数据-模型混合驱动的电动汽车集群行为模拟方法

doi:10.16183/j.cnki.jsjtu.2024.082

上海交通大学学报 ›› 2026, Vol. 60 ›› Issue (1): 51-60.doi: 10.16183/j.cnki.jsjtu.2024.082

• 新型电力系统与综合能源 • 上一篇下一篇

数据-模型混合驱动的电动汽车集群行为模拟方法

刘林¹, 杨丝雨¹, 黄夏楠¹, 陈延滔¹, 徐化帅², 王玲玲²(), 蒋传文²

¹ 国网福建省电力有限公司经济技术研究院, 福州 350012
² 上海交通大学电子信息与电气工程学院, 上海 200240

收稿日期:2024-03-15 修回日期:2024-05-03 接受日期:2024-06-13 出版日期:2026-01-28 发布日期:2026-01-27
通讯作者: 王玲玲 E-mail:himalayart@163.com.
作者简介:刘林(1986—),硕士生,从事能源经济、电力需求预测研究.
基金资助:
国网福建省电力有限公司科技项目(52130N23000A)

Data-Model Hybrid-Driven Simulation Method for Electric Vehicle Fleet Behavior

LIU Lin¹, YANG Siyu¹, HUANG Xianan¹, CHEN Yantao¹, XU Huashuai², WANG Lingling²(), JIANG Chuanwen²

¹ Economic and Technological Research Institute of State Grid Fujian Electric Power Co., Ltd., Fuzhou 350012, China
² School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China

Received:2024-03-15 Revised:2024-05-03 Accepted:2024-06-13 Online:2026-01-28 Published:2026-01-27
Contact: WANG Lingling E-mail:himalayart@163.com.

摘要/Abstract

摘要：

针对实际电动汽车充电负荷数据缺乏、地区差异大、模拟方法复杂等问题,提出由行程链、能耗链和充电链构成的电动汽车行为模拟方法.首先,为了解决实际充电数据缺乏的问题,采用数据驱动方式,基于高斯混合模型推导出车辆行程链构建方法和过程,根据电动汽车和常规汽车行程相似性,得到电动汽车的出行规律;然后,总结电动汽车行驶过程中的能耗模型,在行程链的基础上得到电动汽车的能耗链;最后,综合考虑充电焦虑模型、用户排队情况、充电时间等因素,推导并建立了电动汽车的充电链.对常见充电策略进行模拟,验证了不同充电策略对用户充电成本和电网的影响.

关键词: 电动汽车, 行程链, 能耗链, 充电链, 高斯混合模型

Abstract:

To address the issues of lack of actual electric vehicle (EV) charging load data, significant regional differences, and complex simulation methods, a behavior simulation method for EVs is proposed, which consists of a trip chain, an energy consumption chain, and a charging chain. First, to tackle the scarcity of charging data, a data-driven approach is adopted to derive the construction method and process of vehicle trip chain based on Gaussian mixture model (GMM). The travel patterns of EVs are obtained based on the similarities between EV and conventional vehicle trips. Then, energy consumption models are summarized, creating energy consumption chains based on trip chains. Finally, considering factors like charging anxiety, queueing, and charging time, the EV charging chain is developed. Simulations are conducted on common charging strategies to verify the impacts on user charging costs and the power grid.

Key words: electric vehicle (EV), trip chain, energy consumption chain, charging chain, Gaussian mixture model (GMM)

中图分类号:

TM732

刘林, 杨丝雨, 黄夏楠, 陈延滔, 徐化帅, 王玲玲, 蒋传文. 数据-模型混合驱动的电动汽车集群行为模拟方法[J]. 上海交通大学学报, 2026, 60(1): 51-60.

LIU Lin, YANG Siyu, HUANG Xianan, CHEN Yantao, XU Huashuai, WANG Lingling, JIANG Chuanwen. Data-Model Hybrid-Driven Simulation Method for Electric Vehicle Fleet Behavior[J]. Journal of Shanghai Jiao Tong University, 2026, 60(1): 51-60.

