基于动态随机规划的共享电动汽车充电站规划  

Planning of Charging Stations for Shared Electric Vehicle Based on Dynamic Stochastic Programming

作　　者：汪洋[1] 汤宽武 赖科星 赵正晖 熊军 刘文亮 WANG Yang;TANG Kuanwu;LAI Kexing;ZHAO Zhenghui;XIONG Jun;LIU Wenliang(School of Electrical and Information Engineering,Jiangsu University,Zhenjiang 212013,Jiangsu Province,China;Hitachi ABB Power Grids,San Jose CA 95134,USA;State Grid Xiamen Power Supply Company,Xiamen 361000,Fujian Province,China)

机构地区：[1]江苏大学电气信息工程学院,江苏省镇江市212013 [2]日立ABB电网公司,美国加利福尼亚州圣何塞95134 [3]国网厦门供电公司,福建省厦门市361000

出　　处：《电网技术》2022年第9期3485-3495,共11页Power System Technology

基　　金：国家电网福建省电力有限公司科技项目(52134021005Z)。

摘　　要：电动汽车共享企业与充电站建设公司的合作呈递增趋势,探讨充电站建设与电动汽车共享服务调度的协同优化问题具有重要现实意义。该文采用动态随机规划框架整合了多周期下的各种运行场景,以反映交通网络、充电站建设和维护成本、司机和乘客行程等的不确定性。在共享电动汽车车队实时运营模型基础上,建立了共享电动汽车充电站布局与类型选择优化模型。该模型旨在通过确定充电设施最优位置和类型,来提高乘客福利以及降低运送总成本。以实际交通网络为例进行数值研究,验证了所提出模型的实用性和有效性。结果表明,通过对电动汽车充电站安装预算、权重因子和充电价格等影响因素的调整可以根据实际需求来确定共享电动汽车充电站布局与类型选择,通过计算随机解的值和对预测误差和成本分析的讨论,验证了动态随机框架的优越性。The cooperation between car-sharing enterprises and charging station construction companies is increasing,which calls for the co-optimization of the charging station planning and scheduling of the electric vehicles in the car-sharing business.In this paper,the dynamic stochastic programming framework is exploited to integrate a variety of operating scenarios under multi-cycles which reflect the uncertainties of transportation networks,the construction and maintenance costs of the charging stations,the drivers’ and passengers’ journeys,etc.Based on the real-time operation model of shared electric vehicle fleet,this paper develops an optimization model for the electric vehicle charging stations’ placements and type selections.The proposed model aims to improve the welfare of all passengers and reduce the total cost of fulfilling passengers’ pickup & delivery requirements simultaneously by determining the optimal locations and types of charging facilities.Using a real-world transportation network,the numerical study is conducted verifying the practicability and effectiveness of the proposed model.The results show that the locations and types of shared electric vehicle charging stations can be determined according to the actual demand by adjusting the installation budget of electric vehicle charging station,the weight factors and the charging prices,etc.The advantages of dynamic stochastic framework are verified by calculating value-of-stochastic-solution and discussing the forecasting error and cost analysis.

关 键 词：汽车共享 电动汽车 电动汽车充电站规划 动态随机规划 

分 类 号：TM721[电气工程—电力系统及自动化]