Electric Vehicle Charging Station Planning Research Based On Multivariate Real-time Information

Electric Vehicel plays an important role in reducing carbon emissions and energy consumption in transportation,which has the advantages of energy saving,environmental protection and low noise,and its ownership is growing rapidly under the vigorous promotion of governments around the world.However,the travel characteristics of different types of EVs result in highly stochastic and uncertain spatial and temporal distribution of EV charging load users.As a result,EV charging stations,as public facilities providing charging services to EV users,need to build more charging facilities to dissipate the increasing EV charging load in the future.Proper planning of EV charging stations is essential for promoting the widespread use of EVs.This paper aims to investigate the travel characteristics of EV users,the stochastic and uncertain spatial and temporal distribution of EV charging load,as well as the prediction of charging demand information for charging stations.Additionally,the paper examines methods of planning EV charging stations.Firstly,the travel characteristics of different types of EV users are analyzed to characterize the travel characteristics of EV users in terms of travel origin and destination points,first trip moment of the day,stopping time and first trip end moment of the day,and a private car travel model based on the travel chain mode and a cab travel model based on the travel probability matrix are established.Secondly,a multi-source real-time information-based EV charging station demands model is proposed to predict the spatial and temporal distribution of EV user charging demand and charging station demand information.A dynamic transportation road network model is constructed based on the road network topology information,traffic congestion,and speed-flow relationship.The Dijkstra algorithm is improved to plan the shortest time-consuming travel routes for users,and the Monte Carlo method is utilized to predict the spatial and temporal distribution of EV users’ charging demand,taking into account the effects of vehicle speed and ambient temperature on vehicle power consumption.A charging station selection model is developed considering the dynamic charging demand of users,and charging information about charging stations is obtained statistically.Thirdly,a location and capacity sett model for EV charging stations is proposed,which takes into account the interests in both charging station operators and EV users.The actual charging information about each candidate site combination is obtained through the charging station demand model.Based on the M/M/c queuing theory method,the capacity allocation of each charging station are carried out,with the objective function of minimizing the sum of the construction and operation costs of the charging station and the economic losses of EV users(including time loss and power consumption).An optimization models on the location and capacity determination of electric vehicle charging stations is established,and a two-level particle swarms optimization algorithm is used to optimize the solution.The results show that the construction location and number of charging piles of the six planned charging stations in the region are reasonable and appropriate,ensuring the interests in both parties while minimizing the overall annual economic cost.The proposed planning method is reasonable and effective.