图/表 12

图1

图2

表1

图3

表2

表3

图4

图5

图6

图7

图8

表4

参考文献 23

[1]	李林晏, 韩爽, 乔延辉, 等. 面向高比例新能源并网场景的风光-电动车协同调度方法[J]. 上海交通大学学报, 2022, 56(5): 554-563. doi: 10.16183/j.cnki.jsjtu.2022.040
	LI Linyan, HAN Shuang, QIAO Yanhui, et al. A wind-solar-electric vehicles coordination scheduling method for high proportion new energy grid-connected scenarios[J]. Journal of Shanghai Jiao Tong University, 2022, 56(5): 554-563.
[2]	盛裕杰, 郭庆来, 刘梦洁, 等. 多源数据融合的用户充电行为分析与充电设施规划实践[J]. 电力系统自动化, 2022, 46(12): 151-162.
	SHENG Yujie, GUO Qinglai, LIU Mengjie, et al. User charging behavior analysis and charging facility planning practice based on multi-source data fusion[J]. Automation of Electric Power Systems, 2022, 46(12): 151-162.
[3]	DONG X H, MU Y F, JIA H J, et al. Planning of fast EV charging stations on a round freeway[J]. IEEE Transactions on Sustainable Energy, 2016, 7(4): 1452-1461. doi: 10.1109/TSTE.2016.2547891 URL
[4]	付凤杰, 方雅秀, 董红召, 等. 基于历史行驶路线的电动汽车充电站布局优化[J]. 电力系统自动化, 2018, 42(12): 72-80.
	FU Fengjie, FANG Yaxiu, DONG Hongzhao, et al. Optimized allocation of charging stations for electric vehicles based on historical trajectories[J]. Automation of Electric Power Systems, 2018, 42(12): 72-80.
[5]	ALJOHANI T M, EBRAHIM A, MOHAMMED O. Real-time metadata-driven routing optimization for electric vehicle energy consumption minimization using deep reinforcement learning and Markov chain model[J]. Electric Power Systems Research, 2021, 192: 106962. doi: 10.1016/j.epsr.2020.106962 URL
[6]	IVERSEN E B, MØLLER J K, MORALES J M, et al. Inhomogeneous Markov models for describing driving patterns[J]. IEEE Transactions on Smart Grid, 2017, 8(2): 581-588.
[7]	ROLINK J, REHTANZ C. Large-scale modeling of grid-connected electric vehicles[J]. IEEE Transactions on Power Delivery, 2013, 28(2): 894-902. doi: 10.1109/TPWRD.2012.2236364 URL
[8]	李恒杰, 朱江皓, 傅晓飞, 等. 基于集成学习的电动汽车充电站超短期负荷预测[J]. 上海交通大学学报, 2022, 56(8): 1004-1013. doi: 10.16183/j.cnki.jsjtu.2021.486
	LI Hengjie, ZHU Jianghao, FU Xiaofei, et al. Ultra-short-term load forecasting of electric vehicle charging stations based on ensemble learning[J]. Journal of Shanghai Jiao Tong University, 2022, 56(8): 1004-1013.
[9]	WANG D, GAO J Y, LI P, et al. Modeling of plug-in electric vehicle travel patterns and charging load based on trip chain generation[J]. Journal of Power Sources, 2017, 359: 468-479. doi: 10.1016/j.jpowsour.2017.05.036 URL
[10]	ZHANG J, WANG Z P, MILLER E J, et al. Charging demand prediction in Beijing based on real-world electric vehicle data[J]. Journal of Energy Storage, 2023, 57: 106294. doi: 10.1016/j.est.2022.106294 URL
