Journal of Frontiers of Computer Science and Technology ›› 2022, Vol. 16 ›› Issue (2): 428-437.DOI: 10.3778/j.issn.1673-9418.2008101

• Artificial Intelligence • Previous Articles     Next Articles

Optimization of Charging Station Network Based on Maximum Usage Benefit

MENG Xiangfu+(), YANG Yu, ZHANG Yongku, ZHANG Xiaoyan, CHEN Roubing, WANG Ze   

  1. College of Electronic and Information Engineering, Liaoning Technical University, Huludao, Liaoning 125105, China
  • Received:2020-07-21 Revised:2020-09-30 Online:2022-02-01 Published:2020-11-06
  • About author:MENG Xiangfu, born in 1981, professor, Ph.D. supervisor. His research interests include spatial data management, recommender system and Web database query.
    YANG Yu, born in 1994, M.S. candidate. Her research interests include city computing, deep learning and data mining.
    ZHANG Yongku, born in 1974, associate professor. His research interests include database system and data mining.
    ZHANG Xiaoyan, born in 1983, engineer. Her research interests include spatio-textual data query and city computing.
    CHEN Roubing, born in 2001. Her research interests include data analytics and city computing.
    WANG Ze, born in 2000. His research interests include data mining and machine learning algorithms.
  • Supported by:
    National Natural Science Foundation of China(61772249);Research Project of Liaoning Provincial Education Department(LJ2019QL017);Research Project of Liaoning Provincial Education Department(LJKZ0355)

面向最大使用效益的充电站网络优化方法

孟祥福+(), 杨玉, 张永库, 张霄雁, 陈柔冰, 王泽   

  1. 辽宁工程技术大学 电子与信息工程学院,辽宁 葫芦岛 125105
  • 通讯作者: + E-mail: marxi@126.com
  • 作者简介:孟祥福(1981—),男,辽宁朝阳人,教授,博士生导师,主要研究方向为空间数据管理、推荐系统、数据库查询。
    杨玉(1994—),女,广西贺州人,硕士研究生,主要研究方向为城市计算、深度学习、数据挖掘。
    张永库(1974—),男,辽宁阜新人,副教授,主要研究方向为数据库系统、数据挖掘。
    张霄雁(1983—),女,山东烟台人,工程师,主要研究方向为空间-文本数据查询、城市计算。
    陈柔冰(2001—),女,辽宁朝阳人,主要研究方向为数据分析、城市计算。
    王泽(2000—),男,辽宁营口人,主要研究方向为数据挖掘、机器学习算法。
  • 基金资助:
    国家自然科学基金(61772249);辽宁省教育厅科学研究项目(LJ2019QL017);辽宁省教育厅科学研究项目(LJKZ0355)

Abstract:

China’s government has built a certain scale of charging station network. To address the problems of low charging station network utilization, the existence of a large number of redundant charging stations, and the diffi-culty of charging, this paper proposes a data-driven charging station network optimization approach. Firstly, this paper simulates the charging behavior of EV (electric vehicle), establishes queuing systems for charging stations in different time stamps, and then estimates the arrival rates among charging stations. On this basis, the spatial characteristics of urban EV charging behavior are analyzed, which is used to explore the urban EV charging hotspots. Then, the interactions among competitive dependencies, geographic features and user charging preferences among charging stations are modeled, and a scoring function is proposed to assess the usage benefit of charging station in the network. Lastly, a charging station network optimization model which aims at maximizing the usage benefit of charging station network is developed, and a heuristic network expansion algorithm based on charging hotspots is also presented to solve the model, and thus the optimal charging station network layout can be obtained. Taking a typical urban area as an example to conduct experiments, the experimental results demonstrate that the method proposed in this paper can not only improve charging station utilization while eliminating redundant stations, but also identify charging station network congestion areas, which can provide decision support for government planning departments to solve the charging difficulty.

Key words: optimization of charging station network, electric vehicle (EV), redundant charging station

摘要:

我国政府目前已建成一定规模的充电站网络。针对充电站网络利用率低,存在大量冗余站点和电动汽车充电难等问题,提出了一种数据驱动的充电站网络优化方法。首先,该方法模拟电动汽车充电行为,对不同时间戳内的充电站分别建立队列系统,进而估计充电站间的到达率情况。在此基础上,分析城市电动汽车的充电行为空间特征,用于挖掘城市电动汽车的充电热点。然后,对充电站间的竞争依赖关系、地理位置特征及用户充电偏好间的相互作用进行建模,进而提出了充电站在网络中的使用效益评分函数。最后,建立了以最大化充电站网络使用效益为目标的充电站网络优化模型,并提出了基于充电热点的启发式网络扩展算法进行模型求解,从而获取最佳充电站网络布局。以一个典型的城区为例进行的实验测试结果表明,该方法不仅能在消除冗余站点的同时提高充电站利用率,而且能够识别充电站网络拥堵区域,为政府规划部门解决充电难问题提供了决策支持。

关键词: 充电站网络优化, 电动汽车(EV), 冗余充电站

CLC Number: