基于用地属性的充电站备选点模型构建

黄禹¹，刘杰^1*，黄健畅¹，林荔萍²

1.山东交通学院交通与物流工程学院，山东济南250357；2.山东省交通科学研究院，山东济南250031

摘要：为提高电动汽车充电基础设施资源配置效率并降低运营成本，提出一种融合用地属性与多目标优化的充电站选址方法，基于兴趣点（point of interest，POI）分布数据，采用K-means聚类算法作为预处理方法初步确定充电站备选点；通过计算各充电站备选点覆盖范围内POI类型的权重与数量，得到各充电站备选点的重要度，建立最大化充电站备选点重要度目标函数；同时考虑充电站备选点建设、运营成本及用户在途损耗成本，建立最小化充电站总成本目标函数，构建充电站双目标规划选址模型；分别采用传统麻雀搜索算法（sparrow search algorithm，SSA）和融合动态自适应权重、反向学习策略和柯西变异的改进麻雀搜索算法（improved sparrow search algorithm，ISSA）求解模型，以济南市为例进行实例分析。结果表明：ISSA的充电站选址布局能有效避免过于集中或分散问题，服务覆盖范围与布局均衡性显著优于SSA；与SSA相比，ISSA的选址方案使总成本降低7.43%，充电站重要度提高34.34%；ISSA求解得到的充电站的服务缓冲区在各类用地类型上的覆盖率均优于SSA，尤其在商业用地方面，ISSA方案的覆盖率比SSA方案大6.79百分点。通过充电站双目标规划选址模型及优化求解算法能有效提高充电站布局合理性与空间资源配置效率。

关键词：充电站；备选点；重要度；ISSA

Construction of a candidate-site model for charging stations based on land-use attributes

HUANG Yu¹, LIU Jie^1*, HUANG Jianchang¹, LIN Liping²

1.School of Transportation and Logistics Engineering, Shandong Jiaotong University, Jinan 250357, China;

2.Shandong Transportation Research Institute, Jinan 250031, China

Abstract: To improve the resource allocation efficiency of electric vehicle charging infrastructure and reduce operational costs, a charging station site selection method integrating land-use attributes and multi-objective optimization is proposed. Based on the distribution data of points of interest (POI), the K-means clustering algorithm is used as a preprocessing method to initially determine potential charging station locations. By calculating the weight and number of POI types within the coverage area of each potential site, the importance of each site is determined, and an objective function is established to maximize the importance of the charging station candidate points. At the same time, considering the construction and operational costs of the charging stations and the travel loss costs for users, an objective function to minimize the total cost of the charging stations is formulated. A bi-objective planning model for the charging station site selection is built. The model is solved using both the traditional sparrow search algorithm (SSA) and an improved sparrow search algorithm (ISSA), which integrates dynamic adaptive weights, reverse learning strategies, and Cauchy mutations. An example analysis for Jinan city is presented. The results show that the ISSA-based charging station layout effectively avoids issues of excessive concentration or dispersion, with significantly better service coverage and layout balance than SSA. Compared to SSA, the ISSA site selection plan reduces the total cost by 7.43% and increases the charging station importance by 34.34%. The service buffer coverage of the charging stations solved by ISSA outperforms SSA in all types of land-use areas, especially in commercial areas, where the coverage rate of the ISSA plan is 6.79 percentage points higher than that of the SSA plan. The charging station bi-objective planning model and the optimized solving algorithm can effectively improve the rationality of the charging station layout and spatial resource allocation efficiency.

Keywords: charging station; candidate site; importance; ISSA