Abstract: The development of efficient thermal management strategy is critical for hybrid electric drive tracked vehicles (HETVs) due to the severe thermal safety and energy consumption challenges encountered during complex operations. Conventional strategies struggle to balance high-precision temperature control with multi-objective collaborative optimization, while requiring long development cycles and exhibiting weak generalization capabilities. To address these issues, we propose a hierarchical thermal management framework integrating a gated recurrent unit multi-head attention twin delayed deep deterministic policy gradient with model predictive control (GMA-TD3-MPC). This framework dynamically integrates reinforcement learning (RL) and model predictive control (MPC), utilizing a gated recurrent unit with multi-head attention (GRU-MHA) module to optimize energy consumption and temperature control precision under cyclic conditions; meanwhile, it implements a dynamic threshold triggering mechanism to seamlessly transfer control to the MPC controller when approaching thermal safety limits. Our simulation results demonstrate that compared to baseline strategies, the proposed method accelerates convergence by approximately 28% and 40% over deep deterministic policy gradient (DDPG) and TD3, respectively. In a standard temperature environment (25 ℃) under off-road conditions, compared to standalone MPC, the proposed strategy reduces temperature fluctuation ranges in high-temperature and low-temperature circuits by 44.19% and 6.45%, respectively, while achieving a 5.54% reduction in the total energy consumption and a 10.63% decrease in the peak power demand. Furthermore, under high-temperature conditions (45 ℃), the strategy reduces the total energy consumption by 13.41%.

基于GMA-TD3-MPC算法考虑系统安全与能耗的混合动力履带车热管理策略

[1]Arrinda M, Batalla J, Vicente I, et al., 2025. Model predictive control as the cloud control strategy for a battery thermal management system. Energy Reports, 13: 4635- 4646.

[2]Atkinson C, McCane B, Szymanski L, et al., 2021. Pseudo-rehearsal: achieving deep reinforcement learning without catastrophic forgetting. Neurocomputing, 428: 291- 307.

[3]Çeliksöz D, Kılıç V, 2024. Series-hybrid powertrains: advancing mobility control in electric tracked vehicle technology. World Electric Vehicle Journal, 15(2): 47.

[4]Chen JH, He P, Cai SJ, et al., 2024a. Modeling and temperature control of a water-cooled PEMFC system using intelligent algorithms. Applied Energy, 372: 123790.

[5]Chen JH, He P, He ZH, et al., 2024b. Multi-objective decoupling control of thermal management system for PEM fuel cell. Energy and AI, 18: 100447.

[6]Cobbe K, Klimov O, Hesse C, et al., 2018. Quantifying generalization in reinforcement learning. arXiv:1812. 02341.

[7]Darici S, Canli E, Dogan S, et al., 2012. Determination of heat transfer rate and pressure drop performance of an intercooler for heavy duty engines. International Journal of Arts & Sciences, 5(7): 43- 57.

[8]Dong F, Wang ZM, Ni J, 2020. The effect of different coolants on emissions and fuel consumption of a diesel engine. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 42(7): 359.

[9]Ghate A, Sundar A, Zhu QL, et al., 2024. Development of an integrated energy and thermal planner for a series hybrid off-road autonomous tracked vehicle. Energy Conversion and Management, 322: 119163.

[10]Golubev T, Canull L, Edel Z, 2024. Modeling the impact of thermal management on electric vehicle energy consumption and driving range. 40th Semiconductor Thermal Measurement, Modeling & Management Symposium, p. 1- 7.

[11]Guo R, Sun ZY, Luo MH, 2024. Energy-efficient battery thermal management strategy for range extended electric vehicles based on model predictive control and dynamic programming. Energy, 307: 132769.

[12]Guo X, Peng JK, He HW, et al., 2025. Integrated thermal-energy management for electric vehicles in high-temperature conditions using hierarchical reinforcement learning. Expert Systems with Applications, 276: 127221.

[13]Han J, Yu S, Yi S, 2017. Advanced thermal management of automotive fuel cells using a model reference adaptive control algorithm. International Journal of Hydrogen Energy, 42(7): 4328- 4341.

[14]He HW, Su QC, Huang RC, et al., 2024. Enabling intelligent transferable energy management of series hybrid electric tracked vehicle across motion dimensions via soft actor-critic algorithm. Energy, 294: 130933.

[15]He P, Chen JH, Zhang CZ, et al., 2024. Optimization of modeling and temperature control of air-cooled PEMFC based on TLBO-DE. Energy and AI, 18: 100430.

[16]Hori Y, 2004. Future vehicle driven by electricity and control—research on four-wheel-motored “UOT electric march II”. IEEE Transactions on Industrial Electronics, 51(5): 954- 962.

[17]Jiang KR, Tian ZB, Wen T, et al., 2025. Collaborative optimization strategy of hydrogen fuel cell train energy and thermal management system based on deep reinforcement learning. Applied Energy, 393: 126057.

[18]Karaoglan MU, 2021. Drive cycle performance analysis for various powertrain alternatives of zero-emission battery electric military vehicles. International Journal of Global Warming, 25(3-4): 482- 497.

