关键词:
充电桩布局
K-Means聚类
LSTM-ARIMA模型
遗传算法
整数规划
摘要:
在“双碳”目标的引领下,发展新能源汽车成为了我国汽车产业高质量发展的战略选择,具有广阔的发展空间。然而,充电桩作为新能源汽车的配套基础设施之一,存在着区域分布不均、数量较少的现象。本文以长沙市为例,考虑其9个区的街道数量、占地面积、人口数量和GDP四种因素,利用K-Means聚类方法将该市划分为现代化发展区、生态友好区和传统经济区。进一步,结合该市新能源汽车的历史销售数据,采用LSTM-ARIMA模型预测了该市2024年各月新能源汽车的销售量。接着,以最大化投资回报为目标,考虑地方财政支出、充电桩总数量、充电桩成本范围和用户充电习惯等四个因素,构建了整数规划模型。最后,利用遗传算法求解得出了长沙市各区各街道的充电桩数量。结果表明,相较于2023年,充电桩的空间利用率提升了21.47%。与新能源汽车销售量预测的针对各区各街道的充电桩数量分配增长趋势一致,说明该布局方案在满足充电需求和提升基础设施利用效率方面具有积极作用,提出了一种合理的布局方案。Under the guidance of the “dual carbon” goal, the development of new energy vehicles has become a strategic choice for the high-quality development of China’s automotive industry, with broad development space. However, as one of the supporting infrastructure for new energy vehicles, charging stations have uneven regional distribution and limited quantity. This article takes Changsha City as an example and considers four factors: the number of streets, land area, population, and GDP in its nine districts. Using K-Means clustering method, the city is divided into modern development zone, eco-friendly zone, and traditional economic zone. Furthermore, based on the historical sales data of new energy vehicles in the city, the LSTM-ARIMA model was used to predict the sales volume of new energy vehicles in each month of 2024. Next, with the goal of maximizing investment return, an integer programming model was constructed considering four factors: local fiscal expenditure, total number of charging stations, range of charging station costs, and user charging habits. Finally, the number of charging stations in each district and street of Changsha city was determined using genetic algorithm. The results show that compared to 2023, the space utilization rate of charging piles has increased by 21.47%. Consistent with the growth trend of the distribution of charging piles for each district and street predicted by the sales volume of new energy vehicles, this layout scheme has a positive effect on meeting charging demand and improving infrastructure utilizatio