基于拓展交通均衡的高速路网充电桩优化部署

Optimized Deployment of Charging Piles Based on Extended Traffic Equilibrium in Highway

电动汽车(Electric Vehicles, EV)在中长途出行中仍面临行驶里程问题,完善高速公路充电设施有助于缓解驾驶人的充电焦虑。为刻画EV用户的充电选择行为,以EV剩余里程和充电服务区平均必要服务时间为自变量构建效用函数,进而提出了基于多元Logit的充电选择模型。结合交通分配模型与充电选择模型,建立了拓展交通均衡模型,用于获取每个服务区的充电需求。在此基础上,以系统充电服务时间最小化为目标,建立了基于拓展交通均衡模型的充电桩双层规划模型,用于获取每个服务区最优的充电桩数量。针对大型路网下拓展交通均衡模型求解效率低的问题,将其转化为非线性互补问题进行求解,采用遗传算法与GAMS软件联合求解双层规划模型。以湖南省部分区域高速路网为例,对所提充电桩优化部署模型进行可行性分析,结果显示:相较于均匀分布的充电桩部署方案,优化后的方案降低了53.1%的系统服务时间。此外,通过敏感性分析发现,提高EV用户对路网信息的感知水平有助于提高部分服务区充电桩利用率,减少空充电需求服务区的数量。研究成果对高速公路充电桩的部署具有指导意义,有助于提高高速公路充电服务水平。

Electric vehicles(EVs) face mileage issues during medium-to long-distance travel. Improving highway charging facilities can alleviate charging problems and improve travel service levels. To characterize the charging selection behavior of EV users, a discrete selection model based on a multivariate logit was established, where utility functions were proposed using the remaining mileage of EVs and the average necessary service time in the charging service district(CSD). An extended traffic equilibrium model was developed to obtain the charging demand for each CSD, which combined the traffic assignment model with the CSD selection model. Considering the minimization of the system charging service time, a bi-level programming model for charging stations was proposed based on an extended traffic equilibrium model to obtain the optimal number of charging stations in each CSD. However, solving traffic equilibrium models for large-scale road networks is difficult. The traffic equilibrium assignment problem was transformed into a nonlinear complementarity problem(NCP) for the solution. A genetic algorithm and a nonlinear complementarity algorithm were used to solve the bi-level programming model. The Hunan Provincial Expressway Network was used to verify the proposed model. The results show that the optimized distribution plan reduces the system service time by 53.1% compared with the uniformly distributed charging pile deployment plan. In addition, a sensitivity analysis revealed that an increase in EV users' perception of information can help improve the utilization rate of charging stations in some CSDs and reduce the number of empty CSDs. This study provides significant guidance for the deployment of charging facilities on highways to improve service levels.