Charging Choice Behavior Based Location Optimization For EV Charging Facilities

Electric vehicles(EVs)become a significant direction for cities5 low carbon traffic development,but the relative lower driving range has hindered the promotion of EVs.Therefore,developing a larger amount of charging network is a major method for promoting EV development.Nowadays,however,public charging facilities in China are facing the problem of lower utilization rate,and one of the reasons would be a lack of charging demand estimation and location planning method for charging facilities which considers EV drivers'daily trips and activities.It is necessary to study the charging demand estimation and location planning method in light of comprehensively considering EV drivers'daily activities and charging behaviors.The research of charging choice behavior,charging demand estimation,and location method of charging facilities are summarized,and the pros and cons of these research are analyzed.In light of that,the data was collected by using an online SP survey,involving risk attitude and decision-making mechanism,and the charging choice behaviors of EV drivers are analyzed.Then,the charging decision process of EV drivers and the charging demand estimation based on trip chain are proposed.Assuming that EV drivers will maintain their existing trips and activities,location optimization model for charging facilities is proposed,and multiple scenarios are analyzed based on taking Beijing as a case study.In detail,the research and findings of this paper are summarized as follows:(1)Characteristic analysis of EV charging choice behaviorBased on SP survey from EV drivers,the choice date is obtained by scenario experimental design,attitudinal questions,and socio-demographic information,and the relationship between the above factors and charging frequency is analyzed.The respondents of our survey are younger,higher educated,and have a lower income compared to the general population.The results show that SOC,charging price,parking price,excess range,the activity of the destination,and minimum SOC may affect charging choice behavior.(2)Modeling EV drivers'charging choice behaviorIn light of the characteristic analysis,involving risk attitude and preference heterogeneity,binary logit(BL),latent class BL,and hybrid choice model of charging choice behaviors are proposed.Then,the decision heuristics of attribute non-attendance is considered,analyzing the influence of socio-demographics,charging condition,vehicle condition,destination activity,successive activity,and risk attitude on charging choice preference.The results show that the EV drivers are divided into risk averse class,focusing primarily on the amount of excess or buffer range they have available to complete their next trip,and risk seeking class,balancing price against their current state of charge.(3)Establishing location optimization model aiming to maximize EV drivers'daily trips and activitiesLocation optimization models for static and dynamic charging demands are proposed.In charging demand estimation of the latter model,the charging decision process is established,and EV drivers make charging decisions based on existing trip chains and dynamic occupancy conditions of chargers.Then,the occupancy conditions of chargers are updated simultaneously according to the charging decisions,achieving the interaction between charging behaviors and charging facilities.In location optimization,assuming that EV drivers will not change their daily trips and activities,the location optimization models aiming to maximize the daily activities of EV drivers are proposed while considering the restriction of the entire number of chargers.The control variables of optimization model for static charging demand are based on traffic analysis zones(TAZs),assuming that the charging facilities could fully fulfill the charging demand.Taking a step forward,the control variables of optimization model for dynamic charging demand are the numbers of chargers in TAZs,considering the time variation of charging demand and the occupancy conditions of chargers,being applicable for optimize the exact numbers of chargers in each TAZ by giving a fixed number of entire chargers.(4)Analyzing the location optimization of charging facilities with their demand-fulfilled effect based on a case study of BeijingUsing the actual road network and drivers9 daily travel data,the region within the Fourth Ring Road in Beijing is taken as a study area.Then,the location solutions ofpublic chargers are proposed and analyzed,and the spatial distribution and service levels of chargers are discussed.The existing and optimizing location of chargers are compared,and scenarios under different numbers of entire chargers and ownership rates of home chargers are analyzed.The results show that optimized locations could effectively reduce the missed trips and improve the utilization rate(from 7.2%to 15.3%).In the spatial perspective,the optimized locations distribute more evenly and well matches the distribution of charging demand.The result of sensitivity analysis shows that increasing more public chargers or home chargers could both reduce the missed trips,while it will highly reduce the utilization rates of chargers.Therefore,considering the demand fulfilled effect of charging network and the reduction of utilization rates simultaneously is significant when deploying charging network.