Charging And Routing Scheduling Method Of Electric Vehicle Based On Internet Of Things

With the development of internet of things,the increasing integration of electric vehicles brings a great challenge to power grid and transportation network.We can perceive the battery level of electric vehicles and operation status of charging facilities by placing the sensors in electric vehicles and charging stations.It is a great challenge to schedule the electric vehicles to support some certain services in power grid and transportation network while maintaining the stability and capacity of power grid and transportation network.This paper aims to study the task scheduling and energy management of power grid and transportation network.We do the research on the task scheduling and energy management of electric vehicles and their influence about the community loads.We study the impact of EV charging scheduling on community load under multiple constraints,the joint optimization of EV routing and charging scheduling.We make up for the shortcomings of the existing work of EV charging and routing scheduling in the traffic network and the power grid.The main work of this paper is as follows,(1)First,the energy management and task scheduling problems of delay-tolerant tasks in communities are studied.In this paper,we designed a user dissatisfaction function for the delay-tolerant task.Based on Lyapunov optimization method,the delay-tolerant task scheduling algorithm is proposed to minimize the average cost per unit time of the community and solve the problem of the community’s energy and heat demand.Then,a renewable energy sharing strategy is proposed to schedule idle energy by reinforcement learning algorithm,which keeps the user dissatisfaction within a certain range and minimizes the average cost per unit time of the community.The advantages of the renewable energy sharing strategy are verified by simulation.(2)Second,we first study the charging scheduling problem of electric vehicles with deterministic environmental information in the community,then study the online charging scheduling problem of electric vehicles considering random arrival and the uncertainty of demand constraint to minimize the charging cost of electric vehicles and reduce the influence of electric vehicle charging on the total community load.Different from the traditional charging problem where the EV charging rate is a discrete value,the charging behavior of EV is modeled as a Markov decision process.Then asynchronous actor-critic algorithm is used to solve the online continuous charging problem of EV.Furthermore,an improved asynchronous actor-critic algorithm is proposed to reduce the dimension of state space,and the efficiency and performance of the algorithm are verified by simulation.(3)Third,based on the research about online charging scheduling of electric vehicles in the community,a charging and discharging model of electric vehicles was designed with physical constraints such as customers’ waiting time and battery degradation.Based on the Lyapunov method,a suboptimal charging algorithm with the limited customers’ waiting time is proposed to study the cost optimization problem.The suboptimal charging algorithm can provide the selection criteria of EV charging action.Based on the prior experience,by modeling the optimization problem as Markov process,a charging scheduling algorithm for electric vehicles based on reinforcement learning was proposed to obtain the global optimal charging scheduling scheme.The simulation results showed that the algorithm has a good performance and good computational efficiency.(4)Then electric vehicles play an important role in the transportation network.To study the charging scheduling problem of electric vehicles in the power grid,it is necessary to consider the actual routing constraints in the transportation network.Based on the existing offline routing and charging scheduling problem of electric vehicles,this paper puts forward a new charging joint routing and charging scheduling framework and its mathematical model of autonomous electric vehicles.Then we designed a customized joint routing and charging scheduling optimization algorithm based on benders decomposition method,to solve the mixed linear programming problem about integer routing variables and continuous charging variables,and similar trajectory measurement method is used to improve the efficiency of the algorithm.(5)Last,we studied the online joint routing and charging scheduling problems of electric vehicles.A mathematical model of joint routing scheduling of electric vehicles was proposed based on clustering method according to the taxi-hailing mode in the traffic network service,which considers the pickup and delivery problem,to minimize the passengers’ waiting time.To keep the total power of the community within a certain limit,a two-dimensional backpack charging problem of autonomous electric vehicles is proposed.Our simulation results show the feasibility of the model and algorithm.