[11]	TAO S, LIAO K Y, XIAO X N, et al. Charging demand for electric vehicle based on stochastic analysis of trip chain[J]. IET Generation, Transmission & Distribution, 2016, 10(11): 2689-2698. doi: 10.1049/gtd2.v10.11 URL
[12]	BUTAKOV V A, IOANNOU P A. Driver/vehicle response diagnostic system for the vehicle-following case[J]. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(5): 1947-1957. doi: 10.1109/TITS.2014.2305735 URL
[13]	赵书强, 周靖仁, 李志伟, 等. 基于出行链理论的电动汽车充电需求分析方法[J]. 电力自动化设备, 2017, 37(8): 105-112.
	ZHAO Shuqiang, ZHOU Jingren, LI Zhiwei, et al. EV charging demand analysis based on trip chain theory[J]. Electric Power Automation Equipment, 2017, 37(8): 105-112.
[14]	周志华. 机器学习[M]. 北京: 清华大学出版社, 2016: 206-209.
	ZHOU Zhihua. Machine learning[M]. Beijing: Tsinghua University Press, 2016: 206-209.
[15]	WANG Z W, SHEN C, LIU F. A conditional model of wind power forecast errors and its application in scenario generation[J]. Applied Energy, 2018, 212: 771-785. doi: 10.1016/j.apenergy.2017.12.039 URL
[16]	WU X K, FREESE D, CABRERA A, et al. Electric vehicles’ energy consumption measurement and estimation[J]. Transportation Research Part D: Transport and Environment, 2015, 34: 52-67. doi: 10.1016/j.trd.2014.10.007 URL
[17]	ARIAS M B, BAE S. Electric vehicle charging demand forecasting model based on big data technologies[J]. Applied Energy, 2016, 183: 327-339. doi: 10.1016/j.apenergy.2016.08.080 URL
[18]	DE CAUWER C, VAN MIERLO J, COOSEMANS T. Energy consumption prediction for electric vehicles based on real-world data[J]. Energies, 2015, 8(8): 8573-8593. doi: 10.3390/en8088573 URL
[19]	MANGIPINTO A, LOMBARDI F, SANVITO F D, et al. Impact of mass-scale deployment of electric vehicles and benefits of smart charging across all European countries[J]. Applied Energy, 2022, 312: 118676. doi: 10.1016/j.apenergy.2022.118676 URL
[20]	LIN H Y, FU K, WANG Y, et al. Characteristics of electric vehicle charging demand at multiple types of location—Application of an agent-based trip chain model[J]. Energy, 2019, 188: 116122. doi: 10.1016/j.energy.2019.116122 URL
[21]	中国汽车技术研究中心有限公司. 中国汽车行驶工况第1部分.轻型汽车:GB/T38146.1—2019[S]. 天津: 全国汽车标准化技术委员会, 2019.
	China Automotive Technology Research Center Co., Ltd.. Chinaautomotivetestcycle—Part1.Light-dutyvehicles:GB/T38146.1—2019[S]. Tianjin: National Automotive Standardization Technical Committee, 2019.
[22]	蒋怡静, 于艾清, 黄敏丽. 考虑用户满意度的电动汽车时空双尺度有序充电引导策略[J]. 中国电力, 2020, 53(4): 122-130.
	JIANG Yijing, YU Aiqing, HUANG Minli. Coordinated charging guiding strategy for electric vehicles in temporalspatial dimension considering user satisfaction degree[J]. Electric Power, 2020, 53(4): 122-130.
[23]	IRLE R. USA plug-in sales for 2018 full year[EB/OL]. (2018-01-26)[2024-03-15]. https://ev-volumes.com/news/ev/usa-plug-in-sales-for-2018-full-year/.

