Electric Vehicle Charging Load Forecasting And User Portrait Considering Road Network-power Grid Interaction

The charging load generated by the rapid development of electric vehicles(EV)will bring challenges to the power grid,so it is necessary to strengthen the interaction between EV and power grid.Therefore,from the perspective of EV users,this paper proposes charging time and charging location recommendation strategies to reduce charging cost;Then,on the basis of determining the charging time and location,the information interaction between road network and power grid is further considered to accurately predict the temporal and spatial distribution of EV charging load,which provides data basis for the interaction between EVs and power grid;Finally,through the charging data,the user portraits of EVs considering the market response potential are established to determine the scheduling objects for the interaction between EVs and power grid,so as to achieve precision marketing.This paper mainly focuses on the charging recommendation,charging load forecasting,user portrait and other charging related issues of EVs considering the interaction between road network and power grid.(1)From the user’s point of view,this paper proposes an optimal charging time and location recommendation strategy for EVs with the goal of reducing charging cost.Firstly,based on the trip chain and OD matrix,the travel behavior models of electric private cars and electric taxis are established;secondly,the time value of EVs is calculated,the charging time and time of use(TOU)price are considered,and the optimal charging time period is obtained with the objective of minimizing the unit electricity cost;secondly,the driving time,queuing time,charging time and time value are considered in the optimal charging time period Calculate the time cost,consider the demand and charging price to calculate the electricity cost,and recommend the best charging location to minimize the total cost.(2)This paper presents a method of EV charging Load Forecasting Considering the interaction between road network and power grid.Firstly,considering the charging recommendation strategy and other charging methods,the charging characteristics of EVs are determined;secondly,based on the car following model,the micro traffic model of the vehicle driving process in the road network is established;secondly,considering the grid voltage,the charging price of the charging station is determined,and the network grid interactive charging load forecasting framework driven by the price is established;finally,Monte Carlo simulation is used Methods the travel behavior and charging situation of EVs were simulated,and the temporal and spatial distribution of charging load of EVs was predicted.(3)This paper proposes a method to build user portrait of EV considering market response potential.Firstly,the market response potential index system of EV users is constructed from the three dimensions of electricity response potential,time response potential and spatial response potential;secondly,the residual capacity and TOU price are selected as the characteristic quantity,and the users sensitive to TOU price are identified based on support vector machine;secondly,the distance,time and electricity charge are selected as the characteristic quantity,and the charging potential is identified based on neural network Finally,through the analysis of the predicted charging data,the portraits of EV users with different market response potential are obtained to verify the effectiveness of the proposed method.(4)Finally,the main research results are summarized,and the problems that need to be further studied in the future are analyzed and prospected.