Research On The Method Of Urban Electric Vehicle Charging Infrastructure Layout

The development of electric vehicle(EV)is important for sustainable transportation to achieve the goal of the carbon emissions peak.However,both the limitation of EV’s driving range and the inefficiency of the charging service level cause users’mileage anxiety,which constrains EV usage.Thus,many countries have formulated corresponding EV charging facility(ECVF)development goals and policies to solve this problem.Transportation system of large cities plays an important role in developing sustainable transportation;research on EVCF layout in large cities has become a hot topic.In this dissertation,a comprehensive literature review is carried out from two aspects:charging choice behavior research and EVCF layout research,and the analysis of residents’characteristics based on multi-source data is conducted,on the basis of which,the following research are conducted:1.Study on the method of forecasting the production of charging demand(PCD)based on residents’characteristics.The PCD includes two parts:one is the estimation of the number of EV users with charging needs(PUC)and the other is the estimation of the amount of electric energy that EVCF users need(PEC).Among them,PUC refers to the total number of EV users who need charging after arriving at the destination per unit time;PEC aims to quantify the total electric energy of EVCF users need after arriving at the destination per unit time.Firstly,the characteristics of energy consumption distribution per mile of EVs are explored,on the basis of which,the logical decision mechanism of charging or not for EV users is developed considering EV’s energy consumption per unit mile and the residents’characteristics of daily trips,and charging behavior.Secondly,the method of forecasting PCD is developed to analyze the spatial and temporal distribution characteristics of PUC and PEC,which enables consider the influence of urban road traffic environment on EV’s energy consumption.Thirdly,the algorithm is designed for the application of the method of forecasting the PCD proposed;and correspondingly Agent-based simulation model is adopted to identify the PCD of EVCF users need after they arrive at end of trips.2.Study on the method of forecasting the attraction of charging demand(ACD)based on prospect theory.The ACD includes two points,one is the attraction of EVCF users(AOEU)and the other is the attraction of electric energy of EVCF users(AOEE).AOEU refers to the total number of EVCF users attracted by the EVCF at a location per unit time,AOEE refers to the total electric energy supplied by the EVCF at a location per unit time.Firstly,the decision mechanism of EV user’s joint choice behaviors of charging mode and charging location(CMCL)is generated by considering the subjective influencing factors(such as trip characteristics,urban road traffic,service level of EVCF)and objective influencing factors(such as charging behavior preference,cognition of information,risk perception).According to the decision mechanism of joint choice behaviors of CMCL of EVCF users,a prospect theory-based joint choice behavior model of CMCL is innovatively established,which aims to estimate the ACD of the EVCF at a location.The joint choice behavior model of CMCL includes following five steps:the definitions of perceived general cost and expected general cost of charging behaviors,the calculations of the perceived value,perceived weight,choice possibility of each charging behavior.In addition,the case algorithm is designed for the application of the method of forecasting the ACD proposed in this paper.Agent-based simulation model based on the joint choice behavior model of CMCL is applied to get the results of temporal and spatial distribution of the ACD for a given EVCF layout scheme.3.Study on the method of EVCF placement based on the coupling of supply and demand.Firstly,a map matching-based EVCF placement model(EVCFPM)is established based on multi-indicator comprehensive decision-making model,in order to get the candidates for EVCF layout.The evaluation indexes of EVCFPM are selected from urban road network traffic status,spatial and temporal distribution of PUC,trip attraction,land price.Secondly,a mathematical programming model(MPM)is built with the objective of minimizing the total loss cost of EV users with charging needs(loss cost includes the accessing cost to EVCF and waiting time cost for charging).The service intensity of EVCF at each candidate are considered as the constrain indexes,aims to define the constrains in MPM.In addition,the ACD forecasting model is embedded into the MPM,so as to consider the influence of the coupling of EVCF supply and EVCF demand on optimizing the EVCF layout scheme.In this paper,the static EVCF layout model(SEVCFLM)and dynamic EVCF layout model(DEVCFLM)are established respectively based on the principles of MPM.The case algorithm is designed for the application of SEVCFLM and DEVCFLM proposed in this paper.Compared with the existing studies,the method of EVCF layout proposed in this paper can visually quantify the benefits of the layout scheme and ensure the practicality and effectiveness of the layout scheme.More importantly,it helps to further analyze the concrete form of"the chicken-and-egg problem"between the supply and demand of EVCF.4.Using Xi’an,China as a case study,the proposed method of EVCF layout is demonstrated.Firstly,the case study of forecasting the charging demand is conducted by giving the specific value of parameters in PCD model,and the characteristics of spatial distribution of charging demand is analysis.Secondly,the scenario of SEVCFLM is defined(mainly includes the default values of parameters and default input data),and the optimal solution of SEVCFLM is generated by genetic algorithm(named as static scheme),and the characteristics of the optimal solution of SEVCFLM(such as the spatial distribution of charging facilities,the spatial and temporal distribution of ACD),the effectiveness of the optimal solution of SEVCFLM,and the psychological cost impact on the optimal solution of SEVCFLM are analyzed.Thirdly,12 scenarios of DEVCFLM are defined based on the analysis results of the optimal solution of SEVCFLM,and the corresponding optimal solution of DEVCFLM for each dynamic scenario is also respectively generated by genetic algorithm(named dynamic scheme).After comparing the spatial distribution characteristics of service capacity both of the current EVCF and the static scheme,the effectiveness and practicality of the method of EVCF layout proposed in this paper is verified.The results of the optimal solution of SEVCFLM show that:(1)the service performance of fast EVCF is much higher than that of slow EVCF;(2)the total loss cost of EV users with charging needs is mainly composed of the accessing cost to EVCF.The results of the optimal solution of DEVCFLM show that:(1)under the condition that the charging service level is maintained constant,the EVCF scale(or EVCF service capacity)is positively to the EV sharing rate and negatived to the charging power,while the battery capacity has no significant influence on it;(2)the average loss cost of EVCF user is stable at around 4.4 RMB for any scenarios of EVCF layout,which indicates that the method of EVCF layout proposed in this paper has strong robustness.According to the analysis results of static scheme and dynamic scheme,some suggestions are found for layouting EVCF as follows:(1)the spatial distribution characteristics of PUC and PEC should be taken into consideration in the EVCF layout;(2)the assessment of service attractiveness of EVCF can improve the scientificity of the EVCF layout scheme;(3)it is necessary to delineate the planning area and determine the level of the area for the scientific EVCF layout;(4)the deployment process of fast EVCF should take into account the regional land characteristics;(5)the stations of EVCF in the urban center should be small and dispersed as possible,and the stations of EVCF in the peripheral areas of the city should be large and concentrated;(6)Areas with high trip attraction and high trip accessibility should be installed more EVCF than other areas.Compared with existing studies,the contributions of this paper include:(1)developed a method of forecasting PCD based on residents’characteristics,so as to improve the accuracy of spatial and temporal prediction of PCD;(2)proposed a method of forecasting ACD based on prospect theory,so as to fill the research gap of assigning the ACD to different EVCF;(3)proposed a method of EVCF displacement based on supply-demand coupling,so as to further analyze the concrete form of"the chicken-and-egg problem"between the supply of EVCF and the demand of EVCF.