Open Access

ARTICLE

Online Estimation Method of Train Wheel-Rail Adhesion Coefficient Based on Parameter Estimation

Yi Zhang¹, Wenliang Zhu^1,*, Hanbin Wang¹, Chun Tian², Jiajun Zhou²

1 School of Urban Rail Transportation, Shanghai University of Engineering Science, Shanghai, 201620, China
2 School of Transportation, Tongji University, Shanghai, 201834, China

* Corresponding Author: Wenliang Zhu. Email:

Computer Modeling in Engineering & Sciences 2025, 144(3), 2873-2891. https://doi.org/10.32604/cmes.2025.068951

Received 10 June 2025; Accepted 05 September 2025; Issue published 30 September 2025

Abstract

Aiming to address the challenge of directly measuring the real-time adhesion coefficient between wheels and rails, this paper proposes an online estimation algorithm for the adhesion coefficient based on parameter estimation. Firstly, a force analysis of the single-wheel pair model of the train is conducted to derive the calculation relationship for the wheel-rail adhesion coefficient in train dynamics. Then, an estimator based on parameter estimation is designed, and its stability is verified. This estimator is combined with the wheelset force analysis to estimate the wheel-rail adhesion coefficient. Finally, the approach is validated through joint simulations on the MATLAB/Simulink and AMESim platforms, as well as a hardware-in-the-loop semi-physical simulation experimental platform that accounts for system delay and noise conditions. The results indicate that the proposed algorithm effectively tracks changes in the adhesion coefficient during train braking, including the decrease in adhesion when the train brakes and slides, and the overall increase as the train speed decreases. The effectiveness of the algorithm was verified by setting different test conditions. The results show that the estimation algorithm can accurately estimate the adhesion coefficient, and through error analysis, it is found that the error between the estimated value of the adhesion coefficient and the theoretical value of the adhesion coefficient is within 5%. The adhesion coefficient obtained through the online estimation method based on the parameter estimation proposed in this paper demonstrates strong followability in both simulation and practical applications.

---

Keywords

---