Battery Consistency Charge-Discharge Control Strategy Based on Multi-Agent Reinforcement Learning
Min LI1, Boyu ZHOU1, Leilei JIANG1, Tian QIU2, Jin CHEN1
Author information+
Show less
文章历史+
Received
Published
2025-02-27
2025-06-24
Issue Date
2026-07-23
PDF (2700K)
摘要
锂离子电池作为电动汽车的核心动力源,具有高能量密度、长寿命循环和环保等优异性能,但由于电池单体间普遍存在的不一致性,使电池组整体性能下降、寿命缩短,甚至引发安全事故。为解决多阶段复杂充放电工况下电池组的一致性控制问题,提出一种基于多智能体强化学习(multi-agent reinforcement learning, MARL)的一致性充放电控制策略:构建电池二阶RC等效电路模型,模拟电池组荷电状态(state of charge, SOC)的变化,为智能体与环境的交互提供更真实的动态行为模型;对一致性控制问题进行强化学习建模,将智能体与电池单体对应,确定目标函数与奖励函数,构建电池组一致性控制多智能体强化学习MARL模型;采用近端策略优化算法(PPO)对MARL模型进行优化,得到最优的一致性充放电策略。仿真实验结果表明,在包含32个电池单体的6阶段充放电工况下,该策略能在35 min内将SOC一致性误差抑制在1%以内,实现了复杂工况下的精确、高效一致性控制。
Abstract
As the core power source of electric vehicles, lithium-ion batteries offer excellent performance characteristics such as high energy density, long cycle life, and environmental friendliness. However, inconsistencies among individual battery cells often lead to degraded overall performance, reduced lifespan, and even safety hazards. To solve the problem of consistent control of battery packs under multi-stage complex charging and discharging conditions, a consistency charging and discharging control strategy based on multi-agent reinforcement learning (MARL) is proposed: the second-order RC equivalent circuit model of the battery is constructed to simulate the changes of the state of charge (SOC) of the battery pack, providing a more realistic dynamic behavior model for the interaction between the agent and the environment; The consistency control problem is modeled through reinforcement learning, with agents corresponding to individual battery cells, the objective function and reward function determined, and a multi-agent reinforcement learning (MARL) model for consistency control of battery packs is constructed. The proximal strategy optimization algorithm is adopted to optimize the MARL model and obtain the optimal consistent charge and discharge strategy. The simulation experiment results show that under the six-stage charge and discharge conditions containing 32 battery cells, this strategy can suppress the SOC consistency error within 1% within 35 minutes, achieving precise and efficient consistency control under complex working conditions.
目前,电池组荷电状态(state of charge,SOC)均衡控制的方法主要分为被动均衡和主动均衡两大类[7-8]。被动均衡使用电阻消耗高SOC电池模块/单体中的能量。Amin等[9]探讨了利用金属氧化物半导体场效应管(metal-oxide-semiconductor field-effect transistor,MOSFET)内部电阻进行被动均衡的方法,并分析了其在小容量和大容量电池中的效果。Song等[10]提出了一种新的被动均衡算法,通过电池容量和电阻的变化来优化均衡效果。吴青峰等[11]提出了一种基于分流电阻的电池组间电池健康状态(State of Health, SOH)被动均衡方案,研究分流电阻阻值计算和接入切除法则,实现锂电池SOH均衡。被动均衡的控制电路和算法都比较简单,但会将大量的电量以热量的形式消耗,降低电池的整体效率,并可能造成过热等问题[12]。主动均衡通过将SOC高的电池单体/模块电量转移到少的单体/模块上,实现电池组SOC平衡。现有研究主要针对均衡拓扑,Shah等[13]通过使用高电压单体为低电压单体充电,实现主动均衡,并展示了在电动汽车电池管理系统中的实际应用。Samanta等[14]提出一种基于双向直流-直流转换器(direct current to direct current converter,DC-DC)的有源电池均衡拓扑,极大减少了有源元件和电源开关的数量。Rau等[15]开发了一种以电感为基础的主动均衡拓扑结构,提高了电池系统的整体性能和寿命。Manjunath等[16]介绍了一种基于两级模块的电池间有源均衡拓扑,该拓扑基于改进的降压-升压转换器,适用于串联锂离子电池组一致性控制,降低了控制的复杂度。上述主动均衡方法降低了电池组一致性控制过程中的能耗[17],但拓扑构建思路较为简单,一致性控制效率较低。此外,一些研究通过可重构电池系统实现一致性控制。可重构电池通过在电池组中添加多组开关电路,并控制开关对电池组中的模块/单体进行选择性充放电,从而实现电池组一致性控制[18-20]。虽然这类电路提高了电池的充放电效率,但开关电路存在固有的短路风险。当控制策略错误或硬件故障出现时,电池组可能短路引发火灾或爆炸[21-22]。上述研究表明,提升一致性控制的拓扑结构有助于实现更精细的控制效果,但受限于电路设计的复杂度与成本因素,仅靠硬件手段仍存在局限。因此,拓扑需要结合高效的控制策略,进一步提升整体一致性控制的性能。
ZHANGQ. Comparative analysis of EV and ICE vehicles in passenger and commercial vehicle scenarios[J]. Green Petroleum & Petrochemicals, 2022, 7(4): 6-9. DOI: 10.3969/j.issn.2095-0942.2022.04.002(Ch ).
