Train Slide Prediction and Risk Assessment Using Vehicle-Signal Data: A Data-Model Fusion Method

Abstract

Insufficient wheel–rail adhesion during braking poses a significant threat to operational safety and accelerates component wear. Although anti-slide valves modulate braking force to mitigate sliding, the resulting speed fluctuations can inadvertently trigger emergency braking commands from the signaling system, potentially exacerbating the slide. Existing detection methods are largely reactive, relying on post-event identification rather than predictive foresight. To bridge this gap, an onboard module for predictive risk warning is proposed. Utilizing real-world vehicle and signaling data, a cross-attention CNN–GRU model is proposed for the accurate prediction of whole-train wheel slide probability. In parallel, a train dynamics model projects the braking trajectory to assess the displacement deviation relative to target positions and safety margins. By integrating the data-driven slide probability with the model-based displacement deviation, a dynamic risk matrix is constructed to assess different levels of train slide risk. With an optimal prediction horizon of 2.5 s and established alarm thresholds, the proposed module provides actionable inputs for the train operation control system. Experimental results demonstrate that the proposed method effectively predicts both wheel sliding and the associated overrun risk. This enables the signaling system to proactively intervene in sliding control, preemptively mitigating sliding occurrences and enhancing overall operational safety.

Keywords

Urban rail transit / Wheel slide prediction / Data model fusion / Slide risk assessment

References

[1] Ivanov V, Savitski D, Shyrokau B. A survey of traction control and antilock braking systems of full electric vehicles with individually controlled electric motors. IEEE Trans Veh Technol, 2015, 64(9): 3878-3896

[2] Oliveira DN, Camporez RM, Neto CAR, Strey NF, Scandian C. Effects of temperature and load on sliding wear between wheel and rail steels. Tribol Int, 2025, 210 Article ID: 110804

[3] Megna G, Bracciali A. Gearless track-friendly metro with guided independently rotating wheels. Urban Rail Transit, 2021, 7(4): 285

[4] Qiao Z. Optimization of train anti-slip control strategies for urban rail transit multi-system coordination. Urban Mass Transit, 2024, 27(11): 62-6570

[5] Wang H, He H, Yang Y. Research on design scheme of train speed protection under rain and snow mode for signal system of urban rail transit. Railw Signal Commun, 2024, 60(10): 74-79

[6] Mathur K, Kucuk F (2023) Optimized slip and slide prevention control for electric train units. In: 2023 IEEE 3rd international conference on sustainable energy and future electric transportation (SEFET), pp 1–6. https://doi.org/10.1109/SeFeT57834.2023.10244859

[7] Kim K, Kong S-H, Jeon S-Y. Slip and slide detection and adaptive information sharing algorithms for high-speed train navigation systems. IEEE Trans Intell Transp Syst, 2015, 16(6): 3193-3203

[8] Anuszczyk J, Gocek A, Pacholski K, Dominikowski B. Generating and modeling of braking curve and the assessment of the quality of automatic stopping of the train. Transp Probl, 2018, 13: 13-25

[9] Wen T, Wang J, Liu L, Cai B, Roberts C. Train operational state estimation: technologies, challenges, and future directions. Fund Res, 2025

[10] Ohishi K (2008) Realization of fine motion control based on disturbance observer. In: 2008 10th IEEE international workshop on advanced motion control, pp 1–8. https://doi.org/10.1109/AMC.2008.4516032

[11] Liu J, Peng Q, Huang Z, Liu W, Li H. Enhanced sliding mode control and online estimation of optimal slip ratio for railway vehicle braking systems. Int J Precis Eng Manuf, 2018, 19: 655-664

[12] Zhao K, Li P, Zhang C, He J, Li Y, Yin T. Online accurate estimation of the wheel-rail adhesion coefficient and optimal adhesion antiskid control of heavy-haul electric locomotives based on asymmetric barrier lyapunov function. J Sens, 2018, 2018(1): 2740679

[13] Pichlík P, Bauer J. Analysis of the locomotive wheel slip controller operation during low velocity. IEEE Trans Intell Transp Syst, 2021, 22(3): 1543-1552

[14] Yi L, Jiang W, Yi Y, Li J, Xie C. Freight train speed active disturbance rejection tracking and wheel anti-slip based on fuzzy neural network with DBO optimization. Electr Eng, 2023, 106: 631-644

[15] Gan H, Wang S, Lu J, Ou H. Adhesion control of high-speed train based on improved nonlinear Kalman filter. Appl Sci, 2025

