基于DQN的智能电动汽车动态充电调度算法研究

doi:10.16088/j.issn.1001-6600.2024092101

广西师范大学学报（自然科学版） ›› 2025, Vol. 43 ›› Issue (5): 52-63.doi: 10.16088/j.issn.1001-6600.2024092101

基于DQN的智能电动汽车动态充电调度算法研究

黄媛艳^1,3, 路旋^1,3, 詹凯杰^1,3, 曾海勇^1,2,3*

1.广西类脑计算与智能芯片重点实验室(广西师范大学),广西桂林 541004;
2.广西高校集成电路与微系统重点实验室(广西师范大学),广西桂林 541004;
3.广西师范大学电子与信息工程学院/集成电路学院,广西桂林 541004

收稿日期:2024-09-21 修回日期:2024-11-05 出版日期:2025-09-05 发布日期:2025-08-05
通讯作者: 曾海勇(1989—),男,湖南郴州人,广西师范大学副教授,博士。E-mail: zenghaiyong@gxnu.edu.cn
基金资助:
国家自然科学基金青年基金(62301172);广西自然科学基金青年基金(2023JJB170030);广西青年创新人才专项(桂科AD23026197);广西类脑计算与智能芯片主任基金(BCIC-23-Z5)

Study on Intelligent EV Dynamic Charging Scheduling Algorithm Based on DQN

HUANG Yuanyan^1,3, LU Xuan^1,3, ZHAN Kaijie^1,3, ZENG Haiyong^1,2,3*

1. Guangxi Key Laboratory of Brain-inspired Computing and Intelligent Chips (Guangxi Normal University), Guilin Guangxi 541004, China;
2. Key Laboratory of Integrated Circuits and Microsystems in Guangxi Universities (Guangxi Normal University), Guilin Guangxi 541004, China;
3. School of Electronics and Information Engineering/School of Integrated Circuits, Guangxi Normal University, Guilin Guangxi 541004, China

Received:2024-09-21 Revised:2024-11-05 Online:2025-09-05 Published:2025-08-05

摘要/Abstract

摘要： 随着电动汽车的广泛应用和环保政策的推行,合理调度电动汽车的充电行为对满足用户需求、保障电网稳定性变得愈发重要。针对智能电网中电动汽车车主的差异化充电需求和传统机器学习算法存在的维数灾难问题,本文提出一种基于深度Q网络(deep Q-network, DQN)的电动汽车充电调度算法。该算法利用深度强化学习的优势,结合实时电价和车辆状态等信息,动态调整电动汽车的充放电行为,旨在最大化经济效益并优化电网负荷。实验结果表明,与非受控式充电策略相比,本文提出的算法在30 d的测试期内可降低充电成本约65.6%,能够有效适应电网实时价格波动和用户需求变化,减少高峰充电需求,有效降低充电成本,提高电网的稳定性。

关键词: 电动汽车充电调度, 深度强化学习, 能量优化管理, 车联网, 智能电网, 信息交互

Abstract: With the widespread adoption of electric vehicles (EVs) and the implementation of environmental policies, efficiently scheduling EV charging behaviors has become increasingly important for meeting user demands and can ensure grid stability. Addressing the different charging needs of EV owners in smart grids and the curse of dimensionality problem in traditional machine learning algorithms, this paper proposes an EV charging scheduling algorithm based on Deep Q-Network (DQN). This algorithm leverages the advantages of deep reinforcement learning, combining real-time electricity prices and vehicle status information to dynamically adjust EV charging and discharging behaviors, aiming to maximize economic benefits and optimize grid load. Experimental results show that compared with uncontrolled charging strategies, the proposed algorithm can reduce charging costs by approximately 65.6% over a 30-day test period. It effectively adapts to real-time grid price fluctuations and changes in user demands, reduces peak charging requirements, effectively lowers charging costs, and improves grid stability.

Key words: electric vehicle charging scheduling, deep reinforcement learning, energy optimization management, vehicle-to-grid, smart grid, information interactive

中图分类号: U491.8;TM73

黄媛艳, 路旋, 詹凯杰, 曾海勇. 基于DQN的智能电动汽车动态充电调度算法研究[J]. 广西师范大学学报（自然科学版）, 2025, 43(5): 52-63.