[19]Li K, Bhattacharyya S, Dinh TQ, et al., 2025. Advanced model predictive control for energy-efficient thermal management in intelligent electric refrigerated vans. 27th International Conference on Advanced Communications Technology, p. 273- 278.

[20]Liu HF, Wen MS, Yang HB, et al., 2021. A review of thermal management system and control strategy for automotive engines. Journal of Energy Engineering, 147(2): 03121001.

[21]Ma Y, Ma Q, Liu YQ, et al., 2024. Two-level optimization strategy for vehicle speed and battery thermal management in connected and automated EVs. Applied Energy, 361: 122928.

[22]Mancarella A, Marello O, 2023. Effect of coolant temperature on performance and emissions of a compression ignition engine running on conventional diesel and hydrotreated vegetable oil (HVO). Energies, 16(1): 144.

[23]Nam K, Ahn C, 2025. Energy-efficient battery thermal management in electric vehicles using artificial-neural-network-based model predictive control. World Electric Vehicle Journal, 16(5): 279.

[24]Nguyen TT, Nguyen ND, Nahavandi S, 2020. Deep reinforcement learning for multiagent systems: a review of challenges, solutions, and applications. IEEE Transactions on Cybernetics, 50(9): 3826- 3839.

[25]Norouzi A, Heidarifar H, Borhan H, et al., 2023. Integrating machine learning and model predictive control for automotive applications: a review and future directions. Engineering Applications of Artificial Intelligence, 120: 105878.

[26]Scolaro E, Beligoj M, Estevez MP, et al., 2021. Electrification of agricultural machinery: a review. IEEE Access, 9: 164520- 164541.

[27]Shoai Naini S, Miller RS, Rizoo D, et al., 2023. A model reference adaptive controller for an electric motor thermal management system in autonomous vehicles. SAE International Journal of Electrified Vehicles, 12(1): 3- 16.

[28]Sundar A, Vannarath A, Prucka R, et al., 2023. The influence of cooling air-path restrictions on fuel consumption of a series hybrid electric off-road tracked vehicle. Energy & Propulsion Conference & Exhibition.

[29]Sundar A, Ghate A, Zhu QL, et al., 2026. An integrated power, energy, and thermal management strategy using cascaded control for off-road autonomous hybrid vehicles. Applied Energy, 406: 127269.

[30]Tang AH, Wu ZK, Xu TT, et al., 2024. Week-level early warning strategy for thermal runaway risk based on real-scenario operating data of electric vehicles. eTransportation, 19: 100308.

[31]Tian J, Jing HC, Xiong Y, et al., 2026. Experimental study of the effect of electrode geometry on the ignition and flame development of NH ₃/air mixtures in nanosecond plasma discharges. Combustion and Flame, 286: 114809.

[32]Wang D, Zheng WF, Li SQ, et al., 2025. Impact analysis of uncertainty in thermal resistor-capacitor models on model predictive control performance. Energy and Buildings, 328: 115112.

[33]Wang F, Harindintwali JD, Yuan ZZ, et al., 2021. Technologies and perspectives for achieving carbon neutrality. Innovation, 2(4): 100180.

[34]Wang TY, Wu J, Wu JW, et al., 2025. Energy consumption analysis and performance evaluation of electric vehicle integrated thermal management system experiments. Applied Thermal Engineering, 269: 126002.

[35]Wu BY, Jia Z, Li ZG, et al., 2021. Different exhaust temperature management technologies for heavy-duty diesel engines with regard to thermal efficiency. Applied Thermal Engineering, 186: 116495.

[36]Xi YG, Li DW, Lin S, 2013. Model predictive control—status and challenges. Acta Automatica Sinica, 39(3): 222- 236.

[37]Yu YB, Jiang JY, Min HT, et al., 2023. Integrated energy and thermal management strategy for extended range electric vehicle based on battery temperature and state-of-charge global planning. Energy Conversion and Management, 288: 117154.

[38]Yuan RY, Fletcher T, Ahmedov A, et al., 2020. Modelling and co-simulation of hybrid vehicles: a thermal management perspective. Applied Thermal Engineering, 180: 115883.

[39]Zhai L, Yan JHY, Liu AG, et al., 2025. Structure design of stator oil circuit and thermal characteristics analysis for high-speed oil-cooled flat wire motor. Transactions of Beijing Institute of Technology, 45(6): 600- 608 (in Chinese).

[40]Zhang J, Liu B, Zhou YJ, et al., 2025. Hierarchical energy management strategy for fuel cell hybrid electric vehicles based on improved TD3 considering interior temperature comfort. Applied Thermal Engineering, 276: 126965.

[41]Zhao X, Ma XW, Chen BY, et al., 2022. Challenges toward carbon neutrality in China: strategies and countermeasures. Resources, Conservation and Recycling, 176: 105959.

[42]Zhu QL, Kumar A, Sundar A, et al., 2022. Development of a series hybrid electrified powertrain for a high speed tracked vehicle based on driving cycle simulation. SAE International Journal of Advances and Current Practices in Mobility, 4(4): 1403- 1412.