The Influence Of Electric Vehicle Charging Load Considering Traffic Flow On Distribution Network And Its Control Strategy

With the shortage of fossil energy,serious environmental pollution and other issues becoming increasingly prominent,the development and social popularization of electric vehicle(EV)has become an inevitable trend,and the market penetration is realized with high growth rate.The travel behavior of electric vehicle owners leads to the uncertainty of charging time and space,and the uncertainty of charging behavior of electric vehicles brings huge impact to the power system.Therefore,this paper studies from the following three aspects:Firstly,the traffic flow prediction of charging station based on combined forecasting model is studied.LSTM-SVR combination algorithm model is constructed;The 5min granularity data of PEMS traffic flow in California is processed into 15min granularity data.The traffic flow characteristics of PEMS weekdays and weekends are analyzed,and the model evaluation scenarios are determined;LSTM-SVR combined algorithm model is used to forecast the traffic flow on weekdays and weekends of a monitored section in PEMS data set.Simulation results show that LSTM-SVR combined forecasting model performs better than LSTM and BP single forecasting model in the prediction accuracy of traffic flow on weekdays and weekends.Secondly,the model of disordered charging load of electric vehicles based on traffic flow prediction model and its influence on the power flow of distribution network are carried out.Based on the traffic flow prediction model,the traffic flow data of each charging point of distribution network are predicted,and the number of electric vehicles with charging demand is calculated according to the conversion theory of charging traffic flow;The charging load of electric vehicle with daily time scale is calculated by Monte Carlo method;The active and reactive power of charging are input into the standard 33 node distribution network,and the influence of electric vehicle charging on peak valley difference,voltage deviation and network loss under different permeability is studied.The simulation results show that the large-scale electric vehicle grid connection will lead to the increase of load peak valley difference,voltage deviation and network loss,the higher the permeability of electric vehicles,the more obvious the above phenomena.Finally,the real-time optimal scheduling of cluster electric vehicles based on dynamic TOU price is studied.Based on the research results of the first two chapters,the real-time optimal scheduling model architecture and optimization strategy of electric vehicles are designed;Based on the initial state of charge(SOC)and power demand of electric vehicle(EV),the charging and discharging types of EV are analyzed;The dynamic time of use(TOU)price responding to the load curve fluctuation of distribution network is formulated;A real-time optimal scheduling strategy including distribution network loss and EV charging and discharging cost is constructed.The simulation results show that the proposed orderly charging and discharging model can stabilize the load curve to a certain extent and reduce the distribution network loss and user cost.This paper studies the influence of electric vehicle load considering traffic flow on the distribution network and its control measures.Through the prediction of electric vehicle traffic flow,it can capture the spatiotemporal uncertainty of vehicle owners’ travel,reduce the cumulative error of charging load modeling,accurately analyze its impact on the distribution network and formulate effective control measures,which has certain reference significance for stabilizing the safe operation of power grid.