Predictive energy management for urban electric bus transit: An edge-enhanced Graph Transformer approach

Abstract

Achieving energy-optimal urban mobility requires precise predictive energy management for electric buses (EBs). However, existing trip-level predictive models largely overlook the topological structure and complex driving dynamics of stop-route networks. This study proposes an edge-enhanced Graph Transformer framework that models each trip as a directed graph, with stops as nodes and inter-stop micro-trips as edges, to predict trip-level energy consumption rates (ECR). Using real-world data from 219 EBs across 47 routes in Guangzhou, the model integrates stop, route, kinetic, and environmental features. A novel edge enhancement mechanism is introduced to strengthen micro-trip dynamics within the attention computation. Feature contributions are quantitatively analyzed using SHAP values, and internal decision-making patterns are interpreted through attention weight visualization. The proposed model achieves MAE of 0.0756 kWh/km, RMSE of 0.0972 kWh/km, MAPE of 10.14%, and R² of 0.77 on unseen routes, outperforming all baselines. Edge enhancement yields MAPE reductions of 8.51% and 5.92% for Graph Transformer and GCN, respectively. Mechanism analysis confirms that edge features dominate feature fusion with approximately 62% attention weight allocation. The achieved prediction accuracy falls within the typical SOC safety margins maintained in EB battery management, supporting practical deployment in charging scheduling and energy-aware route planning.

Graphical abstract

Keywords

electric bus / energy consumption prediction / Graph Transformer / energy management / SHAP

References

[1] Aamodt ACory KConey K (2021). Electrifying transit: A guidebook for implementing battery electric buses. Golden, CO: National Renewable Energy Laboratory, NREL/TP-7A40–76932

[2] Abdelaty H, Al-Obaidi A, Mohamed M, Farag H E Z, (2021). Machine learning prediction models for battery-electric bus energy consumption in transit. Transportation Research Part D, Transport and Environment, 96: 102868

[3] Chen Y, Zhang Y, Sun R, (2021). Data-driven estimation of energy consumption for electric bus under real-world driving conditions. Transportation Research Part D, Transport and Environment, 98: 102969

[4] Dong C, Xiong Z, Li N, Yu X, Liang M, Zhang C, Li Y, Wang H, (2025a). A real-time prediction framework for energy consumption of electric buses using integrated machine learning algorithms. Transportation Research Part E, Logistics and Transportation Review, 194: 103884

[5] Dong C, Xiong Z, Zhang C, Li N, Li Y, Xie N, Zhang J, Wang H, (2025b). A transformer-based approach for deep feature extraction and energy consumption prediction of electric buses based on driving distances. Applied Energy, 380: 123941

[6] Dong G, Witlox F, Bie Y, (2024). Decentralizing e-bus charging infrastructure deployment leads to economic and environmental benefits. Communications in Transportation Research, 4: 100139

[7] El-Taweel N A, Zidan A, Farag H E Z, (2021). Novel electric bus energy consumption model based on probabilistic synthetic speed profile integrated with HVAC. IEEE Transactions on Intelligent Transportation Systems, 22( 3): 1517–1531

[8] Gallet M, Massier T, Hamacher T, (2018). Estimation of the energy demand of electric buses based on real-world data for large-scale public transport networks. Applied Energy, 230: 344–356

[9] Geng M, Li J, Xia Y, Chen X, (2023). A physics-informed Transformer model for vehicle trajectory prediction on highways. Transportation Research Part C, Emerging Technologies, 154: 104272

[10] Hendriks J N, Sturmberg B C P, (2024). An integrated model of electric bus energy consumption and optimised depot charging. npj Sustainable Mobility and Transport, 1( 1): 10

[11] Jiang J, Yu Y, Min H, Cao Q, Sun W, Zhang Z, Luo C, (2023). Trip-level energy consumption prediction model for electric bus combining Markov-based speed profile generation and Gaussian processing regression. Energy, 263: 125866

