Architecture And Algorithm Of Energy Trading System For Urban PHEV

With the continuous development of auto industry and Internet of Vehicles,electric vehicles,as mobile storage battery,have brought the new opportunity and challenge to V2X(Vehicle-to-X,X can be Vehicle,road,person,etc.)technology development.In view of the special environment of urban Internet of vehicles,energy scheduling of plug in hybrid electric vehicle(PHEV)has the following two problems.First,the social hotspots in urban areas are far away from the main power grid,and the development of vehicle-to-grid(V2G)has the problems of small coverage,large loss,high time cost and mismatch between supply and demand.Second,the centralized V2 G model has potential risks of single point attack and monopoly compromise.PHEV refuses to contribute surplus energy due to security threats and lack of interest incentive.In view of these problems,some scholars proposed to add parked vehicles as distributed mobile batteries into the energy system of the Internet of vehicles,based on the statistical characteristics that most vehicles are parked and idle at 95% of their time and the number of parked vehicles in hotspots is very large.However,there is still very little research in this area,and most of it is focused on microgrids supported by remote cloud computing.Therefore,this thesis deeply analyzes the above problems and provides effective solutions.The main contributions and innovations are explained as follows.(1)Fog computing is located between cloud center and Internet of things devices,and is close to the end user to process the processing and storage tasks.For real time perception and wide geographic distribution,this thesis proposed a new Internet of vehicles architecture based on fog computing to support the local energy transaction of PHEV and save time cost and energy loss of bypass charging under the V2 G scenario.The new architecture introduces a new entity fog computing energy center.Thus,different utility maximization models and solving algorithms are designed for the following two situations: 1)nonprofit-driven fog computing energy center,whose goal is to make PHEV benefit from local charging and discharging operations;2)profit-driven fog computing energy center,whose goal is to maximize its own profits while ensuring that each PHEV gets non-negative transaction utility.Simulation results showed that the model is fair and energy-saving,the proposed algorithm is low complexity,and the improved algorithm is superior to the original algorithm in terms of convergence speed,final value and the uniformity of Pareto solution set.(2)Aiming at the problem of user privacy protection and transaction record tamperproof,this thesis proposed to reduce the system's dependence on trusted third parties by using the alliance block chain under the Internet of vehicles architecture based on fog computing.In addition,a more efficient and reliable consensus algorithm DPOSP was formed by improving the practical byzantine fault tolerance(PBFT)algorithm and incorporating it into the delegated proof of stake(DPOS)algorithm.In order to encourage PHEV to participate in local energy trading,an energy iteration double auction(EIDA)algorithm was designed to solve the social welfare maximization problem model,so as to obtain the optimal charging and discharging decision and energy pricing.Finally,genetic algorithm and Lagrange algorithm were used to realize EIDA.Simulation results showed that the proposed consensus algorithm has good energy saving,fault tolerance and bad node processing ability,the results of model solution algorithm was in line with the law of the market,and the value of optimization target,cost performance of genetic algorithm were better than these of Lagrange multiplier algorithm.