LAIX, CHENQ W, GUH H, et al. Life cycle assessment of lithium-ion batteries for carbon-peaking and carbon-neutrality: Framework, methods, and progress[J]. Journal of Mechanical Engineering, 2022, 58(22): 3-18. DOI: 10.3901/JME.2022.22.003(Ch ).
XIAOX, TIANP G, YUL, et al. Status and prospect of safety studies of cascade power battery energy storage system[J]. Journal of Electrical Engineering, 2022, 17(1): 206-224. DOI: 10.11985/2022.01.027(Ch ).
[7]
RAJASHEKARAK. Present status and future trends in electric vehicle propulsion technologies[J]. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2013, 1(1): 3-10. DOI: 10.1109/JESTPE.2013.2259614 .
[8]
KAMELM, SANKARANARAYANANV, ZANER, et al. State-of-charge balancing with parallel and series output connected battery power modules[J]. IEEE Transactions on Power Electronics, 2022, 37(6): 6669-6677. DOI: 10.1109/TPEL.2022.3143835 .
[9]
OMARIBAZ B, ZHANGL J, SUND B. Review of battery cell balancing methodologies for optimizing battery pack performance in electric vehicles[J]. IEEE Access, 2019, 7: 129335-129352. DOI: 10.1109/ACCESS.2019.2940090 .
[10]
JIANGB W, LIUY J, HUANGX L, et al. A new battery active balancing method with supercapacitor considering regeneration process[C]//IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society, New York: IEEE Press, 2020: 2364-2369. DOI: 10.1109/IECON43393.2020.9254839 .
[11]
CARTERJ, FANZ, CAOJ. Cell equalisation circuits: A review[J]. Journal of Power Sources, 2020, 448: 227489. DOI: 10.1016/j.jpowsour.2019.227489 .
[12]
AMIN, ISMAILK, NUGROHOA, et al. Passive balancing battery management system using MOSFET internal resistance as balancing resistor[C]//2017 International Conference on Sustainable Energy Engineering and Application (ICSEEA), New York: IEEE Press, 2017: 151-155. DOI: 10.1109/ICSEEA.2017.8267701 .
[13]
SONGH, LEES. Study on the systematic design of a passive balancing algorithm applying variable voltage deviation[J]. Electronics, 2023, 12(12): 2587. DOI: 10.3390/electronics12122587 .
WUQ F, YANGK Y, YANGY T, et al. Passive SOH balancing scheme for lithium battery based on switching shunt resistance[J]. Acta Energiae Solaris Sinica, 2024, 45(12): 520-527. DOI: 10.19912/j.0254-0096.tynxb.2023-1310 (Ch ).
[16]
KHALIDA, STEVENSONA, SARWATA I. Performance analysis of commercial passive balancing battery management system operation using a hardware-in-the-loop testbed[J]. Energies, 2021, 14(23): 8037. DOI: 10.3390/en14238037 .
[17]
SHAHS, MURALIM, GANDHIP. A practical approach of active cell balancing in a battery management system[C]//2018 IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES), New York: IEEE Press, 2018: 1-6. DOI: 10.1109/PEDES.2018.8707811 .
[18]
SAMANTAA, CHOWDHURIS. Active cell balancing of lithium-ion battery pack using dual DC-DC converter and auxiliary lead-acid battery[J]. Journal of Energy Storage, 2021, 33: 102109. DOI: 10.1016/j.est.2020.102109 .
[19]
RAU J, ZHANGH W, BISCHOFV, et al. Development of a smart battery system with active balancing for a mobile robot platform[C]//International Conference on the Efficiency and Performance Engineering Network. Cham: Springer Nature Switzerland, 2023: 385-395. DOI: 10.1007/978-3-031-49421-5_31 .
