The Scheduling Strategy Of Electric Vehicles Charging And Discharging Based On Vehicle-To-grid Mode In The Smart Grid

With the increasingly severe energy crisis and environmental pollution problems,electric vehicles(EVs)have been strongly supported in all over the world for its good environmental performance and the advantages of energy for power.As the scale of electric vehicle application increases rapidly,EVs disorderly charging has brought the serious negative impact of power system security and economic operation.Facing the future development of energy bidirectional exchange between EVs and smart grid,optimizing the charging and discharging behavior of EVs has become a hot research topic.This theis analyzes the research status of smart grid and charging control strategies for EVs,and discusses the influence of EVs disorderly charging on power grid.Vehicle-to-Grid(V2G)technology and implementation methods are emphatically studied.According to centralized and distributed controlable modes,the charging power distribution method,the centralized V2 G scheduling strategy and distributed V2 G scheduling strategy for EVs are proposed in this thesis.Firstly,it is of great significance to study how to make use of the available power to allocate equitably for EVs.A dynamic weighted power allocation method for EVs under distribution network is proposed.Considering distribution network constraints,battery constraints and time constraints,establishing evaluation index of charging fairness includes: the charging time,residence time,battery statu and each vehicle historical charging integrity.Entropy weight method is used to obtain the charging weight of each vehicle.And with weighing Max-Min fair allocation algorithm,the limited power resources are distributed to EVs fairly.This method dynamically controls the charging power of electric vehicle according to the load of power grid,avoids overload of distribution transformer,and improves customer satisfaction effectively.Secondly,in order to realize centralized V2 G,an interactive dispatching strategy between EVs and grid based on clustering optimization is proposed.Firstly,according to the battery,time and charge-discharge conversion constraints,the charging vehicles are divided into regular and control group in different periods by charging pile.The former can charge disorderly,and the latter including charging vehicle group and discharging vehicle group.Then,the division information and load information of the car group are counted in the charge pile controller.So the scheduling load of control group can be optimized to minimize the total load variance during the control period.The weight coefficient of each vehicle is generated by the travel constraint of each vehicle.The dynamic weighted power allocation algorithm is used to control the charging and discharging power which is not higher than the dispatchable load value.This method can meet the travel demand of EVs,suppress the fluctuation of power grid,and reduce the system operation and user charging costs.It is simple and beneficial to practical application.Finally,aiming at the problems of high complexity,high cost and high dependence on communication system of centralized control,a distributed orderly charging and discharging management strategy for EVs is put forward.It is implemented by charging piles independently without the real-time monitoring and controlling of centralized communication network.The time distribution of the next day's electric vehicle's charging load is optimized through the predicted load information.Then decision strategy maker generates the probability table of charging and discharging margin in each period of the next day and sends it to charging piles.When daytime online control is begin,the charging pile generates the charging and feeding plan independently according to the battery status,user demand information and probability distribution table of the electric vehicle,and displays it to the user.The user decides whether to respond to the strategy.The results indicate that this method can guide EVs to realize peak load shifting and take account of the interests for both the grid and users.It is suitable for decentralized management of large-scale EVs.