Charging Station Planning And Distribution Network Reliability Assessment Considering The Space-time Distribution Of Electric Vehicles

Nowadays,environment and energy issues are looming large.The widespread use of traditional vehicles is one of major causes of these problems.Electric vehicle,which is more environmentally friendly,is variously developing to be a desirable alternative.Different from other loads in power grid,the travel and charging and discharging of electric vehicle are featured by space-time distribution.On the one hand,a large number of charging loads with the characteristic of dynamic space-time distribution will change the property of power load.This will lead to increasing difficulty in predicting load,demand for new charging infrastructure and other problems.On the other hand,large-scale electric vehicle batteries can be used as reserve power to support power grid.Therefore,it is quite necessary to study the integration of electric vehicle with space-time distribution into power grid.Based on the analysis of electric vehicle's travel and charging and discharging on road network and power grid,this paper predicts the space-time distribution of a large-scale electric vehicle charging load.On this basis,the optimal location and power capacity of fast charging stations are determined.It also evaluates the enhancement effect of electric vehicle with capacity of vehicle to grid interaction on the reliability of distribution network.The research process presents as follows:Prediction on space-time distribution of electric vehicle charging load based on Markov Decision Process.Firstly,a space-time transfer model of various electric vehicles is established on the basis of Trip Chain Theory.Aiming at achieving shortest travel time and considering randomness of route choice,a process of real-time and dynamic route choice is simulated based on Markov Decision Process.Then,with the measured data of electric vehicle,temperature and transportation energy consumption are modelled,through which energy consumption of per unit distance is calculated.Combined with electric vehicle charging conditions and travel routes,the space-time distribution of electric vehicle charging load is calculated by means of Monte Carlo Method.Lastly,taking certain city as an example,it is proved that the travel simulation method based on Markov Decision Process can fully simulate the randomness of path selection.This method avoids the problem that the electric vehicle of the same origin-destination in the shortest path algorithm is always charged at a fixed place.The predicted space-time distribution of charging load is more consistent with the actual situation.Locating and sizing method of electric vehicle fast charging stations based on quasi-dynamic traffic.Firstly,based on the Voronoi diagram theory,a charging station planning model based on partitioning method is established,and the problems in the process of calculating the charging cost of electric vehicles are analyzed.Then,on the basis of prediction on space-time distribution of electric vehicle charging load and taking into account the impact of traffic jam on electric vehicle travel,a simulation method of space-time distribution of electric vehicle is proposed based on quasi-dynamic traffic.In this way,contradiction between accuracy and computing efficiency of dynamic traffic simulation is resolved.Targeted at minimizing the cost of electric vehicle charging as well as electric vehicle charging infrastructure,a model for locating and sizing of fast charging stations is established which are based on space-time distribution of electric vehicle charging demand.Moreover,by comparing with the charging station planning model based on Voronoi,it proves that the planning method proposed in the paper that takes into account the travel behavior of electric vehicle can accurately calculate the actual charging cost of electric vehicle.The quasi-dynamic traffic simulation method can simulate the influence of real-time path resistance changes on path selection,and the planning result is more accurate.Evaluation on distribution network reliability based on vehicle to grid and quasi-dynamic traffic.Firstly,electric vehicle's travel on road network and charging and discharging on distribution network are simulated based on quasi-dynamic traffic model.In order to calculate electric vehicle's capacity to supply power to faulty island,the way of time advance of traditional Monte Carlo Method is improved and computing efficiency of analogue simulation is promoted.Lastly,according to indexes of distribution network reliability and those of defined electric vehicle reliability,the impact of the number of electric vehicle,battery and resistance of road on both electric vehicle charging reliability and distribution network reliability are evaluated.