Research On The Optimal Charging Strategy Of Electric Vehicle Charging Station With Photovoltaic Power

With the growth of the number of electric vehicles,the charging load of electric vehicles will have a certain impact on the power grid.The electric vehicle charging station with photovoltaic power supply can not only improve this problem,but also help to save energy and reduce emissions.However,the charging power of electric vehicles in the photovoltaic charging station and the generating power of photovoltaic power supply have strong randomness and volatility,and the electric vehicles in the station are disordered charging will cause the waste of photovoltaic power generation and will have a certain impact on the distribution network.In order to realize the on-site consumption of photovoltaic power generation in the photovoltaic charging station and reduce the impact of the electric vehicle’s disorderly charging load on the grid,it is necessary to predict the photovoltaic power generation in the station one day in advance,and guide users to charge orderly according to the predicted results.This paper takes the photovoltaic charging station to be built in a university in the north of Henan Province as the research object,and puts forward the optimal charging strategy of the electric vehicle charging station with photovoltaic power supply.First of all,two combined forecasting models of photovoltaic power generation are proposed.Support vector machine(SVM)prediction model with linear kernel function,polynomial kernel function and gaussian radial basis kernel function as kernel function is constructed as a single model in the combination prediction model,and then BP neural network and deep belief network are used to optimize the prediction results of the three single models respectively,and carried out simulation verification,the simulation results show that the mean absolute percentage error of the combined forecasting model of photovoltaic power generation based on support vector machine and deep belief network is up to 15.50%,which is more accurate than other forecasting models.Therefore,this model is selected to predict the photovoltaic power generation in the photovoltaic charging station.Secondly,according to the prediction results of photovoltaic power generation,an orderly charging optimization strategy is proposed.Taking the charging price difference of each period and the peak-valley-flat property of each charging period as independent variables,and the charging power of electric vehicles as the dependent variable,an ordered charging power model is established.The goal is that the orderly charging power can effectively track the predicted power of photovoltaic power generation,and with the constraints of charging price,distribution capacity and charging price difference,the genetic algorithm is used to solve the optimal charging price difference and time attribute of each time period of the next day,and then the optimal charging price of each time period of the next day is calculated,and carried out simulation verification,the simulation results show that the strategy can achieve 99%absorption rate of photovoltaic power generation in the photovoltaic charging station.Finally,in order to apply the method in practice,the information management platform of photovoltaic charging station is designed,and the relevant methods proposed in this paper are embedded in the platform,and the architecture of the platform and the function of the software system are designed.Through the charging optimization strategy of this paper,it provides a reference for the future operation mode of the photovoltaic charging station to be built and the same type of photovoltaic charging station in a university in the north of Henan Province.