Research On Electric Vehicle Charging Station Planning And Operation Considering Transportation Network With Photovoltaic And Battery Energy Storage System

With the rapid development of the world economy,fossil resources such as oil and coal are increasingly exhausted,the global climate is warming,and the human living environment is continuously deteriorating.Exhaust emissions from fuel vehicles are one of the important sources of air pollution.As a new energy vehicle,electric vehicles(EVs)have attracted more and more attention due to their advantages of less pollution,low emissions,and high energy efficiency.However,as the number of electric vehicles continues to grow,the demand for charging infrastructure is also increasing.How to improve the charging infrastructure has become a bottleneck restricting the development of electric vehicles.In other words,reasonably planning the location and capacity of charging stations is important for development of the electric vehicle industry and the safe and stable operation of the power system.This thesis takes public fast charging stations for EVs and fast charging stations for bus system as the research objects.For the fast public charging stations,the optimal location is determined on the basis of fully considering the traffic flow of the transportation network and the load characteristics of the distribution system.In the meantime,considering photovoltaic(PV)power generation and battery energy storage systems(BESS),the economic operation of charging stations is also discussed.The gravity model is adopted to determine the traffic flow when studying the charging load,and the interconnection between traffic nodes is fully considered.In order to minimize the number of charging stations and then reduce the total investment accordingly,the research on fast charging stations for bus system is conducted by clustering the adjacent bus stations.For the operation cost,not only the charging cost of electric buses(EBs)and the energy loss are included,but also the time cost of buses heading to the charging stations in taken into accout.Different from previous studies,this thesis only focuses on the development of electric vehicles in small and medium-sized cities.The followings are the main research work of this thesis:(1)In human society,apart from the existence of universal gravitation in the physical world,similar gravitational phenomena also widely exist in the flow of people,objects and information among different places.According to the traffic data obtained,this thesis uses the gravity model to calculate the traffic flow,considering the distance between each intersection and the center of the main road,as well as the distance between each intersection and the center of the city.It then determines the charging probability of electric vehicles by studying the probability distribution of electric vehicle mileage and the probability distribution of mileage per period,as well as the relationship between state of charge(SOC)and mileage;finally based on the traffic flow of the transportation network,the penetration rate of electric vehicles,etc.Factors that determine the charging demand of electric vehicles will be considered.(2)A queuing model is established for charging stations according to the temporal and spatial distribution of charging demand,and the minimum number of chargers installed in each charging station is determined with the model.Then,with the minimum total social cost as the objective function,the optimal planning model of the charging stations is achieved.In order to avoid the optimization solution falling into the local optimum,the problem is solved by the simulated annealing particle swarm algorithm.When calculating the total social cost,in addition to consider the commonly used investment cost of the investors,the cost of power consumption from the user to the charging station,the cost of waiting for charging,and the time cost of the user on the road are also included.The Voronoi diagram is used to divide the service scope of the charging station to realize the final decision for site selection.Finally,the optimization calculation and the comparison of the schemes are carried out for the case study,and the planning scheme obtained by the simulated annealing particle swarm algorithm is better than the classical particle swarm algorithm.(3)The electrification of public transportation system is an important part of urban traffic electrification,and in general the electrification of public transportation is the first to achieve for urban traffic electrification.On the basis of comprehensive consideration of the bus operation network and power distribution network,research on the fast charging station planning of the bus system is investigated.The affinity propagation clustering algorithm is used to cluster the bus stations to reduce the search space of the particle swarm algorithm.Then the discrete binary particle swarm optimization algorithm is adopted to find the optimal solution for the deployment of fast charging stations.The goal of optimization is to reduce the redundancy of equipment in the charging station and the higher cost caused by excessive charging stations.Finally,by taking into account the demand of the electrification of the public transportation system in the planning city and promoting decarbonization in the area,the bus charging station planning is optimized for the current situation and different departure intervals,and it is found that as the departure interval decreases,the vehicle scale and charging demand will increase,thereby the total investment of charging stations is increasing.(4)In order to effectively achieve carbon neutrality,solar cell modules are laid on the shed of the charging station,and energy storage system is added to reduce the uncertainty of photovoltaic power generation and improve the economic benefits of the operation of the charging station.More accurate prediction of PV power generation can better schedule the coordinated operation between PV&BESS.Single model forecasting has limitations.This thesis proposes a combined forecasting method that considers different weather conditions.K-mediods clustering algorithm is used to analyze the influence of radiation intensity,temperature,humidity and their changing differences on photovoltaic output power under different weather scenarios.Through the analysis of historical data of photovoltaic power generation,it is found that the effect of simply distinguishing by weather condition is not ideal.This thesis combines seasonal effects and weather conditions for analysis,which can improve the accuracy of cluster analysis effectively.Then,according to the results of weather clustering,a combined forecasting model of photovoltaic power generation based on Bi-directional long short term memory(Bi-LSTM)network for different weather conditions is established.Finally,an example is used to compared with other methods to verify the effectiveness of the proposed model.(5)The coordinated charging and discharging of PV&BESS can realize the operation optimization of the fast charging station and the economic operation of the distribution network.Considering the strategy of discharging and charging at peak and valley tariff respectively,a charging station operation model was proposed to reduce the power purchase cost of operators and the peak-valley load difference of distribution lines.In the calculation of the power purchase cost of the operator,the peak and valley tariff period implemented by the power grid company at the present stage is fully considered.The power purchase cost is analyzed step-by-step,and a refined model for optimizing the operation of the electric vehicle fast charging station with PV&BESS is established.Through comparison,it is found that the two objectives mentioned above are well optimized embedded with the PV&BESS.