基于LSTM神经网络的电动汽车充电站需求响应特性封装及配电网优化运行

doi:10.12204/j.issn.1000-7229.2021.06.008

电力建设 ›› 2021, Vol. 42 ›› Issue (6): 76-85.doi: 10.12204/j.issn.1000-7229.2021.06.008

• 电动汽车参与电网调度的关键技术·栏目主持傅质馨副教授· • 上一篇下一篇

基于LSTM神经网络的电动汽车充电站需求响应特性封装及配电网优化运行

薛溟枫¹, 毛晓波¹, 潘湧涛¹, 杨艳红², 赵振兴², 李彦君²

1.国网江苏省电力有限公司无锡供电分公司, 江苏省无锡市 214000
2.中国科学院电工研究所,北京市 100190

收稿日期:2020-11-27 出版日期:2021-06-01 发布日期:2021-05-28
通讯作者: 杨艳红
作者简介:薛溟枫(1976),男,本科,工程师,研究方向为需求侧管理、有序用电、综合能源服务等;|毛晓波(1989),男,硕士,工程师,研究方向为需求侧管理、综合能源服务等;|潘湧涛(1988),男,本科,工程师,研究方向为综合能源服务、市场化售电等;|赵振兴(1984),男,助理研究员,研究方向为储能微电网、综合能源系统优化运行;|李彦君(1991),男,硕士研究生,研究方向为深度学习在电力系统方向的应用。
基金资助:
国网江苏省电力有限公司科技项目(J2019078)

Demand Response Package Model of Electric Vehicle Charging Station Based on LSTM Neural Network and Optimal Operation of Distribution Network

XUE Mingfeng¹, MAO Xiaobo¹, PAN Yongtao¹, YANG Yanhong², ZHAO Zhenxing², LI Yanjun²

1. Wuxi Power Supply Company of State Grid Jiangsu Electric Power Co., Ltd., Wuxi 214000, Jiangsu Province, China
2. Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, China

Received:2020-11-27 Online:2021-06-01 Published:2021-05-28
Contact: YANG Yanhong
Supported by:
State Grid Jiangsu Electric Power Co., Ltd. Research Program(J2019078)

摘要/Abstract

摘要：

随着电动汽车的快速增长,大规模电动汽车充电具有随机性、时空耦合性的特点,对配电网运行电压造成越限风险。通过基于价格的需求响应,引导电动汽车在大时空范围有序合理地充电成为重要的技术手段。文章研究基于数据驱动的电动汽车充电站需求响应特性及其参与配电网运行优化调度问题,首先提出单体电动汽车充电模型和计及交通网络拓扑结构的电动汽车行驶特性,建立了区域电动汽车充电站负荷需求响应计算方法;在此基础上,提出了基于LSTM深度神经网络的电动汽车充电站需求响应模型封装方法,得到电动汽车充电站充电成本和充电功率响应之间的映射模型;接着,构建了考虑电动汽车充电站需求响应的区域配电网电压运行优化模型,并采用粒子群算法进行求解;最后,通过对包含3个充电站的33 节点系统的算例对比分析,验证了所述电动汽车充电站需求响应及其参与配电网优化运行方法的有效性,为数据驱动方法解决电动汽车充电和需求响应问题提供借鉴。

关键词: 电动汽车充电站, LSTM神经网络, 需求响应, 封装模型

Abstract:

With the rapid growth of electric vehicles (EVs), the charging characteristics of large-scale EVs are randomness and spatiotemporal coupling, which poses a risk of exceeding the limit on the operating voltage of distribution network. Through the demand response (DR) based on price, it has become an important technical means to guide the orderly and reasonable charging of EVs in a large space-time range. In this paper, the DR characteristics of EV charging station based on data-driven and its participation in the operation optimization of distribution network are studied. Firstly, the charging model of single EV and the driving characteristics of EV considering the topological structure of traffic network are proposed, and the load simulation calculation method of regional EV charging station is established. On this basis, the electric power system based on LSTM deep neural network is proposed. The mapping model between charging cost and power response of EV charging station is obtained by encapsulating the DR model of EV charging station. Furthermore, a voltage operation optimization model of regional distribution network considering the DR of EV charging station is constructed, and the model is solved with particle swarm optimization algorithm. Finally, the comparison and analysis of the 33-node system with 3 charging stations verify the effectiveness of the proposed method of EV charging station DR and its participation in distribution network operation optimization. It provides reference for data-driven method to solve the problem of EV charging and demand response.

