The disorderly charging behavior of large-scale electric vehicles (EVs) is recognized to cause prolonged queuing times, increased charging costs, unbalanced utilization of charging station resources, and aggravated voltage fluctuations in the power grid, thereby undermining system stability. To address these challenges, a hierarchical graph reinforcement learning (HGRL) algorithm was proposed to optimize EV charging navigation strategies for coordinated multi-agent benefits. An EV charging model that accounts for user heterogeneity was developed by incorporating mileage anxiety, temporal and economic preferences, and energy consumption. Based on this model, a multi-agent objective function was formulated to minimize the total time cost, economic cost, mileage anxiety, and driving energy consumption of users. To effectively solve the problem, the charging navigation task was modeled as a two-layer stochastic optimization problem encompassing both charging station destination selection and route planning. The proposed HGRL method is designed to overcome the curse of dimensionality in high-dimensional state spaces and the challenges of stochastic uncertainty in optimization. It is demonstrated through simulation results that the HGRL algorithm effectively optimizes the interests of multiple stakeholders—including EV users, charging stations, and the power grid—while providing personalized and intelligent charging navigation strategies that account for user heterogeneity, and exhibits strong adaptability to varying system conditions.
JIANG Chang-xu, CAI Zhi-xuan, YUAN Yu-juan, LIN Jun-jie.
Electric Vehicle Charging Navigation Strategy Considering User Heterogeneity and Multi-agent Interests[J].
Science Technology and Engineering, 2026, 26(27): 11786-11803 DOI:10.12404/j.issn.1671-1815.2506846
有学者从EV用户利益角度出发,建立了充电引导模型,以优化时间成本和经济成本等关键指标。文献[3]在预测交通流量的基础上,提出了一种能够兼顾用户多需求场景和多导航目标的充电引导模型。文献[4]综合考虑了出行时间、出行距离、电动汽车能耗及出行舒适度等因素,提出了一种基于Dijkstra算法的多目标电动汽车路径规划方法。文献[5]利用混合整数非线性规划算法对充电引导策略进行优化,有效降低了EV用户的时间成本与经济成本。鉴于EV充电引导问题涉及EV、充电站、交通路网和电网等多个主体,有学者进一步从车-站-路-网多方协同的角度展开研究。文献[6]基于层次分析法和改进CRITIC(criteria importance though intercrieria correlation)法优化了EV充电站选址,并结合Topsis方法优化充电路径,从而在电动汽车、交通路网、充电站和配电网四个层面实现了利益均衡。文献[7]通过实时电价引导电动汽车充放电,有效降低了微电网运行经济成本和出力波动。文献[8]综合考虑了充电服务收入、政府补贴、基础设施建设成本及运营维护费用等因素,提出了一种以运营商利润最大化为目标的充电站优化配置策略。文献[9]则针对大规模EV充电可能引发的电网电能质量下降、路网道路拥堵和充电桩利用率不足等问题,基于前景理论提出了一种新的EV充电引导方案。
用户里程焦虑是指EV驾驶员因担心车辆剩余电量不足以支撑到达目的地或充电站而产生的一种心理状态。文献[17-18]表明:①EV用户的里程焦虑值主要受剩余电量(state of charge, SOC)的影响;②当SOC低于某一个阈值时,EV用户才会产生里程焦虑;③里程焦虑的变化呈非线性特征,且随着SOC趋近于零,其增长率显著加快。基于上述研究结论,可以假设电动汽车用户的里程焦虑值随着电量的增加而呈凸性递减,当SOC值达到或高于某阈值时,用户里程焦虑逐渐消失。因此,EV的里程焦虑值斜率与车辆SOC值的关系[19-20]可表示为
min fEV(X)=$\frac{{\mathit{T}}_{\mathit{k},\mathit{c},\mathit{t}}-{\mathit{T}}_{\mathit{m}\mathit{i}\mathit{n}}}{{\mathit{T}}_{\mathit{m}\mathit{a}\mathit{x}}-{\mathit{T}}_{\mathit{m}\mathit{i}\mathit{n}}}$+$\frac{{\mathit{F}}_{\mathit{k},\mathit{c},\mathit{t}}-{\mathit{F}}_{\mathit{m}\mathit{i}\mathit{n}}}{{\mathit{F}}_{\mathit{m}\mathit{a}\mathit{x}}-{\mathit{F}}_{\mathit{m}\mathit{i}\mathit{n}}}$+$\frac{{\mathit{D}}_{\mathit{k},\mathit{c},\mathit{t}}-{\mathit{D}}_{\mathit{m}\mathit{i}\mathit{n}}}{{\mathit{D}}_{\mathit{m}\mathit{a}\mathit{x}}-{\mathit{D}}_{\mathit{m}\mathit{i}\mathit{n}}}$+$\frac{{\mathit{B}}_{\mathit{k},\mathit{c},\mathit{t}}-{\mathit{B}}_{\mathit{m}\mathit{i}\mathit{n}}}{{\mathit{B}}_{\mathit{m}\mathit{a}\mathit{x}}-{\mathit{B}}_{\mathit{m}\mathit{i}\mathit{n}}}$
如图13所示为4座充电站的平均电价和完成充电的EV数量。由图13可知,在4个充电站中HGRL策略下完成充电的EV数量最多,因此本文所提策略能够充分利用充电桩资源,同时能够合理分布EV负荷。由第三节可知,EV充电总功率的增加通常会导致节点边际电价提高,但HGRL算法能够充分考虑充电功率异质性信息,在优化EV数量的同时协调总功率分布。相比SGRL2和SDM策略,尽管HGRL策略下CS1与CS2服务的EV数量更多,但电价反而更低,从而进一步提升了电力-交通耦合网络的经济性。与此同时,由CS3服务的EV数量可知,SGRL2和SDM策略未能充分利用其充电桩资源,抵达该站的 EV 数量偏少,导致与 CS3 相连的电网节点负荷较低,因此其平均电价也相对较低。
HuangXueliang, LiuYongdong, ShenFei, et al. Electric vehicle-grid interaction: a review and outlook[J]. Automation of Electric Power Systems, 2024, 48(7): 3-23.
