Hierarchical Coordinated Optimal Scheduling Strategy For Large Scale Electric Vehicles

With the wide application of Electric Vehicle(EV),large-scale EV will become a high power load.If access to the distribution network is not controlled,it may have a negative impact on the operation and management of the power system,and it is easy to coincide with the peak of the daily routine load peak,which is not conducive to the economic operation of the power grid.Considering the high flexibility of the EV charging load,this paper studies the modeling and Simulation of EV large-scale access to the grid and the optimal scheduling strategy,through the planned charging of the EV charging station,and the response charge and discharge function of the EV,to realize the scheduling of large-scale EV in order to slow the solution of the EV of the grid in the day peak load period.Firstly,according to the charging behavior of different kinds of vehicles,an EV load demand calculation model is designed.The Monte Carlo sampling method based on data statistics is used to simulate large scale EV charging load.Based on the development of EV in Changsha in the next thr-ee years,the full day charge power time of EV is obtained by example simulation.Curves and EV charging load related parameters provide data basis for subsequent ordered charging load control.The decision tree method is used to classify the large-scale EV load,define the EV charge/discharge margin,distinguish the EV which can participate in the intelligent charge and discharge according to the charging information,and identify the charging mode that other EV should adopt.The example proves that it is effective,and the decision tree classifier is optimized to make the classification way more convenient.Secondly,a hierarchical EV charge and discharge control model is proposed.The model is divided into two layers,the upper layer is lower for the day before scheduling,real-time scheduling.In the daily scheduling of the upper level,the power purchase problem of the EV charging station is considered as a potential game problem,and a potential function is designed to jointly optimize the cost of power generation and the revenue of the charging station.It is proved by proving that there is a unique Nash equilibrium in the potential game,and the Nash equilibrium is realized by a distributed algorithm,and the Lee Yap Andrianof direct method proves that the Nash equilibrium is globally asymptotically stable,that is,the proposed distributed algorithm can converge from any initial condition to Nash equilibrium.In the real time scheduling of the lower layer,the EV cluster can be selected through the decision tree method,and then a EV ordered charge discharge control method based on the State of Charge(SOC)extension margin is proposed to determine the EV charge and discharge strategy in the cluster.The simulation shows that the proposed two-layer ordered charge discharge control strategy can increase the economic benefit of the charging station while the charging demand of EV is satisfied,and the function of the power·grid load to peak the peak and fill the valley can meet the optimization requirements of many aspects.The hierarchical optimized scheduling strategy can operate in real time and adapt to different number of EV scheduling requirements and has strong scalability.