Key words: electric vehicle charging station, LSTM neural network, demand response, package model

中图分类号:

TM73

薛溟枫, 毛晓波, 潘湧涛, 杨艳红, 赵振兴, 李彦君. 基于LSTM神经网络的电动汽车充电站需求响应特性封装及配电网优化运行[J]. 电力建设, 2021, 42(6): 76-85.

XUE Mingfeng, MAO Xiaobo, PAN Yongtao, YANG Yanhong, ZHAO Zhenxing, LI Yanjun. Demand Response Package Model of Electric Vehicle Charging Station Based on LSTM Neural Network and Optimal Operation of Distribution Network[J]. ELECTRIC POWER CONSTRUCTION, 2021, 42(6): 76-85.

图/表 14

图1 充电行为封装与配电网优化研究框架

Fig.1 Research framework of charging behavior encapsulation and distribution network optimization

图2 包含不同功能区域划分的道路交通网络

Fig.2 Road network accross different functional areas

图3 价格与充电站负载率关系曲线

Fig.3 Relationship between price and charging load rate

图4 充电站负荷曲线

Fig.4 Load curve of charging station

图5 电动汽车充电价格曲线

Fig.5 EV Charging price

图6 LSTM封装训练模型

Fig.6 Training model of LSTM package

图7 PSO优化充电站价格流程

Fig.7 Flow of charging station price optimization based on PSO

图8 配网33节点系统

Fig.8 33-node system of distribution network

图9 3组光伏电源有功出力曲线

Fig.9 Active power output curve of three groups photovoltaic power supply

图10 充电站LSTM封装模型训练误差

Fig.10 Training error of charging station LSTM package model

图11 节点电压评价函数优化曲线

Fig.11 Optimization curve of node voltage evaluation function

图12 配网节点电压优化

Fig.12 Node voltage optimization of distribution network

图13 配网节点电压连续优化

Fig.13 Continuous optimization of node voltages in distribution network

图14 充电站连续优化价格策略

Fig.14 Continuous optimization of price strategy for charging stations

参考文献 20

[1]	张西竹, 刘洵源, 杨文涛, 等. 动态分时电价机制下的电动汽车分层调度策略[J]. 电力建设, 2018,39(12):73-80.
	ZHANG Xizhu, LIU Xunyuan, YANG Wentao, et al. A hierarchical scheduling strategy for electric vehicles under dynamic time-of-use tariff mechanism[J]. Electric Power Construction, 2018,39(12):73-80.
[2]	ZHANG J, YAN J, LIU Y Q, et al. Daily electric vehicle charging load profiles considering demographics of vehicle users[J]. Applied Energy, 2020,274:115063. doi: 10.1016/j.apenergy.2020.115063 URL
[3]	杨健维, 苟方杰, 黄宇, 等. 基于不确定性测度的居民小区电动汽车充电分时电价制定策略[J]. 电网技术, 2018,42(1):96-102.
	YANG Jianwei, GOU Fangjie, HUANG Yu, et al. Residential area electric vehicle charging pricing strategy based on uncertainty measure[J]. Power System Technology, 2018,42(1):96-102.
[4]	MURATORI M, RIZZONI G. Residential demand response: Dynamic energy management and time-varying electricity pricing[J]. IEEE Transactions on Power Systems, 2016,31(2):1108-1117. doi: 10.1109/TPWRS.2015.2414880 URL
[5]	赵亚杰. 电动汽车聚合服务商参与电网辅助服务的调控策略研究[D]. 北京: 华北电力大学, 2019.
	ZHAO Yajie. Research on the regulation strategy of electric vehicle aggregators participating in power grid ancillary services[D]. Beijing: North China Electric Power University, 2019.
[6]	邢强, 杨祺铭, 范军太, 等. 基于数据驱动方式和行为决策的电动汽车快充需求预测模型[J]. 电网技术, 2020,44(7):2439-2453.
	XING Qiang, YANG Qiming, FAN Juntai, et al. Electric vehicle fast charging demand forecasting model based on data-driven approach and human behavior decision-making[J]. Power System Technology, 2020,44(7):2439-2453.
[7]	孙庆凯, 王小君, 张义志, 等. 基于LSTM和多任务学习的综合能源系统多元负荷预测[J]. 电力系统自动化, 2021,45(5):63-70.
	SUN Qingkai, WANG Xiaojun, ZHANG Yizhi, et al. Multiple load prediction of integrated energy system based on long short-term memory and multi-task learning[J]. Automation of Electric Power Systems, 2021,45(5):63-70.
