Research On Electric Vehicle Charging Load Forecasting And Charging Station Siting And Sizing

"Under the background of the goal of "carbon neutrality",the great progress of the EV industry has become an important strategic initiative for countries around the world to cope with the energy crisis and environmental pollution.On the one hand,large-scale electric vehicle access to charging infrastructure can cause a series of problems such as harmonic pollution,increased grid line losses and voltage imbalance in the distribution network.Thus,it is imperative to forecast the charging load of electric vehicles and study the adverse effects they cause.On the other hand,problems such as unreasonable charging station planning and shortage of charging facilities can seriously affect consumers’ willingness to purchase and charging experience of electric vehicles.To sum up,it is important to study the prediction of EV charging load and the location and capacity of charging stations in order to develop the EV industry in China and to achieve the "double carbon" goal.Firstly,this paper proposes a method for predicting the temporal distribution of the charging load of electric vehicles.Electric vehicles are classified into four types according to their usage;subsequently,the statistical analysis of the data is used to obtain the daily mileage of electric vehicles,so as to calculate their charging hours;considering the possibility of electric private cars charging in units during the daytime,while taking into account the influence of time-sharing tariffs,an electric vehicle charging load prediction model is established,and the Monte Carlo method is used to simulate the charging load models of different types of electric vehicles to obtain the load The time distribution curve is obtained.The simulation shows that the large-scale electric vehicles connected to the distribution network will have a negative impact on the smooth operation of the distribution network,while orderly charging can reduce the charging cost of vehicle owners while achieving peak and valley reduction of the distribution network to ensure its smooth operation.Secondly,to address the problem of inaccurate calculation of energy loss during the driving process of electric vehicles in previous studies of electric vehicle charging load prediction,a method for predicting the spatial and temporal distribution of energy consumption of electric vehicles under vehicle speed prediction is proposed.This paper takes the National Household Travel Survey(NHTS)as the base data,introduces the user travel chain,and models the time variables as well as the spatial variables in the travel chain;then,considering that the vehicle speed is not constant during the driving process of electric vehicles,a BP neural network is used to predict the driving speed of electric vehicles,and then,according to the In addition,a fuzzy inference system is established with three input variables: battery state of charge(SOC),next trip driving time and parking time,to generate a charging probability describing the user’s willingness to charge;finally,the load distribution curves of EVs in different charging areas are obtained through Monte Carlo method simulations.The reasonableness and superiority of the method in this paper are proved by simulation results.Finally,in order to address the problem of inaccurate analysis of EV users’ charging behaviour in previous studies,a method for siting and sizing EV charging stations based on a fuzzy inference system is proposed.In this paper,the battery SOC value and next driving mileage are taken as the influencing factors of users’ charging probability,and the fuzzy inference system is used to quantify vehicle owners’ willingness to charge,calculate the total charging load demand in the planning area,and then derive the range of the number of charging stations that need to be built;in addition,the initial station coordinate locations of different numbers of charging stations are initially set using the geometric division method,and the Voronoi diagram is used to divide the service range of each After that,the coordinates of each charging station are determined by genetic algorithm with the objective function of minimizing the charging cost from the charging demand node to the station.The simulation example shows that the method of this paper is feasible and reasonable,and is a guide to the planning of today’s electric vehicle infrastructure.