J Shanghai Jiaotong Univ Sci ›› 2023, Vol. 28 ›› Issue (1): 28-38.doi: 10.1007/s12204-023-2566-2

• • 上一篇    下一篇

基于充电态势感知的充电站负荷预测方法

  

  1. (1. 上海交通大学 电力传输与功率变换控制教育部重点实验室,上海 200240;2. 国网上海市电力公司 电力科学研究院,上海 200437;3. 兰州理工大学 电气工程与信息工程学院,兰州 730050)
  • 收稿日期:2022-03-02 出版日期:2023-01-28 发布日期:2023-02-10

Electric vehicle charging situation awareness for charging station ultra-short-term load forecast

SHI Yiwei1 (史一炜), LIU Zeyu1 (刘泽宇), FENG Donghan1 ∗ (冯冬涵), ZHOU Yun1 ∗ (周 云), ZHANG Kaiyu2 (张开宇), LI Hengjie3 (李恒杰)   

  1. (1. Key Laboratory of Control of Power Transmission and Conversion (Ministry of Education), Shanghai Jiao Tong University, Shanghai 200240, China; 2. Electric Power Research Institute, State Grid Shanghai Municipal Electric Power Company, Shanghai 200122, China; 3. School of Electrical Engineering and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, China)
  • Received:2022-03-02 Online:2023-01-28 Published:2023-02-10

摘要: 电动汽车将成为连接智能交通系统与电力系统的关键节点,先进的汽车电子技术提高了电动车的感知、计算和通信能力。这些技术为充电导航和充电负荷预测提供了条件,为充电拥堵和变压器过载问题提供了一种解决方案。本研究提出了一种保护车主隐私的充电态势感知框架和方法,仅依赖于出行服务商提供的公开信息,可开展充电站侧的超短期负荷预测。案例分析中,选取上海市浦东新区的一个区域开展实验,从在线地图平台和充电服务平台采集数据。结果显示,基于态势感知的方法,在低通信和算力要求下,各站的充电负荷可以提前1分钟以上得到准确预测。本工作为进一步研究充电站优化运行策略和电力市场环境中的交易模式提供了基础。

关键词: 电动汽车,智能交通系统,态势感知,充电负荷预测

Abstract: Electric vehicles (EVs) are expected to be key nodes connecting transportation–electricity–communication networks. Advanced automotive electronics technologies enhance EVs’ perception, computing, and communication capacity, which in turn can boost the operational efficiency of intelligent transportation systems (ITSs). EVs couple the ITS to the power system, providing a promising solution to charging congestion and transformer overload via navigation and forecasting approaches. This study proposes a privacy-preserving EV charging situation awareness framework and method to forecast the ultra-short-term load of charging stations. The proposed method only relies on public information from commercial service providers. In the case study, data are powered by the Baidu LBS cloud and EV-SGCC platform, and the experiment is conducted within an area of Pudong New District in Shanghai. Based on the results, the charging load of charging stations can be adequately forecasted more than 1 min ahead with low communication and computing power requirements. This research provides the basis for further studies on operation optimization and electricity market transaction of charging stations.

Key words: electric vehicle (EV), intelligent transportation system (ITS), situation awareness, charging load forecast

中图分类号: