Research On Electric Vehicle Charging Scheduling And Optimization Strategy For Parks

In response to the dual pressures of energy and the environment,energy vehicles have become an effective way to alleviate the energy crisis and solve the "dual carbon" problem due to their low pollution and high energy efficiency.However,when a large number of disorderly charging electric vehicles are connected to the grid Due to the strong uncertainty,it will have a huge impact on the operation of the power grid.Based on the electric vehicle charging event data set ACN-Data,this paper analyzes the load characteristics of electric vehicles,builds an electric vehicle charging load model and management framework,and applies an adaptive charging network simulation platform,which uses real physical charging.Based on the equipment,it can provide sufficient charging potential for the power infrastructure.Taking this platform as the background,the research on the optimization problem of electric vehicle charging scheduling and the corresponding optimization strategy is carried out.The main contents include:（1）Aiming at the seasonal,nonlinear,and random fluctuations of electric vehicle charging load,a new model based on Seasonal Autoregressive Integrated Moving Average（SARIMA）and Long Short Term Memory（LSTM）hybrid model for electric vehicle charging load prediction method.This method combines the characteristics of the SARIMA,which is good at processing linear data,and the LSTM,which is good at processing nonlinearity and randomness.The prediction accuracy of the load is at least 23.68% higher than that of general machine learning models under MAPE index.It can accurately analyze and predict the load of electric vehicle,it provides a basis for the subsequent optimization of electric vehicle charging scheduling.（2）Aiming at the disordered charging behavior of electric vehicles at charging piles in the park,an optimization strategy for electric vehicle charging scheduling based on convex optimization algorithm is proposed.The time-of-use electricity price is used to increase the enthusiasm of users to participate in orderly charging,and dynamic peak periods and peak price increases are introduced into the user’s charging electricity price,and a dynamic electricity price implementation plan is formulated.A charging optimization model with the lowest user charging cost,the lowest charging peak value,the smallest load fluctuation,and the alleviation of battery degradation as the objective functions is established,and the model is solved by the second-order cone programming algorithm based on the interior point method.Compared with disordered charging behavior,at least 34.90% of the pressure on the grid during peak charging of electric vehicles is reduced,the load fluctuation of the grid during charging of electric vehicles is alleviated,and the degradation problem of batteries in electric vehicles is alleviated to a certain extent,and the cost is reduced.（3）Aiming at the influence of the output uncertainty of distributed energy and electric vehicle charging load on the system,a Lagrangian relaxation method for orderly charging of electric vehicles and scheduling optimization of coordinated control with distributed energy are proposed.This method can stabilize the output fluctuation of distributed energy resources and reduce user costs at the same time.The concept of set pair analysis is introduced to model the uncertainty of renewable energy,and the electric vehicle is used as the response load to establish the charging load response model of the electric vehicle.The model is solved using the Langrangian relaxation method.In the experimental verification,taking into account the seasonal difference in distributed energy output,the optimization results in winter and summer scenarios are discussed.On the basis of ensuring photovoltaic utilization,the demand satisfaction rate of energy consumption is improved,and Reduced charging costs.