摘要
针对电动汽车充电设施选址评价问题,通过分析充电设施使用现状及特点,提出了基于使用效率的充电设施建设选址评价模型。采用平衡数据集SMOTE算法生成数据样本,应用基于自助法重采样技术的随机森林算法构建模型,并应用ROC曲线来度量模型的分类能力与准确性。利用某市的充电设施进行了实例分析,评估候选站的使用效率作为模型应用,其结果表明,利用此模型对电动汽车公用充电设施的建设选址进行评价,准确地评估充电设施的使用率状况,提高了充电设施建设选址的针对性。
Aiming at the problem of electric vehicle charging facilities location evaluation,this paper presents a charging facilities construction site evaluation model based on the use efficiency by analyzing the current situation and characteristics of charging facilities. With the random forest algorithm that is based on boot-strap resampling technique,the data samples are generated by SMOTE( synthetic minority oversampling technique). The accuracy of classification prediction was measured by ROC( receiver operating characteristics) curve,which improves the classification accuracy of random forest algorithm. In this paper,we propose a structural analysis based on charging facilities in a city,and selected candidate stations for model application. The results show that this model can be used to evaluate the construction site of the electric charging facilities of electric vehicles,to evaluate the situation of the utilization rate of the charging facilities and to improve the targeting of the construction of the charging facilities.
作者
田贺平
孙舟
朱洁
陈海洋
赵宇彤
张明珠
陈雁
TIAN He-ping;SUN Zhou;ZHU Jie;CHEN Hai-yang;Zhao Yu-tong;ZHANG Ming-zhu;CHEN Yan(State Grid Beijing Electric Power Research Institute,Beijing 100075,China;State Grid Information & Telecommunication Group,Beijing 100085,China)
出处
《电工电能新技术》
CSCD
北大核心
2018年第8期9-16,共8页
Advanced Technology of Electrical Engineering and Energy
基金
国家电网公司科技项目(52020116000J)
关键词
电动汽车
充电设施
使用效率
选址
随机森林
electric vehicle
charging facilities
use efficiency
site selection
random forest