Charging Station Location Problem Of Electric Vehicles

With the continuous development and progress of our social economy,environmental problems are getting more and more attention.Urban motor vehicle emissions are the main source of air pollution in our country.Electric vehicles,as a kind of new energy products,rapidly rise in the market with its advantages of low pollution,zero emission(low emission),high energy efficiency,etc.However,as the important supporting infrastructure of electric vehicles,the imperfect or unreasonable construction of charging stations severely hindered the development of electric vehicles.At present,it is difficult to charge for electric vehicles in many areas.Therefore,it is necessary to make reasonable planning and management for charging stations.Previous researches on charging station location problem are mainly divided into two categories.The first kind of research is mostly based on the traditional location model,including covering model and flow-capturing model,etc.It belongs to the research category of combinatorial optimization(or discrete optimization).In the second kind of research,the researchers mainly consider the location of charging stations in different situations,including the parking behavior,traffic corridor,network equilibrium,etc.It belongs to the research category of continuous optimization.The research of this paper mainly falls into the first kind of research category.The thought and method of traditional location model are used to formulate discrete models to study the charging station location problem.At the same time,according to the need of specific problem,a corresponding continuous model considering the traveler’s choice behavior is also proposed.Unlike the previous studies,fully considering the characteristics of electric vehicles and charging stations,and traveler’s choice behavior of charging stations,etc.,this paper proposed several kinds of electric vehicle charging station location models.The research work of this paper mainly includes the following aspects:First,compared to other travelers,commuters have greater constancy in time and space.Their commutes are ususlly concentrated in certain short periods and certain areas.It has obvious regularity.Therefore,considering the travel behavior of the commuters,this paper takes the minimization of the weighted sum of the station’s construction cost and the travelers’ access cost to charging stations as the objective,introduces the parameter that means the maximum distance users can accept between the charging station and their destination,and proposes the charging station location model.The model simultaneously handles the problem where to construct the charging stations and how many chargers should be installed in each station.Considering that the charging station is not necessarily constructed on the nodes of the urban traffic network,it may also be constructed on the links,this paper further expands the model,and proposes the genetic algorithm to solve the model.At last,the numerical experiments are carried out based on the application scenarios of a certain area in Beijing.Numerical results show that the location result obtained by the expanded model is more reasonable than that obtained by the basic model.Second,based on the study in the first part,and considering that different users may have different perception of station access cost and charging cost due to personal interest,different income level as well as the nature of work etc.,this paper analyzes the charging selection behavior of the commuters,classifies the travelers according to the nature of their destinations and formulates a charging station location model considering different types of users.The model takes the minimization of the total cost,including charging station’s cost(such as installing cost and management cost,etc.)and the user’s cost(such as charging cost,station access cost,and waiting cost,etc.),as the objective.The numerical experiments are carried out based on the application scenarios of a certain area in Beijing.Third,different levels of electric vehicles charging stations vary in size,construction cost and service capacity.Therefore,based on the study in the previous parts and considering the different charging demand,this paper first proposes a location model of multi-level charging stations to minimize the weighted sum of the station’s construction cost and the users’ cost using the thought of classical location models.Then,in order to better reflect the planners’ decision-making behavior of charging station location problem and travelers’ choice behavior of charging stations,the bi-level programming model was established to further study the location problem of the multi-level charging stations and the genetic algorithm is proposed to solve the bi-level programming model.Finally,the numerical experiments based on the application scenarios of a certain area in Beijing are carried out.The two models can not only determine the location of the charging stations,but also obtain the service scope and corresponding level of each charging station.Under the conditions of meeting the requirements of the objective functions,the models ensure the effective utilization of charging stations.Fourth,further,considering that for travelers,the completion of a trip is influenced by various factors,such as the vehicle range,the charging efficiency,the actual available charging time,etc.,and it may need to charge more than one time to complete one trip.In order to meet the various charging demand in real life and ensure that the travelers can successfully complete the travel,this paper proposes a location model of multiple types of charging stations considering the round-trip behavior of travelers.The model minimizes the construction cost and maximizes the number of the captured users.In the model,the parameter of the minimum safe reserve ratio are introduced to reflect the range anxiety of traveler.The higher the value is,the higher the range anxiety will be.