Many Objective Site Selection Modeling And Optimization Solution Of Electric Vehicle Charging-Swapping Station

With the introduction of policies related to the development of new energy vehicles,electric vehicles have been vigorously promoted.It alleviates the pressure of air pollution and energy shortage faced by the society.However,the current development system and infrastructure facilities of electric vehicles are not perfect."Difficulty in charging","slow charging" and "mileage anxiety" are still the three major factors to the rapid development of Electric vehicles.At the same time,the site selection of charging-swapping infrastructure has a serious impact on the convenience of users and the cost and income of investors.Therefore,it is particularly important to plan the site selection of charging-swapping infrastructure in a rational manner and to build an efficient charging-swapping service system.Current researches mostly focus on the site selection of charging stations.However,in the swapping battery mode,there is some business interaction between the charging station and the swapping station.Therefore,it is necessary to study the joint location of charging station and swapping station.In view of the characteristics of the site selection of charging stations and swapping stations,this paper proposes a many-objective site selection model for charging and swapping stations based on the consideration of various factors of users and enterprises,and designs a many-objective evolutionary algorithm to solve the model.The details are as follows:(1)Most of the current studies focus on the construction cost and service quality in site selection model.However,the site selection of one or two objectives does not result in better site selection decisions.Due to the fact that the site selection model for charging and swapping stations is of immediate interest to users and businesses,this paper proposes a many-objective site selection model for charging-swapping stations,taking into account construction cost,coverage rate,investment cost and satisfaction.According to the characteristics of the model,a segmented multi-integer coding strategy and corresponding crossover and mutation operations are designed,and several classical many-objective evolutionary algorithms are used to solve the established site selection model in a simulation experimental platform.(2)Based on the many-objective site selection model for charging-swapping stations,a many-objective site selection expectation model for chargingswapping stations with uncertain demand is constructed,taking into account the uncertainty of current charging demand.The model first sets electric vehicle ownership as a random variable and then uses the frequency values of multiple sampling results as probability values through a random simulation method to generate an equal number of random variables for the solution set.Finally,a model of classical many-objective optimization algorithm is used to solve the problem.(3)Many-objective evolutionary algorithm has the problems of uneven distribution and conflicting convergence and diversity when solving the optimization problem model.In many-objective algorithm,the matching selection strategy is the operation of selecting high-quality parents in the process of evolutionary optimization.It will directly affect the quality of the generated offspring.However,the reference vector-guided evolutionary algorithm does not design the corresponding matching selection strategy.To improve the efficiency of solving optimization problems,a reference vector guided evolutionary algorithm based on adaptive matching selection strategy is proposed in this paper.Five algorithms are used for experimental comparison on standard test problems DTLZ and Ma F.Experimental results show the effectiveness of the proposed algorithm.