Research On Moving Behavior Of Electric Vehicles After Charging Based On Hybrid Choice Model

In the era of carbon dioxide emissions peak and carbon neutrality goals,electric vehicle(EV)is a main fulcrum for energy transition and the realization of "double carbon" because of its advantages of high energy efficiency and low carbon emissions.With the rapid development of electric vehicles,the contradiction between supply and demand of charging facilities is increasing day by day.The average utilization rate of public charging facilities is low,and charging pile positions are inefficiently occupied sometimes.However,it is difficult to solve the problem fundamentally only by building public charging facilities.To better meet the charging demand of the electric vehicle users,effective optimization measures should be taken with limited resources of charging facilities.From the perspective of charging behavior,the driving away behavior of electric vehicle users after charging is studied,which is of great practical significance for the charging facility operators’ management and the effective use of social public charging resources.Based on the background of electric vehicle users charging in public charging facilities,this paper studies the decision-making behavior after charging,whether to immediately leave the charging pile and park in another parking space after charging,so as to avoid continuous occupation of charging facilities.The main contents and conclusions of this paper are as follows:1.This paper makes a comprehensive review of domestic and foreign research status on charging choice behavior,the influence of latent variables of attitude on charging choice behavior and charging choice behavior model.2.The mechanism of EVs moving behavior is analyzed.On the basis of defining the research scope of this paper,the decision-making process of EVs moving behavior is analyzed,and two latent attitude variables of social responsibility consciousness and perceived switching cost are introduced.The influencing factors of EV users’ choice of driving away are analyzed from three aspects: personal socioeconomic attributes,charging scenes and psychological attitudes.3.Data is obtained through questionnaire survey,and the latent variables of attitudes that were difficult to be directly observed are characterized by measurement index variables.The D-efficient experimental design method is used to generate 24 different choice scenarios and divided into 3 groups.The designed questionnaires combining revealed preference and stated preference are distributed and recycled to electric vehicle users.And data collection,sorting and inspection are carried out.Based on the survey data,the characteristics of EVs moving behavior are analyzed,and the relationship between scene variables,attitude variables,personal socioeconomic attributes and moving ratio is studied.4.On the basis of the variables’ characteristics analysis,considering the influence of latent variables of attitude,the traditional binomial logit model and hybrid choice model(HCM)including structural equation and discrete choice are respectively adopted to construct the choice model of moving behavior.Compared with traditional logit model,HCM significantly improves model fitting and model interpretation.According to the results of the model,the influences of personal attributes,charging type,charging time and place,time spent of moving,penalty cost and latent variables of attitude on EVs’ moving behavior are analyzed.The results show that the penalty cost,scene of charging at 9: 00 at workplace and social responsibility consciousness have a positive impact on the probability of EV users choosing to move their EVs,while time spent of moving,type of fast charging,perceived switching cost,scene of charging at 14:00 at shopping mall and income level have a negative impact on EV users’ moving behavior.5.On the basis of introducing latent variables of attitude to build HCM,the mixed logit(ML)model and latent class logit(LCL)model under HCM framework are built respectively with both latent variables of attitude and preference heterogeneity considered.The results of the two heterogeneity models are compared and analyzed.The results show that the goodness of fit of LCL model is significantly better than that of ML model.The heterogeneity of EVs’ moving behavior is mainly reflected in the perceived time and energy of moving behavior,and the sensitivity of EV users to penalty costs is generally consistent.Finally,according to the analysis results of choice model of moving behavior after charging,the strategy to improve the operation efficiency of charging facilities is proposed.