Research On Optimization Of Charging/discharging Efficiency Of Electric Vehicles In V2G

In recent years,countries around the world have attached increasing importance to environmental protection and sustainable development issues.This has made electric vehicles,which replace traditional fossil fuels with efficient and clean electric energy,stand out due to their environmental protection,energy saving,and convenience features,gradually occupying a share of the automotive market.However,the unregulated use of a large number of electric vehicles without scheduling will lead to traffic congestion,and their disordered charging behavior can have a negative impact on the power grid load.The use of Vehicle-to-Grid(V2G)technology,which enables bidirectional energy flow between electric vehicles and the power grid,is an effective solution.In V2G,a large number of coordinated electric vehicles carry high-capacity batteries and participate in power grid dispatching as energy carriers,thus forming a large-scale distributed energy storage system,so as to achieve the three-in-one goal of reducing the peak and valley of the power grid,improving the utilization rate of renewable energy and bringing additional benefits to electric vehicle owners.This paper focuses on the charging and discharging problem of electric vehicles based on V2 G technology,and proposes a strategy that jointly optimizes the routing planning and charging/discharging decision-making of electric vehicles.This strategy provides a solution that not only satisfies the travel requirements of electric vehicles as transportation tools,but also enables them to participate in grid dispatching as mobile energy storage devices.Taking into account various practical constraints,this paper first proposes a distributed multi-vehicle joint path planning and charging/discharging decision algorithm based on game theory.Secondly,a multi-vehicle joint routing and charging/discharging decision algorithm based on reinforcement learning is proposed to solve the charging/discharging problem of electric vehicles under time constraints.The main contributions of this article are as follows:(1)Taking electric vehicles as the dispatching object,considering the maximum battery capacity,vehicle departure destination,limited charging piles of charging stations and road traffic and other practical constraints,integrating the charging,discharging,waiting and passing behaviors of electric vehicles,and considering the influence of multiple electric vehicles,the problem of maximizing the benefits of vehicle alliance is described as a dynamic cooperative game with complete information.A distributed multi-vehicle Joint Path Planning and Charging/Discharging Decision algorithm(JPPCDD)based on game theory is proposed.Finally,the advantages of JPPCDD are verified by comparative simulation experiments on open data set(PeMS).(2)In order to further reduce the impact of electric vehicle participation in energy dispatching on the travel of vehicle owners,a maximum travel time limit for vehicles is added based on(1).Firstly,Markov decision process(MDP)was established to describe the problem,and route selection and charging/discharging behavior of the vehicle were innovatively integrated in the vehicle action space.Secondly,based on the general multi-agent reinforcement learning algorithm,a Multi-vehicle Joint Routing and Charging/Discharging Decision algorithm(MJRCDD)was proposed to solve the V2 G based charging/discharging benefit optimization problem of multi-electric vehicles under time constraints.Finally,the validity of MJRCDD is verified by simulation experiment and comparison experiment based on PeMS.