[8]	庞传军, 余建明, 冯长有, 等. 基于LSTM自动编码器的电力负荷聚类建模及特性分析[J]. 电力系统自动化, 2020,44(23):57-63.
	PANG Chuanjun, YU Jianming, FENG Changyou, et al. Clustering modeling and characteristic analysis of power load based on long-short-term memory auto-encoder[J]. Automation of Electric Power Systems, 2020,44(23):57-63.
[9]	肖浩, 裴玮, 孔力. 含大规模电动汽车接入的主动配电网多目标优化调度方法[J]. 电工技术学报, 2017,32(S2):179-189.
	XIAO Hao, PEI Wei, KONG Li. Multi-objective optimization scheduling method for active distribution network with large scale electric vehicles[J]. Transactions of China Electrotechnical Society, 2017,32(S2):179-189.
[10]	徐文波, 程华福, 白中华, 等, 多站融合模式下储能电站的优化设计和运行[J]. 供用电, 2019,36(11):84-91.
	XU WenBo, CHENG Huafu, BAI Zhonghua, et al. Optimal design and operation of energy storage power station under multi-station fusion mode[J]. Distribution & Utilization, 2019,36(11):84-91.
[11]	付华, 柳梦雅, 陈子春. 风光储电动汽车换电站多目标运行优化[J]. 电力系统及其自动化学报, 2016,28(4):38-43.
	FU Hua, LIU Mengya, CHEN Zichun. Multi-objective optimization of battery swapping station with wind photovoltaic and energy storage[J]. Proceedings of the CSU-EPSA, 2016,28(4):38-43.
[12]	刘伟佳, 文福拴, 马莉, 等. 计及电动汽车和可控负荷的需求侧能源交易机制[J]. 电力建设, 2019,40(11):24-30.
	LIU Weijia, WEN Fushuan, MA Li, et al. Demand-side transactive energy mechanism considering electric vehicles and controlable loads[J]. Electric Power Construction, 2019,40(11):24-30.
[13]	邢强, 陈中, 黄学良, 等. 基于数据驱动方式的电动汽车充电需求预测模型[J]. 中国电机工程学报, 2020,40(12):3796-3812.
	XING Qiang, CHEN Zhong, HUANG Xueliang, et al. Electric vehicle charging demand forecasting model based on data-driven approach[J]. Proceeding of the CSEE, 2020,40(12):3796-3812.
[14]	卢少平, 应黎明, 王霞, 等. 基于用户出行模拟的电动汽车快充站负荷预测及其优化调度[J]. 电力建设, 2020,41(11):38-48.
	LU Shaoping, YING Liming, WANG Xia, et al. Charging load prediction and optimized scheduling of electric vehicle quick charging station according to user travel simulation[J]. Electric Power Construction, 2020,41(11):38-48.
[15]	吴钉捷, 李晓露. 基于实时出行需求和交通路况的电动汽车充电负荷预测[J]. 电力建设, 2020,41(8):57-67.
	WU Dingjie, LI Xiaolu. Charging load prediction of electric vehicle according to real-time travel demand and traffic conditions[J]. Electric Power Construction, 2020,41(8):57-67.
[16]	MU Y F, WU J Z, JENKINS N, et al. A spatial-temporal model for grid impact analysis of plug-in electric vehicles[J]. Applied Energy, 2014,114:456-465. doi: 10.1016/j.apenergy.2013.10.006 URL
[17]	段雪, 张昌华, 张坤, 等. 电动汽车换电需求时空分布的概率建模[J]. 电网技术, 2019,43(12):4541-4550.
	DUAN Xue, ZHANG Changhua, ZHANG Kun, et al. Probabilistic modeling of battery swapping demand of electric vehicles based on spatio-temporal distribution[J]. Power System Technology, 2019,43(12):4541-4550.
[18]	MA J S, YU X Y, CHEN G, et al. Research on urban accessibility distribution areal model by Floyd algorithm and Kriging interpolation[C]// 2010 18th International Conference on Geoinformatics. June 18-20, 2010, Beijing, China. IEEE, 2010: 1-4.
[19]	邵尹池, 穆云飞, 余晓丹, 等. “车-路-网”模式下电动汽车充电负荷时空预测及其对配电网潮流的影响[J]. 中国电机工程学报, 2017,37(18):5207-5219.
	SHAO Yinchi, MU Yunfei, YU Xiaodan, et al. A spatial-temporal charging load forecast and impact analysis method for distribution network using EVs-traffic-distribution model[J]. Proceedings of the CSEE, 2017,37(18):5207-5219.
[20]	TANG D, QIN B, FENG X, et al. Effective LSTMs for target-dependent sentiment classification[C]// Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics. 2015: 3298-3307.

