Research On Location And Capacity Of Charging Station Based On Electric Vehicles Charging Load Prediction

The establishment of the ’double carbon’ goal has promoted the development of the electric vehicle industry.However,the construction speed of charging stations lags behind the growth rate of electric vehicle charging demand.The charging problem of electric vehicles has become an obstacle to the development of the electric vehicle industry.In this context,how to scientifically and reasonably determine the location and capacity of electric vehicle charging stations has become extremely important.Therefore,this paper studies the location and capacity of charging stations by analyzing the spatial and temporal distribution of charging load and predicting the demand of charging load.The details are as follows:(1)The spatial and temporal distribution of electric vehicle charging load is studied based on user travel behavior and charging behavior.Firstly,a user travel model is established based on the travel chain structure and functional area attributes to describe the user ’s travel behavior;Secondly,considering the different charging habits of electric vehicle users,a user charging decision model is established based on the charging price,state of charge and dwell time of the destination;Then,Monte Carlo simulation is used to simulate the travel behavior and charging behavior of electric vehicles in the region,so as to obtain the spatial and temporal distribution of electric vehicle charging load;Finally,the effectiveness of the method is verified by simulation experiments.(2)Electric vehicle charging load demand forecasting based on improved Random Forest Regression(RFR)model.Firstly,aiming at the problem that the traditional random forest regression model has low prediction accuracy due to improper parameter selection,a prediction method based on Sparrow Search Algorithm(SSA)improved random forest regression model is proposed;Secondly,the sparrow search algorithm is used to optimize the number of decision trees in the random forest regression model and the number of split features selected by each decision tree by using the sparrow search algorithm with strong optimization ability and fast convergence speed,so as to solve the parameter selection problem of the random forest regression model;Then,a charging load prediction model based on Sparrow Search AlgorithmRandom Forest Regression(SSA-RFR)is established to predict the charging load;Finally,the simulation results show that the SSA-RFR charging load prediction model has higher prediction accuracy than the traditional RFR model.(3)Research on location and capacity of electric vehicle charging station based on weighted K-Means clustering.Firstly,the charging load in the region is divided and weighted in time and space.In order to minimize the construction,operation and maintenance costs of charging stations and the charging costs of users,a charging station location and capacity model is established with the service radius,capacity and quantity of charging stations as constraints;Secondly,a weighted K-Means clustering algorithm is proposed to solve the model,and the optimal location and capacity scheme is selected by calculating the contour coefficients of different schemes;Finally,experiments show that the feasibility of the location and capacity model of electric vehicle charging station based on weighted K-Means clustering.