Research On Behavior Characteristics Modeling And Optimal Scheduling Strategy Of Electric Vehicle

In recent years,environmental degradation and energy shortages have become increasingly serious,and electric vehicles(EVs)have developed rapidly as a green energy transportation tool.In the case of disorderly charging,the centralized access of large-scale EVs will have an impact on the safe and stable operation of the power grid.The charge and discharge characteristics of EV batteries are similar to energy storage,with the advantages of large groups,flexible dispatching,and the potential for two-way interaction with grid energy.Through Vehicle to Gird technology and optimization of charging and discharging,EV can assist the power grid in peak regulation,frequency regulation,etc.,to ensure the stable and economic operation of the power grid.This paper took electric vehicles as the object,and carried out research work in three aspects: charging behavior characteristic modeling,load spatiotemporal distribution prediction,and charging and discharging optimization scheduling strategy.First,established EVs’ charging and travel behavior characteristics model.The variables of EV travel behavior and charging behavior are summarized into three types of factors: power,time and space.Based on the actual monitoring data set,analyzed the characteristics of variables,jointly evaluated the fitting effect of various distribution forms,and established the probability distribution model of each variable.Time factor variables considered EVs’ type,typical day,typical functional area and other characteristics,and refined the variable type.Considering the coupling of road network and power grid due to spatial factors,established four models,including traffic road network model,EV spatiotemporal transfer model,vehicle real-time power consumption model,and load characteristic model.Comparing the monitoring data of first-tier cities with simulated vehicle travel and spatial transfer,the simulation results prove the effectiveness of the model.Then,studied the prediction and simulation of spatio-temporal distribution of EV disordered charging.In the single time factor load prediction,used the Monte Carlo simulation method based on the charging probability at the moment to predict the charging load distribution of different types,different functional areas and different typical days.Load prediction considered the coupling of road network and power grid,the user’s charging willingness and EV spatio-temporal transfer jointly determine the charging behavior change,and the spatio-temporal distribution of load and its impact on the power grid are simulated by taking the IEEE33 standard distribution network as an example.Finally,studied the multi-objective optimization scheduling strategy of EV charge and discharge.Clustered clustering based on the adjustable potential of a single EV,and discussed the adjustable margin on the classification of node EVs’ clusters,which provided constraints for subsequent optimization.Taking the node as the optimization unit,analyzed the basic load characteristics of nodes in different functional areas,established a multiobjective optimal dispatching model for charge-discharge,with the goal of minimizing peak-to-valley difference,minimizing load fluctuation and lowest charging cost on eviction side of the power grid,and the optimal dispatching strategy is solved by particle swarm algorithm.Designed IEEE33 standard distribution network study to verify effectiveness of optimization strategy from the aspects of node target,system target,and system voltage drop.