Research On Charging Facility Locations Considering User Heterogeneity And Charging Station Capacity

New energy vehicle has the advantages of zero emission,energy conservation and emission reduction,which make significant contributions to reducing carbon dioxide emissions and controlling global warming.At present,more and more automobile manufacturers begin to develop and manufacture battery electric vehicles(BEV)and other "green vehicles".In our daily life,more and more consumers choose to buy battery electric vehicle.The traffic network is mixed with fuel vehicles and battery electric vehicles,so it is necessary to study the mixed traffic assignment model with two types of cars.In addition,the density of charging facilities is a very important factor that affects the permeability of battery electric vehicles.Rational decision on the location of charging facilities makes a lot of sense for improving the permeability of electric vehicles and protecting the environment.Firstly,this dissertation summarizes the relevant literature at home and abroad,then constructs a mixed traffic assignment model of fuel vehicles and battery electric vehicles which considering charging on the way,what’s more,the model considers the influence of buffer ranges and the capacity constraints of charging stations.By introducing Lagrange multiplier,the Karush-Kuhn-Tucker(KKT)conditions can be derived,it is concluded that the generalized travel cost of fuel vehicles includes traveling time,the generalized travel cost of battery electric vehicles may also includes queue delay.Based on the mixed traffic assignment model,this dissertation also establishes a bi-level programming model for charging facility locations,including charging stations and charging lanes.The upper-level of the model is that the transportation planning manager makes reasonable decisions on the location of charging facilities to improve the performance of the entire transportation network under a given budget.The lowerlevel of the model is a mixed traffic assignment model that takes into account the capacity constraints of the charging stations.Due to the complexity of solving the bilevel programming model,the genetic algorithm is used in this dissertation.The Gradient projection(GP)algorithm can be applied to solve the traffic assignment model with capacity constraints.Therefore,the GP algorithm is applied to solve the lowerlevel model.In the end,the Nguyen-Dupius road network and Sioux Falls road network is used for numerical experiments.This article compares the impact of different distributions of drivers and different charging station capacities on the total travel cost.In addition,on the location of charging facilities,this paper compares the location schemes under different budget levels,and verifies that the genetic algorithm and GP algorithm can effectively solve the bi-level programming model with capacity constraints.The results show that the algorithm converges fast and can get good results in a short time.