[12] Jiang W, Luo J, (2022). Graph neural network for traffic forecasting: A survey. Expert Systems with Applications, 207: 117921

[13] Kong L, Yang H, Li W, Zhang Y, Guan J, Zhou S, (2025). Traffexplainer: A framework toward GNN-based interpretable traffic prediction. IEEE Transactions on Artificial Intelligence, 6( 3): 559–573

[14] Li P, Zhang Y, Zhang Y, Zhang K, Jiang M, (2021a). The effects of dynamic traffic conditions, route characteristics and environmental conditions on trip-based electricity consumption prediction of electric bus. Energy, 218: 119437

[15] Li P, Zhang Y, Zhang Y, Zhang Y, Zhang K, (2021b). Prediction of electric bus energy consumption with stochastic speed profile generation modelling and data driven method based on real-world big data. Applied Energy, 298: 117204

[16] Li X, Wang T, Li J, Tian Y, Tian J, (2022). Energy consumption estimation for electric buses based on a physical and data-driven fusion model. Energies, 15( 11): 4160

[17] Liu Y, Liang H, (2022). A data-driven approach for electric bus energy consumption estimation. IEEE Transactions on Intelligent Transportation Systems, 23( 10): 17027–17038

[18] López F C, Fernández R Á, (2020). Predictive model for energy consumption of battery electric vehicle with consideration of self-uncertainty route factors. Journal of Cleaner Production, 276: 124188

[19] Lu Z, Xing T, Li Y, (2025). Optimization of electric bus vehicle scheduling and charging strategies under Time-of-Use electricity price. Transportation Research Part E, Logistics and Transportation Review, 196: 104021

[20] Lundberg S M, Lee S I, (2017). A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems, 30: 4765–4774

[21] Manzolli J A, Trovão J P, Antunes C H, (2022). A review of electric bus vehicles research topics – Methods and trends. Renewable & Sustainable Energy Reviews, 159: 112211

[22] McCabe D, Ban X, Kulcsár B, (2025). Minimum-delay opportunity charging scheduling for electric buses. Communications in Transportation Research, 5: 100209

[23] Nan S, Tu R, Li T, Sun J, Chen H, (2022). From driving behavior to energy consumption: A novel method to predict the energy consumption of electric bus. Energy, 261: 125188

[24] Pamuła T, Pamuła W, (2020). Estimation of the energy consumption of battery electric buses for public transport networks using real-world data and deep learning. Energies, 13( 9): 2340

[25] Pan Y, Fang W, Ge Z, Li C, Wang C, Guo B, (2024). A hybrid on-line approach for predicting the energy consumption of electric buses based on vehicle dynamics and system identification. Energy, 290: 130205

[26] Rahmani S, Baghbani A, Bouguila N, Patterson Z, (2023). Graph neural networks for intelligent transportation systems: A survey. IEEE Transactions on Intelligent Transportation Systems, 24( 8): 8846–8885

[27] Shao Z, Bell M G H, Wang Z, Geers D G, Yao X, Gao J, (2025). CCDSReFormer: Traffic flow prediction with a criss-crossed dual-stream enhanced rectified transformer model. Communications in Transportation Research, 5: 100189

[28] Shi D, Yang Q, Zhu Q, Jupp D L B, Long Y, (2021). Uncertainties and errors in algorithms for elevation gradients. International Journal of Geographical Information Science, 35( 2): 296–320

[29] Wei Y, Yang X, Xiao X, Ma Z, Zhu T, Dou F, Wu J, Chen A, Gao Z, (2024). Understanding the resilience of urban rail transit: Concepts, reviews, and trends. Engineering, 41: 7–18

[30] Wen Y, Xu P, Li Z, Xu W, Wang X, (2023). RPConvformer: A novel Transformer-based deep neural networks for traffic flow prediction. Expert Systems with Applications, 218: 119587

