shortened and can accurately and safely identify and avoid various obstacle environments. Meanwhile, physical tests verify the navigation capability of this system in narrow channels and dynamic scenarios. This algorithm provides the optimal and practical solution for autonomous systems in complex environments.