基于城市网格属性划分的电动汽车充电需求预测  

EV charging demand prediction based on city grid attribute division

作　　者：张美霞 徐立成 杨秀 ZHANG Meixia;XU Licheng;YANG Xiu(School of Electrical Engineering,Shanghai University of Electric Power,Shanghai 200090,China)

机构地区：[1]上海电力大学电气工程学院,上海200090

出　　处：《电测与仪表》2025年第3期10-19,共10页Electrical Measurement & Instrumentation

基　　金：上海市科委资助项目(18DZ1203200)。

摘　　要：为提高电动汽车充电需求预测研究中对用户-交通二者耦合关系描述的精确性,文章提出一种基于城市网格属性划分的电动汽车充电需求预测方法。将网约车行程数据和基于Python爬取的城市兴趣点数据融合,对研究区域进行功能区精确划分,进而挖掘得到居民出行规律和高频行驶路径等特征数据;考虑电动汽车用户的路径选择行为,结合道路通行数据构建基于用户有限理性的双层路径选择模型;考虑电动汽车的行驶特性和充电特性,建立完整的充电需求预测模型,并将该模型应用到成都市二环区域中,进行不同区域和不同场景下充电需求预测可行性验证。In order to improve the accuracy of the description of the coupling relationship between users and traffic in the study of electric vehicle charging demand prediction,a method of electric vehicle charging demand prediction based on city grid attribute division is proposed.The fusion of online vehicle trip data and Python-based urban in-terest point data is used to accurately divide the study area into functional areas,and then,the characteristics data such as travel patterns of residents and high frequency driving paths are mined.The path selection behavior of elec-tric vehicle users is considered,and a two-layer path selection model based on limited rationality of users is con-structed by combining road traffic data.The driving and charging characteristics of electric vehicles are considered,and a complete charging demand prediction model is built.And the model is applied to the Second Ring Road of Chengdu to verify the feasibility of charging demand in different areas and scenarios.

关 键 词：电动汽车 城市兴趣点 网约车数据 路径选择 充电需求预测 

分 类 号：TM73[电气工程—电力系统及自动化]