[16] Yin S, Peng T, Yang C, Yang C, Chen Z, Gui W. Dynamic hybrid observer-based early slipping fault detection for high-speed train wheelsets. Control Eng Pract, 2024, 142: 105736

[17] Huang J, Jiang B, Xu C, Wang N. Slipping detection of electric locomotive based on empirical wavelet transform, fuzzy entropy algorithm and support vector machine. IEEE Trans Veh Technol, 2021, 70(8): 7558-7570

[18] Abouzeid AF, Guerrero JM, Lejarza-Lasuen L, Muniategui-Aspiazu I, Endemaño-Isasi A, Briz F. Advanced maximum adhesion tracking strategies in railway traction drives. IEEE Trans Transp Electrif, 2024, 10(2): 3645-3660

[19] Chen H, Cai W, Song Y (2014) Wheel skid prediction and antiskid control of high speed trains. In: 17th international IEEE conference on intelligent transportation systems (ITSC), pp 1209–1214. https://doi.org/10.1109/ITSC.2014.6957852

[20] Yang Y, Wang X, Zhang Y. Study on mechanism analysis of skidding prediction for electric vehicle based on time-delay effect of force transmission. World Electr Veh J, 2021

[21] Moaveni B, Fathabadi FR, Molavi A. Supervisory predictive control for wheel slip prevention and tracking of desired speed profile in electric trains. ISA Trans, 2020, 101: 102-115

[22] Namoano B, Ruiz-Carcel C, Emmanouilidis C, Starr AG (2022) Data-driven wheel slip diagnostics for improved railway operations. IFAC-PapersOnLine 55(19):103–108. https://doi.org/10.1016/j.ifacol.2022.09.191. 5th IFAC workshop on advanced maintenance engineering, services and technologies AMEST 2022

[23] Namoano B, Emmanouilidis C, Starr A. Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method. Eng Appl Artif Intell, 2024, 137 Article ID: 109173

[24] Huang J, He S, Feng H, Yu Y. Slipping trend prediction based on improved informer. Appl Sci, 2025

[25] Shrestha S, Wu Q, Spiryagin M. Review of adhesion estimation approaches for rail vehicles. Int J Rail Transp, 2019, 7(2): 79-102

[26] Zewang Y, Wu M, Tian C, Jiajun Z, Chen C. A review on the application of friction models in wheel-rail adhesion calculation. Urban Rail Transit, 2021

[27] Ou D, Tong Y, Wang Y, Dai S, Chang M (2025) Vehicle-signal fusion based autonomous perception method of subway train adhesion characteristic parameters. In: Jia L, Ou D, Liu H, Zong F, Wang P, Zhang M (eds) The proceedings of 2024 international conference on artificial intelligence and autonomous transportation. Springer, Singapore, pp 173–182. https://doi.org/10.1007/978-981-96-3961-8_17

[28] Li X, Gwan C, Zhao S, Gao X, Zhu Y. Multimodal temperature prediction for lithium-ion battery thermal runaway using multi-scale gated fusion and bidirectional cross-attention mechanisms. J Energy Storage, 2025, 116 Article ID: 116098

[29] Khan R, Sameem MSI, Kanwal K, Iqbal M, Hossain M, Ye Z (2022) A deep neural framework for image caption generation using GRU-based attention mechanism. https://doi.org/10.48550/arXiv.2203.01594

[30] Abdul Salam M, Gamal M, Hamed HFA, Sweidan S. Abstractive text summarization using deep learning models: a survey. Int J Data Sci Analyt, 2025

[31] Guo Z, Yang C, Wang D, Liu H. A novel deep learning model integrating CNN and GRU to predict particulate matter concentrations. Process Saf Environ Prot, 2023, 173: 604-613

[32] Du G, Liu N, Zhang D, Li Q, Sun J, Jiang X, Zhu Z. Grounding fault diagnosis of running rails based on a multi-scale one-dimensional convolutional neural network in a dc subway system. Urban Rail Transit, 2024

[33] Park C, Song H, Lee C. S3-Net: SRU-based sentence and self-matching networks for machine reading comprehension. ACM Trans Asian Low-Resour Lang Inf Process, 2020

[34] Qin Q, Wang Z, Li B, Huang A. The HWAM-EMD-GRU forecasting model for short-term passenger flow in an airport light rail transit line. Urban Rail Transit, 2024

Funding

the National Natural Science Foundation of China(52472348)

the Open Project of Shanghai Engineering Research Center of Driverless Train Control of urban Guided Transport(SUTC-2024KT-01)

the State Key Laboratory of Advanced Rail Autonomous Operation(RAO2024K02)

RIGHTS & PERMISSIONS

The Author(s)