HUANG Yuanyan, LU Xuan, ZHAN Kaijie, ZENG Haiyong. Study on Intelligent EV Dynamic Charging Scheduling Algorithm Based on DQN[J]. Journal of Guangxi Normal University(Natural Science Edition), 2025, 43(5): 52-63.

参考文献

[1] 詹华,江昌旭, 苏庆列. 基于分层强化学习的电动汽车充电引导方法[J]. 电力自动化设备, 2022, 42(10): 264-272. DOI: 10.16081/j.epae.202208022.
[2] MANRÍQUEZ F, SAUMA E, AGUADO J, et al. The impact of electric vehicle charging schemes in power system expansion planning[J]. Applied Energy, 2020, 262: 114527. DOI: 10.1016/j.apenergy.2020.114527.
[3] HU J J, YANG G Y, ZIRAS C, et al. Aggregator operation in the balancing market through network-constrained transactive energy[J]. IEEE Transactions on Power Systems, 2019, 34(5): 4071-4080. DOI: 10.1109/TPWRS.2018.2874255.
[4] 田梦秋, 承楠, 李长乐. 6G无线网络场景知识研究综述[J]. 无线电通信技术, 2024, 50(3): 484-495.
[5] 简富俊, 黄黎明, 汪洋, 等. 智能电网通信系统技术的研究[J]. 自动化应用, 2024(S1): 464-466. DOI: 10.19769/j.zdhy.2024.S1.153.
[6] 温昕. 车联网技术在新能源汽车控制系统设计中的应用研究[J]. 装备维修技术, 2024(1): 70-73. DOI: 10.16648/j.cnki.1005-2917.2024.01.017.
[7] 高爽, 戴如鑫. 电动汽车集群参与调频辅助服务市场的充电调控策略[J]. 电力系统自动化, 2023, 47(18): 60-67.
[8] 郭永强, 张靖, 何宇, 等. 考虑电动汽车充放电响应的微电网混合储能配置[J]. 电网与清洁能源, 2022, 38(8): 82-93. DOI: 10.3969/j.issn.1674-3814.2022.08.011.
[9] 吴洲洋, 艾欣, 胡俊杰. 需求侧灵活性资源参与调频辅助服务的备用优化与实时调度[J]. 电力系统自动化, 2021, 45(6): 148-157. DOI: 10.7500/AEPS20200629008.
[10] 付学林. 电动汽车混合模式充电调度策略研究[D]. 成都: 电子科技大学, 2020. DOI: 10.27005/d.cnki.gdzku.2020.002099.
[11] 傅质馨, 朱韦翰, 朱俊澎, 等. 动态路-电耦合网络下电动出租车快速充电引导及其定价策略[J]. 电力自动化设备, 2022, 42(4): 9-17. DOI: 10.16081/j.epae.202201013.
[12] SHEIK M S, IMTHIAS A T P, DEVARAJ D. Optimized charge scheduling of plug-In electric vehicles using modified placement algorithm[C]// 2019 International Conference on Computer Communication and Informatics (ICCCI). January 23-25, 2019. Coimbatore, Tamil Nadu, India. IEEE, 2019: 1-5.. DOI: 10.1109/iccci.2019.8821960.
[13] REN J M, WANG H, WEI Y, et al. A novel genetic algorithm-based emergent electric vehicle charging scheduling scheme[C]// IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society. October 14-17, 2019, Lisbon, Portugal. IEEE, 2019: 4289-4292. DOI: 10.1109/IECON.2019.8926746.
[14] ZHANG J, WANG W, WANG R H, et al. An electric vehicle scheduling strategy for the power grid peak regulation based on V2G[C]// 2023 3rd International Conference on New Energy and Power Engineering (ICNEPE). November 24-26, 2023, Huzhou, China. IEEE, 2023: 1181-1187. DOI: 10.1109/ICNEPE60694.2023.10429591.
[15] CHIŞ A, LUNDÉN J, KOIVUNEN V. Reinforcement learning-based plug-in electric vehicle charging with forecasted price[J]. IEEE Transactions on Vehicular Technology, 2017, 66(5): 3674-3684. DOI: 10.1109/TVT.2016.2603536.
[16] WAN Z Q, LI H P, HE H B, et al. Model-free real-time EV charging scheduling based on deep reinforcement learning[J]. IEEE Transactions on Smart Grid, 2019, 10(5): 5246-5257. DOI: 10.1109/TSG.2018.2879572.
[17] 杜明秋, 李妍, 王标, 等. 电动汽车充电控制的深度增强学习优化方法[J]. 中国电机工程学报, 2019, 39(14): 4042-4049. DOI: 10.13334/j.0258-8013.pcsee.181187.
[18] 李航, 李国杰, 汪可友. 基于深度强化学习的电动汽车实时调度策略[J]. 电力系统自动化, 2020, 44(22): 161-167.
[19] 黄元清, 刘迪迪, 覃光锋, 等. 计及车主需求的电动汽车聚合商能量调度策略[J]. 南方电网技术, 2024, 18(10): 161-170. DOI: 10.13648/j.cnki.issn1674-0629.2024.10.016.
[20] 柴雅琪, 海立卫, 韩颖, 等. 电动汽车站内充电调度优化方法研究[J]. 电工技术, 2024(14): 180-183. DOI: 10.19768/j.cnki.dgjs.2024.14.051.
[21] 彭可兴. 基于强化学习的多智能体协作通信策略研究[D]. 南京: 南京信息工程大学, 2024. DOI: 10.27248/d.cnki.gnjqc.2024.000016.
[22] 王娟, 陈明,支刚, 等. 电动汽车充电负荷的蒙特卡洛预测方法[J]. 云南电力技术, 2024, 52(4): 10-14.
[23] 李培建, 闫锡桐. 一种应用动态规划的通信链路推荐方法[C]// 第三十八届中国(天津)2024'IT、网络、信息技术、电子、仪器仪表创新学术会议论文集. 天津:天津市电子学会, 2024: 235-238. DOI: 10.26914/c.cnkihy.2024.014526.
[24] NORD POOL. Historical market data[DB/OL]. (2018-09-17)[2024-09-22]. http://www.nordpoolgroup.com/historical-market-data/.
[25] MHAISEN N, FETAIS N, MASSOUD A. Real-time scheduling for electric vehicles charging/discharging using reinforcement learning[C]// 2020 IEEE International Conference on Informatics, IoT, and Enabling Technologies (ICIoT). Piscataway, NJ: IEEE, 2020: 1-6. DOI: 10.1109/iciot48696.2020.9089471.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于DQN的智能电动汽车动态充电调度算法研究