[20]
MANJUNATHK, KALPANAR, SINGHB, et al. A two-stage module based cell-to-cell active balancing circuit for series connected lithium-ion battery packs[J]. IEEE Transactions on Energy Conversion, 2023, 38(4): 2282-2297. DOI: 10.1109/TEC.2023.3283424 .
[21]
GHAEMINEZHADN, OUYANGQ, HUX S, et al. Active cell equalization topologies analysis for battery packs: A systematic review[J]. IEEE Transactions on Power Electronics, 2021, 36(8): 9119-9135. DOI: 10.1109/TPEL.2021.3052163 .
[22]
TASHAKORN, KACETLJ, FANGJ Y, et al. Generic dynamically reconfigurable battery with integrated auxiliary output and balancing capability[J]. IEEE Transactions on Power Electronics, 2023, 38(7): 7933-7944. DOI: 10.1109/TPEL.2023.3263809 .
[23]
ALAHMADM, HESSH, MOJARRADIM, et al. Battery switch array system with application for JPL’s rechargeable micro-scale batteries[J]. Journal of Power Sources, 2008, 177(2): 566-578. DOI: 10.1016/j.jpowsour.2007.11.053 .
[24]
KIMH, SHINK G. On dynamic reconfiguration of a large-scale battery system[C]//2009 15th IEEE Real-Time and Embedded Technology and Applications Symposium. New York: IEEE Press, 2009: 87-96. DOI: 10.1109/RTAS.2009.13 .
[25]
ZHAOR, LIUJ, GUJ J. Simulation and experimental study on lithium ion battery short circuit[J]. Applied Energy, 2016, 173: 29-39. DOI: 10.1016/j.apenergy.2016.04.016 .
[26]
ZHANGY Z, XIONGR, HEH W, et al. Long short-term memory recurrent neural network for remaining useful life prediction of lithium-ion batteries[J]. IEEE Transactions on Vehicular Technology, 2018, 67(7): 5695-5705. DOI: 10.1109/TVT.2018.2805189 .
[27]
MAY, DUANP, SUNY S, et al. Equalization of lithium-ion battery pack based on fuzzy logic control in electric vehicle[J]. IEEE Transactions on Industrial Electronics, 2018, 65(8): 6762-6771. DOI:10.1109/TIE.2018.2795578 .
[28]
OUYANGQ, HANW J, ZOUC F, et al. Cell balancing control for lithium-ion battery packs: A hierarchical optimal approach[J]. IEEE Transactions on Industrial Informatics, 2020, 16(8): 5065-5075. DOI:10.1109/TII.2019.2950818 .
[29]
MCHENJ, BEHALA, LIZ J, et al. Active battery cell balancing by real-time model predictive control for extending electric vehicle driving range[J]. IEEE Transactions on Automation Science and Engineering, 2024, 21(3): 4003-4015. DOI:10.1109/TASE.2023.3291679 .
[30]
FANT E, LIUS M, YANGH, et al. A fast active balancing strategy based on model predictive control for lithium-ion battery packs[J]. Energy, 2023, 279: 128028. DOI:10.1016/j.energy.2023.128028 .
WUW J, WUJ, GUOH T, et al. Cascaded equalization control of lithium batteries based on variable discourse domain fuzzy PID algorithm[J]. Journal of Electronic Measurement and Instrumentation, 2024, 38(8): 137-144. DOI: 10.13382/j.jemi.B2407281(Ch ).
KARMAKARS, BERAT K, BOHREA K. Novel PI controller and ANN controllers-based passive cell balancing for battery management system[J]. IEEE Transactions on Industry Applications, 2023, 59(6): 7623-7634. DOI:10.1109/TIA.2023.3299886 .
[35]
CASPARM, SCHÜRMANNT, ANNEKENM, et al. Active balancing control for distributed battery systems based on cooperative game theory[J]. Journal of Energy Storage, 2023, 68: 107585. DOI:10.1016/j.est.2023.107585 .
CHENG C. Consistency intelligent control and low temperature flexible preheating of lithium-ion battery pack[D]. Jinan: Shandong University, 2023. DOI: 10.27272/d.cnki.gshdu.2023.003180(Ch ).
[38]
LIAOL, LIH G, LIH J, et al. Research on equalization scheme of lithium-ion battery packs based on consistency control strategy[J]. Journal of Energy Storage, 2023, 73: 109193. DOI:10.1016/j.est.2023.109193 .