[31] Xu Z, Wang J, Lund P D, Zhang Y, (2023). Analysis of energy consumption for electric buses based on low-frequency real-world data. Transportation Research Part D, Transport and Environment, 122: 103857

[32] Yang Y, Cheng J, Liu Y, (2024). An overview of solutions to the bus bunching problem in urban bus systems. Frontiers of Engineering Management, 11( 4): 661–675

[33] Yang Y, Zhan J, Liu Y, Qu X, (2026a). Cross-city transfer learning for optimal e-scooter parking station deployment: Evidence from 25 European cities. ENGINEERING Management, 13( 1): 194–212

[34] Yang Y, Zhan J, Liu Y, Wang Q, (2025). Cross-city transfer learning: Applications and challenges for smart cities and sustainable transportation. Communications in Transportation Research, 5: 100206

[35] Yang Y, Zhan J, Xu M, Liu Y, Qu X, (2026b). Toward climate-neutral urban mobility: Understanding shared e-scooter carbon emission patterns through multi-city evidence in Europe. Transportation Research Part A, Policy and Practice, 203: 104736

[36] Zeng Z, Qu X, (2023). What’s next for battery-electric bus charging systems. Communications in Transportation Research, 3: 100094

[37] Zhang J, Wang Z, Liu P, Zhang Z, (2020). Energy consumption analysis and prediction of electric vehicles based on real-world driving data. Applied Energy, 275: 115408

[38] Zhang K, Feng X, Wu L, He Z, (2022). Trajectory prediction for autonomous driving using spatial-temporal graph attention transformer. IEEE Transactions on Intelligent Transportation Systems, 23( 11): 22343–22353

[39] Zhang X, Zhang Z, Liu Y, Xu Z, Qu X, (2024a). A review of machine learning approaches for electric vehicle energy consumption modelling in urban transportation. Renewable Energy, 234: 121243

[40] Zhang Y, Zhou Q, Wang J, Kouvelas A, Makridis M A, (2025a). CASAformer: Congestion-aware sparse attention transformer for traffic speed prediction. Communications in Transportation Research, 5: 100174

[41] Zhang Z, Wang S, Ye B, Ma Y, (2025b). A feature prediction-based method for energy consumption prediction of electric buses. Energy, 314: 134345

[42] Zhang Z, Ye B, Wang S, Ma Y, (2024b). Analysis and estimation of energy consumption of electric buses using real-world data. Transportation Research Part D, Transport and Environment, 126: 104017

[43] Zhao L, Song Y, Zhang C, Liu Y, Wang P, Lin T, Deng M, Li H, (2020). T-GCN: A temporal graph convolutional network for traffic prediction. IEEE Transactions on Intelligent Transportation Systems, 21( 9): 3848–3858

[44] Zhao Y, Chen X, Liu P, Nielsen C P, McElroy M B, (2025). Future ultrafast charging stations for electric vehicles in China: Charging patterns, grid impacts and solutions, and upgrade costs. Engineering, 48: 309–322

[45] Zhong L, Zeng Z, Huang Z, Shi X, Bie Y, (2024). Joint optimization of electric bus charging and energy storage system scheduling. Frontiers of Engineering Management, 11( 4): 676–696

[46] Zhou F, Yang Q, Zhong T, Chen D, Zhang N, (2021). Variational graph neural networks for road traffic prediction in intelligent transportation systems. IEEE Transactions on Industrial Informatics, 17( 4): 2802–2812

[47] Zhou Y, Wang H, Wang Y, Li R, (2022). Robust optimization for integrated planning of electric-bus charger deployment and charging scheduling. Transportation Research Part D, Transport and Environment, 110: 103410

[48] Zhu Q, Huang Y, Feng Lee C, Liu P, Zhang J, Wik T, (2025). Predicting electric vehicle energy consumption from field data using machine learning. IEEE Transactions on Transportation Electrification, 11( 1): 2120–2132

RIGHTS & PERMISSIONS

Higher Education Press