Study on Intelligent EV Dynamic Charging Scheduling Algorithm Based on DQN

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

Metrics

本文评价

推荐阅读 10

[1]	黄仁慧, 张锐锋, 文晓浩, 闭金杰, 黄守麟, 李廷会. 基于复数协方差卷积神经网络的运动想象脑电信号解码方法[J]. 广西师范大学学报（自然科学版）, 2025, 43(3): 43-56.
[2]	汤亮, 陈博文, 牛一森, 马荣庚. 基于YOLOv8的雾天车辆行人实时检测方法[J]. 广西师范大学学报（自然科学版）, 2025, 43(3): 72-83.
[3]	梁胤杰, 南新元, 蔡鑫, 李云鹏, 勾海光. 基于数据增广与改进YOLOv8的桥梁缺陷检测[J]. 广西师范大学学报（自然科学版）, 2025, 43(3): 84-97.
[4]	田晟, 陈东. 基于深度强化学习的网联燃料电池混合动力汽车生态驾驶联合优化方法[J]. 广西师范大学学报（自然科学版）, 2024, 42(6): 67-80.
[5]	陈秀锋, 王成鑫, 赵凤阳, 杨凯, 谷可鑫. 改进DQN算法的单点交叉口信号控制方法[J]. 广西师范大学学报（自然科学版）, 2024, 42(6): 81-88.
[6]	唐峯竹, 唐欣, 李春海, 李晓欢. 基于深度强化学习的多无人机任务动态分配[J]. 广西师范大学学报（自然科学版）, 2021, 39(6): 63-71.