Research On Pre-evaluation Method Of Charging Station Connected To Distribution Network

In the context of global environmental pollution and energy crisis,new energy vehicles represented by electric vehicles(EVs)have received worldwide attention from all walks of life.Among them,the widespread deployment of charging network is considered to be the only way to promote the development of EVs.However,because of the lack of electric power resources and other reasons,it is difficult to access charging stations in many parts of China,and there is a shortage of systematic method to reasonably evaluate the influence of charging station access.In view of this,this paper proposes a complete set of pre-evaluation method for the impact of charging station access,aiming at the pre-assessment need of charging station to be connected to the power grid.Through the probability characterization of base load and the prediction of charging station load to be accessed,the quantitative assessment of line heavy-overload risk and economy after charging station access is realized.Based on the assessment results,a comprehensive decision scoring method is proposed.It provides a useful reference for pre-evaluation of charging station access in power shortage area.Firstly,the probability distribution characterization method of typical base load of distribution network lines is studied.Depending on the variability of the load distribution,the base load is grouped into three categories: holiday load,Sunday load and ordinary load.Then,based on the data distribution characteristics of the three types of load,the probability distribution characterization method the three types of load is determined,and the probability model of each type of load is established respectively.Secondly,the prediction method of load of charging station which would be connected is designed.Then the K-means clustering algorithm is used to cluster the historical data of charging stations.On the basis of evaluating indexes of clustering algorithms,the actual scene information of charging stations behind the data is introduced to evaluate the physical rationality of the evaluation clustering results.Finally,11 typical load patterns are extracted,which can be used to predict the load of charging stations to be connected.Thirdly,the pre-evaluation method of charging connected to the distribution network is proposed.A few evaluation indexes are designed from two aspects: the risk of heavy-overload after the charging station is connected,and a comprehensive scoring method is proposed by combining the AHP(analytic hierarchy process),EWM(entropy weight method)and fuzzy theory.Finally,based on the above research results,charging station access schemes in each scenario are designed and the evaluation method proposed is used to evaluate and score each access scheme.Then the rationality analysis of the evaluation results is carried out to verify the effectiveness of the pre-evaluation method proposed in this paper,and provide relevant reference for the actual application scenario of the power grid side.