车型	整车质量/ kg	电池容量/ (kW·h)	CLTC续航里程/km
比亚迪海豹2023	1885	61.4	550
蔚来ET5	2165	75	560
小鹏P7i	1980	86.2	702
特斯拉Model S	2089	100	715
上汽大众ID.6 X	2150	63.2	460

参数	原始数据均值	拟合数据均值	误差/%
出行次数	2.77	2.60	-6.14
起始时间/min	566.05	555.61	-1.84
终止时间/min	1009.49	1004.54	-0.49
行驶时间/min	22.63	22.94	1.37
停车时间/min	147.52	153.68	4.18
行程距离/km	17.44	17.73	1.66
起始地点	2.67	2.66	-0.37
终止地点	2.71	2.77	2.21

行程类型	文献[9]预测误差/%	朴素贝叶斯模型预测误差/%	本文预测误差/%
HO&OH	-13.92	-14.23	-5.07
WO&OW	14.10	12.00	-5.74
HW&WH	8.98	9.53	5.54
OO	14.60	12.21	11.47

充电方式	充电发生次数比例/%	充电平均成本/元	充电总成本/元
无序充电	19.9	15.23	11410.46
避峰充电	15.9	12.71	7594.31

数据-模型混合驱动的电动汽车集群行为模拟方法

Data-Model Hybrid-Driven Simulation Method for Electric Vehicle Fleet Behavior

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 23

相关文章 15

编辑推荐

Metrics

本文评价

[1]	刘永江, 郭杉, 贾俊青, 刘小恺, 蔡文超, 曾龙. 多种充电模式协同的规模化电动汽车分层充电方法[J]. 上海交通大学学报, 2025, 59(9): 1304-1314.
[2]	胡龙, 方八零, 樊飞龙, 陈达伟, 李新喜, 曾润. 用户-基站-充电站能量互动和储能共享优化方法[J]. 上海交通大学学报, 2025, 59(7): 877-888.
[3]	. 基于多种影响因素的电动汽车充电负荷建模方法[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(6): 1232-1241.
[4]	孙欣, 江海林, 谢敬东, 王思敏, 王森. 基于用户信用指数的园区电动汽车充放电调度机制[J]. 上海交通大学学报, 2025, 59(2): 200-207.
[5]	韩一鸣, 贺彬, 杨博, 李嘉乐. 考虑行驶特性的电动汽车充电站联合电储能系统最优规划[J]. 上海交通大学学报, 2025, 59(11): 1720-1731.
[6]	潘益, 王明深, 缪惠宇, 袁晓冬, 韩华春. 车网互动技术激励机制演化分析[J]. 上海交通大学学报, 2025, 59(11): 1637-1646.
[7]	白文超, 班明飞, 宋梦, 夏世威, 李知艺, 宋文龙. 电动汽车-无人机联合救援系统协调调度模型[J]. 上海交通大学学报, 2024, 58(9): 1443-1453.
[8]	邱革非, 冯泽华, 沈赋, 何超, 何虹辉, 刘铠铭. 考虑车网互动的园区电网动态双层能量管理策略[J]. 上海交通大学学报, 2024, 58(6): 916-925.
[9]	张程, 匡宇, 陈文兴, 郑杨. 计及电动汽车充电方式与多能耦合的综合能源系统低碳经济优化运行[J]. 上海交通大学学报, 2024, 58(5): 669-681.
[10]	孙毅, 葛明洋, 王献春, 鲍荟谕, 杨泓玥, 姚陶. 考虑补偿激励的电动汽车多区域优化调度策略[J]. 上海交通大学学报, 2024, 58(5): 636-646.
[11]	王精, 邢海军, 王华昕, 彭思佳. 考虑电动汽车及负荷聚合商参与的综合能源系统优化调度[J]. 上海交通大学学报, 2023, 57(7): 814-823.
[12]	胡铭轩, 乔钧, 张执南. 连续康复训练动作分割与评估[J]. 上海交通大学学报, 2023, 57(5): 533-544.
[13]	张文安, 高伟展, 刘安东. 基于动态运动原语和自适应控制的机器人技能学习[J]. 上海交通大学学报, 2023, 57(3): 354-365.
[14]	. 考虑电动汽车行为异质性的鲁棒性充电需求预测方法和充电网络规划研究[J]. J Shanghai Jiaotong Univ Sci, 2023, 28(1): 136-149.
[15]	. 基于充电态势感知的充电站负荷预测方法[J]. J Shanghai Jiaotong Univ Sci, 2023, 28(1): 28-38.