Research On Location Of Electric Vehicle Charging Station Considering Charging Interruption And Uncertain Demand

With zero pollution and low noise,electric vehicles can reduce dependence on oil and reduce carbon emissions,and as "carbon peaking and carbon neutral" has become an important strategic goal for China’s future development in recent years,the electric vehicle industry is developing rapidly.However,the development of the electric vehicle industry also faces a number of challenges,with the imbalance between electric vehicle charging facilities and demand remaining one of the key factors limiting the development of the electric vehicle industry.The mismatch between the location of charging infrastructure and the location of charging demand can prevent some EV users from having their needs met in a timely manner,causing travel anxiety for these users and becoming a constraint to the development of the EV industry.Therefore,charging stations are a key part of the configuration of new energy electric vehicles,and the scientific location layout has important theoretical and practical significance for the development of the electric vehicle industry.First of all,the design planning basis of electric vehicle charging stations is analysed.From the conflicting perspectives of operators and users,a robust optimisation model for EV charging station locations is constructed,taking into account the two realities of facility disruption and demand uncertainty.The model takes into account the operator’s construction costs and operation and maintenance costs,and the user’s position,taking into account the user’s convenience and aiming to reduce the user’s charging costs;and introduces the concept of virtual standby emergency facilities to meet the charging demand at each demand point.This will result in penalty costs if demand point users have to be served by emergency facilities.As the model contains uncertain demand,it is transformed into an equivalent robust optimisation model for solution following a robust optimisation approach.Next,a performance efficient improved mayfly algorithm was designed.Using the basic mayfly algorithm as a framework,a modified Tent chaotic sequence is used to initialise the mayfly population,and an anti-attraction velocity update mechanism is introduced to guide the mayfly velocity update according to the characteristics of the MA algorithm,as well as a dimension-by-dimension centre of gravity reverse learning variation of the globally optimal mayfly to help the algorithm jump out of the local optimum and accelerate convergence.Comparative simulation experiments based on 12 standard test functions and some CEC2017 test functions show that MMOA has significant advantages over other comparative algorithms in terms of convergence speed,optimisation finding accuracy and stability.Finally,the validity of the model and the superiority of the improved Mayfly algorithm are verified in conjunction with the analysis of the algorithms.The improved mayfly algorithm was used to solve the charging station siting model for a specific area,and the coordinates of each charging station and the distribution of demand for each facility point were obtained.By comparing the algorithm with four other algorithms,it was found that the improved mayfly algorithm had the best solution and the lowest system cost.A sensitivity analysis of several parameters of the siting model was then carried out to investigate the impact of parameters such as the proportion of demand disturbance,the probability of facility disruption,and the unit penalty cost on the total system cost.Different decision makers can control the robustness of the model based on risk preferences to avoid over-optimism or over-conservatism.