• CN: 11-2187/TH
  • ISSN: 0577-6686

Journal of Mechanical Engineering ›› 2025, Vol. 61 ›› Issue (16): 283-292.doi: 10.3901/JME.2025.16.283

Previous Articles    

Optimization of EMU Operation Stability Based on Evolutionary Control of Wheel Tread Hollow Worn

WANG Zegen, GONG Dao, ZHOU Jinsong, LIU Guangyu   

  1. Institute of Rail Transit, Tongji University, Shanghai 201804
  • Accepted:2024-09-12 Online:2025-02-05 Published:2025-02-05

Abstract: During the normal service process of a certain type of EMU, there was a significant decrease in the operation stability of the train due to the continuous development of wheel tread wear. In order to improve the operation stability of the vehicle and ensure the safety of the train, the tracking tests were carried out on the wheel tread wear and the vibration characteristics of the frame. An optimization method of EMU operation stability based on the evolution control of wheel tread wear is proposed to solve the problem of low stability from the root cause of tread wear. This method proposes to use the hollow worn position, hollow worn depth and hollow worn width to define the hollow worn state of the wheel tread, and then establishes the mapping relationship model between tread hollow worn parameters and equivalent taper based on BP neural network, and analyzes the characteristics of the hollow worn parameters corresponding to the low equivalent taper, and then carries out orthogonal optimization on the key vehicle dynamic parameters to improve the hollow worn state of the wheel tread, and finally improves the vehicle's operation stability. The results show that after optimization by this method, the hollow worn depth of the wheel tread with the same mileage is reduced by 14.46%, and the hollow worn state of wheel tread is significantly improved; the equivalent conicity of the wheel with the same mileage is reduced by 16.39%, and the lateral vibration acceleration of the frame is reduced by 32.02%, so the operation stability of the vehicle is also significantly improved; at the same time, the ride comfort and safety of the vehicle still meet the standard requirements. This study synergistically considers the evolution of tread hollow wear and stability degradation, and the proposed optimization method can provide a certain reference for the treatment of stability degradation problems in other high-speed trains.

Key words: EMU trains, hollow worn tread, stability, equivalent conicity, BP neural network

CLC Number: