| The popularization of electric vehicles(EVs)is of great significance for reducing carbon emissions,realizing the transformation of energy structure,solving the shortage of resource and environment problems,and accelerating sustainable development.Therefore,based on the vehicle and traffic data in a certain region,the planning location of battery-swapping stations(BSSs)and centralized charging stations(CCSs)under battery-swapping mode(BSM)is studied.The hosting capability of EVs is analyzed simultaneously considering centralized slow charging,decentralized slow charging,and centralized fast charging.The main contribution of this thesis is as follows:A service radius method(SRM)for quantitative analysis of BSS service demand is proposed combing with the scale of multi-type EVs predicted by system dynamics(SD)model.Then,the spatial-temporal distribution of battery-swapping EV(BSEV)service demand relying on the location of BSSs is obtained combining with the spatial-temporal distribution data of regional traffic flow of vehicles.Besides,an improved differential evolution algorithm combing with Monte Carlo Searching(IDEA-MCS)method is proposed to optimize the location of BSSs respectively in pessimistic scenario,normal scenario,and optimistic scenario.Through the cross-comparison and analysis among different scenarios,the maximum average conversion profit is taken as the optimal results of BSSs.Compared with DEA method and IDEA method,the advantages of IDEA-MCS optimization algorithm are verified.To settle the planning problem of CCSs in the BSM,the charging and battery distribution strategies of CCS are proposed.The start distribution time,daily charging quantity,charging period,and the range of charging time are determined.After verifying the rationality of charging and battery distribution strategies,the allowed range of charging number of batteries is determined.Aiming at the minimum annual total conversion cost,the charging number of batteries and daily charging period in CCS are optimized by the method of non-linear mixed integer programming with different scales of EVs.Considering the impacts on distribution network caused by multiple charging modes of EVs,a preliminary study on the hosting capability of EVs in distribution network is carried out.Based on the analysis of charging process and charging demand of EVs,the charging load characteristics under different charging modes are determined.The maximum daily hosting capability,instantaneous penetration rate and daily average penetration rate of EVs are analyzed and calculated by improved differential evolution algorithm(IDEA)in distribution network. |