CaiJunpeng, ChenCaibo, ChenDewang, et al. Research on electric vehicle charging guidance strategy[J]. Modern Electronics Technique, 2021, 44(1): 137-143.
HuJunjie, PanYi, XuChengming, et al. Fast guidance strategy for electric vehicle charging based on dynamic traffic inference[J]. Transactions of China Electrotechnical Society, 2025, 40(9): 2880-2896.
ZhaoJun, LiJianhua, YaoDunmin. Route optimization of electric vehicles considering travel comfort[J]. Science Technology and Engineering, 2024, 24(17): 7368-7375.
[9]
HuangY, LiuM, ZhangY. Electric vehicle entire-trip navigation and charging reservation method based on a high-speed communication network[J]. International Journal of Electrical Power & Energy Systems, 2023, 150: 109070.
LiuLijun, ChenChang, HuXin, et al. Ordered charging strategy for electric vehicles based on the information coupling of vehicle-road-station-grid[J]. High Voltage Engineering, 2024, 50(2): 693-703.
XuJiaqi, GuoHongxia, ZouGuilin. Two-stageoptimal scheduling of microgrid with electric vehicles based on real-time electricity price[J]. Science Technology and Engineering, 2023, 23(13): 5571-5578.
MaLongfei, ZhangBaoqun, TangQi, et al. Charging station configuration strategy for maximizing operator profits under multiple constraints[J]. Science Technology and Engineering, 2025, 25(26): 11202-11209.
ZhangCong, PengKe, XiaoChuanliang, et al. EV charging guiding strategy based on coordination of “EVs-road-network”[J]. Electric Power Automation Equipment, 2022, 42(10): 125-133.
[18]
LiY, SuS, ZhangM, et al. Multi-agent graph reinforcement learning method for electric vehicle on-route charging guidance in coupled transportation electrification[J]. IEEE Transactions on Sustainable Energy, 2023, 15(2): 1180-1193.
[19]
SuS, LiY, YamashitaK, et al. Electric vehicle charging guidance strategy considering “traffic network-charging station-driver” modeling: a multi-agent deep reinforcement learning based approach[J]. IEEE Transactions on Transportation Electrification, 2024, 10(3): 4653-4666.
[20]
JiangC, ZhouL, ZhengJ H, et al. Electric vehicle charging navigation strategy in coupled smart grid and transportation network: a hierarchical reinforcement learning approach[J]. International Journal of Electrical Power and Energy Systems, 2024, 157: 109823.
ZhanHua, JiangChangxu, SuQinglie. Electric vehicle charging navigation method based on hierarchical reinforcement learning[J]. Electric Power Automation Equipment, 2022, 42(10): 264-272.
[23]
XingQ, XuY, ChenZ. A bilevel graph reinforcement learning method for electric vehicle fleet charging guidance[J]. IEEE Transactions on Smart Grid, 2023, 14(4): 3309-3312.
JiangChangxu, LuYuejun, ShaoZhenguo, et al. Collaborative optimization operation of integrated electric power and traffic network based on graph neural network multi-agent reinforcement learning[J]. High Voltage Engineering, 2023, 49(11): 4622-4631.
SongYuhang, ChenYufang, WeiYanling, et al. Charging path planning for electric vehicles based on reinforcement learning environment design strategy[J]. Automation of Electric Power Systems, 2024, 48(11): 184-196.
[28]
LiuH, WangD Z W. Locating multiple types of charging facilities for battery electric vehicles[J]. Transportation Research Part B: Methodological, 2017, 103: 30-55.
[29]
GuoF, YangJ, LuJ. The battery charging station location problem: impact of users’ range anxiety and distance convenience[J]. Transportation Research Part E: Logistics and Transportation Review, 2018, 114: 1-18.
[30]
YangJ, GuoF, ZhangM. Optimal planning of swapping/charging station network with customer satisfaction[J]. Transportation Research Part E: Logistics and Transportation Review, 2017, 103: 174-197.
[31]
XuM, YangH, WangS. Mitigate the range anxiety: siting battery charging stations for electric vehicle drivers[J]. Transportation Research Part C: Emerging Technologies, 2020, 114: 164-188.
JiangChangxu, LuYuejun, YuanYujuan, et al. Highway charging station and distribution network expansion planning considering range anxiety[J]. Power System Technology, 2025, 49(9): 3881-3890.
HuYongshi, QingChenyang, LüShuhao, et al. Research on the location selection of electric vehicle charging stations considering user satisfaction and range anxiety[J]. Journal of Wuhan University of Technology(Transportation Science & Engineering), 2025, 49(4): 706-711, 718.
[36]
KucukogluI, DewilR, CattrysseD. The electric vehicle routing problem and its variations: a literature review[J]. Computers & Industrial Engineering, 2021, 161: 107650.
JiangChangxu, YuanYujuan, LiuChenxi, et al. Multi-agent graph reinforcement learning-based electric vehicle charging guidance strategy driven by user comprehensive satisfaction[J]. Power System Technology, 2026, 50(5): 1985-2000.