Research On Electric Vehicle Charging Demand Forecast And Location Selection Of Charging Station Based On Data-driven

At present,all countries are facing a serious energy and environmental crisis.With the advantages of energy saving and environmental protection,electric vehicles(EVs)have become the focus of attention in various countries.The energy supply of EV comes from electric energy,and its travel has high flexibility and randomness.When large-scale EVs are connected to the power grid,it will not only increase the burden of the power grid,but also bring many hidden dangers to the safe operation of the power grid.At the same time,due to the construction of charging stations do not fully consider various influencing factors,resulting in difficult and expensive charging problems frequently.The contradiction between unbalanced and inadequate development of charging stations and charging demand is increasing day by day.Therefore,it is of great practical significance to carry out the research on charging load prediction and layout optimization of charging stations.Based on this,this paper conducts research from the following aspects:(1)Prediction of spatial-temporal distribution of EV charging demand.Accurate prediction of EV charging demand is a prerequisite for site selection of charging facilities.By obtaining the daily driving track data of EVs,the user’s travel characteristic parameters and vehicle charging characteristic parameters are extracted,including single trip track,daily first trip time,trip origin and destination,stay time,basic vehicle parameters,etc.The user travel activity chain and charging model of single EV were built,and the probability density function of travel characteristics was obtained,and then the charging demand prediction model based on Monte Carlo simulation was constructed.Finally,the influence of the change of vehicle size on the temporal and spatial distribution of EV charging load is analyzed.(2)A charging station siting model is constructed.In order to understand the influence of charging demand on site selection,this paper establishes two different site selection models by drawing on the ideas of P median model and set coverage model:the location selection model aiming at minimum cost considering the static charging demand;according to the prediction results of charging demand,a location model aiming at the minimum comprehensive cost is established,in which the influence of different location schemes on carbon emissions is considered.The site selection model is considered from two perspectives: user and enterprise.Consider the cost factors of charging station from the operator’s point of view: construction cost,operation cost,land cost,etc.;on the other hand,the convenience of charging and the time cost of charging are considered from the user’s perspective.Meanwhile,considering the problem that new charging stations are needed to meet the increasing charging demand,a method is designed to quickly determine the location of new charging stations by using the hollow circle property of Voronoi diagram.(3)The model solving algorithm is improved.The location problem of charging station conforms to the P median problem in the discrete location model.Firstly,the immune algorithm and particle swarm optimization algorithm are combined and improved,which can solve the problems of slow speed,poor accuracy and many adjustment parameters when using a single algorithm.The improved algorithm is applied to the location model of charging station considering the static charging demand,and the effectiveness of the improved algorithm is verified by simulation experiments.Then the improved whale optimization algorithm is used to solve the problems of low accuracy and insufficient searching ability of the original algorithm.The effectiveness of the improved algorithm is analyzed by 15 benchmark test functions,and the improved whale optimization algorithm is applied to solve the multiobjective location model.In summary,this paper focuses on the siting problem of electric vehicle charging stations under dynamic charging demand.A siting layout model is established considering the interests of both users and enterprises.Based on the characteristics of large-scale,multi-constrained and non-linear charging station siting problem,the model solving algorithm is purposely improved to improve the convergence speed and solution accuracy of the algorithm.It provides a reference for the prediction of EV charging demand and the planning of charging station.