基于LSTM神经网络的电动汽车充电站需求响应特性封装及配电网优化运行

Demand Response Package Model of Electric Vehicle Charging Station Based on LSTM Neural Network and Optimal Operation of Distribution Network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献 20

相关文章 15

编辑推荐

Metrics

本文评价

[1]	郭祚刚, 徐敏, 李睿智, 袁智勇, 谈赢杰, 陈柏沅, 雷金勇, 刘念. 基于VaR理论的综合能源零售市场最优策略[J]. 电力建设, 2021, 42(6): 9-16.
[2]	林铭蓉, 胡志坚, 高明鑫, 陈锦鹏. 考虑需求响应和电动汽车负荷路-电耦合特性的配电网可靠性评估[J]. 电力建设, 2021, 42(6): 86-95.
[3]	段秦刚, 陈永椿, 王一, 吴明兴, 别佩, 董萍. 电力现货市场下激励型需求响应交易机制及出清模型[J]. 电力建设, 2021, 42(6): 145-156.
[4]	张高, 薛松, 范孟华, 张寒. 面向我国电力市场的需求响应市场化交易机制设计[J]. 电力建设, 2021, 42(4): 132-140.
[5]	王瀚琳, 刘洋, 许立雄, 陈贤邦, 谭思维. 考虑风电消纳的区域多微网分层协调优化模型[J]. 电力建设, 2020, 41(8): 87-98.
[6]	李恒强，王骏，蒋星燃，匡仲琴，徐石明，杨永标. 面向电网调峰辅助服务的空调负荷调控策略及其虚拟机组建模[J]. 电力建设, 2020, 41(7): 49-57.
[7]	赵晨，何宇俊，罗钢，龚超，赵越，张轩，陈启鑫. 基于泛在互联的虚拟电厂参与实时市场模型[J]. 电力建设, 2020, 41(6): 36-43.
[8]	苏俊妮，张鑫，赵俊炜，张锐，曲明辉，杨艳红. 计及需求响应和储能的电-气综合能源系统优化调度[J]. 电力建设, 2020, 41(5): 1-8.
[9]	都成，魏震波. 基于模糊激励型需求响应的微电网两阶段优化调度模型[J]. 电力建设, 2020, 41(5): 37-44.
[10]	李鹏，杨莘博，李慧璇，田春筝，李锰，谭忠富. 计及多能源多需求响应手段的园区综合能源系统优化调度模型[J]. 电力建设, 2020, 41(5): 45-57.
[11]	祁兵，叶欣，李彬，陈宋宋，李源非，石坤. 考虑IEC标准的用户侧电力物联网体系架构研究[J]. 电力建设, 2020, 41(5): 92-99.
[12]	许志恒，张勇军，余兆荣，王玕. 考虑激励型需求侧响应的电-气互联系统两阶段随机优化调度[J]. 电力建设, 2020, 41(4): 81-89.
[13]	张军六，樊伟，谭忠富，鞠立伟，德格吉日夫，杨莘博，孙婧霞. 计及需求响应的气电互联虚拟电厂多目标调度优化模型[J]. 电力建设, 2020, 41(2): 1-10.
[14]	周洪益，钱苇航，柏晶晶，卫志农，孙国强，臧海祥. 能源区块链的典型应用场景分析及项目实践[J]. 电力建设, 2020, 41(2): 11-20.
[15]	王澜灵, 刘俊勇, 许立雄, 贺心达, 唐杰, 房钢. 计及实时需求响应的综合能源信息物理系统可靠性分析[J]. 电力建设, 2020, 41(12): 25-40.