| As the number of electric vehicles increases,their travel behavior not only affects the operation of the traffic network through traffic parameters such as traffic speed and traffic flow,but also affects the operation of the power grid through the real-time load of charging stations.Therefore,the charging scheduling of EVs based on the information of power grid and traffic network is helpful to improve the operation status of power grid and traffic network and promote the use of EVs.This article addresses the issue of difficulty in balancing charging path optimization and charging time and space optimization in current research on electric vehicle charging scheduling.Starting from the difference in charging demand,a charging scheduling strategy of "fast charging guidance and slow charging control" is proposed based on the differentiated charging demand.The main research contents of this paper are as follows:(1)The theoretical basis related to EV charging scheduling such as charging load influencing factors,charging path optimization methods and charging scheduling strategy evaluation indexes are elaborated.Firstly,the influencing factors of EV charging load are studied and the charging load model is constructed based on Monte Carlo method.Secondly,the mainstream algorithms of path optimization are explained and simulated for the charging path optimization problem faced by the on-road energy replenishment of electric vehicles.Finally,the evaluation indexes of grid side,transportation network side and electric vehicle side are clarified to provide the theoretical basis for quantitative evaluation of subsequent charging scheduling strategies.(2)A charging path optimization strategy that integrates traffic network information and grid information is developed to address the on-route energy replenishment needs of EVs to their destinations,overcoming the problem that the existing path optimization strategy does not consider multiple sources of information comprehensively and thus has poor optimization effects.The strategy first constructs a "traffic network-grid-electric vehicle" system model,based on which the Floyd algorithm and Monte Carlo method are used to predict the spatial and temporal distribution of charging loads at fast charging stations,and then develops a spatial and temporal dual-scale incentive tariff based on the predicted loads and the power-price model.Finally,the traffic network information and grid information are assigned to the actual road sections of the traffic network in turn,and the multi-objective decision model for charging station selection based on regret theory is established by considering the driving distance,driving time,charging queue waiting time,charging cost and other information.The simulation results of the actual road network show that the proposed strategy optimizes the spatial and temporal distribution of load and improves the travel experience of users.(3)Aiming at the differentiated charging demand of EVs when they arrive at the destination for energy replenishment,the differentiated charging demand is quantified as the charging demand urgency index,and the EV charging time scheduling strategy with user charging cost as the optimization target under different charging demands is studied,which overcomes the shortcomings of the existing charging time scheduling strategy that ignores the path navigation on the traffic network side and the differentiated charging demand on the grid side before users are connected to the grid.The strategy first recommends the optimal driving path for users before grid connection based on the path optimization strategy and real-time traffic network information,and constructs an electric vehicle model for reaching the destination to replenish energy by combining the travel chain and charging demand urgency.Then,it takes EV charging satisfaction as the optimization object and minimizes user charging cost as the optimization objective to schedule the charging load of distribution network on the time scale,while taking into account the peak and valley reduction of distribution network load.Finally,Latin hypercube sampling is used to conduct simulation experiments under two EV charging scenarios,three EV cluster sizes and three types of charging strategies.The results show that the proposed strategy significantly reduces the charging cost of users while achieving peak and valley reduction in the distribution network.(4)For the flexible charging demand of EVs when they arrive at their destinations for energy replenishment,a second-order cone planning-based EV charging time and space scheduling strategy is studied with the optimization objective of minimizing distribution network losses,which ensures the safe and economic operation of the distribution network,and reduces the charging cost of users.Based on solving the emergency charging demand of EVs,the strategy treats EVs with flexible charging demand as mobile energy-carrying devices of the distribution network,and then adopts the second-order cone planning theory to schedule the charging behavior of EVs on spatio-temporal scale with the distribution network as the optimization object.The results show that the total network loss is reduced by 17.561%and the node voltage minimum is increased by 6.955% under the proposed strategy,which further makes up for the shortcomings that the scheduling effect of single area charging time scheduling strategy in the office area charging scenario is yet to be optimized and the potential of flexible charging load scheduling is not fully exploited. |
