Community electric vehicle load forecasting based on time series distribution

Abstract

Electric vehicle charging load forecasting helps power system planning management and optimization. This paper proposes a residential private electric vehicle load forecasting model based on positive-reverse valley time charging, which predicts the load while satisfying the orderly charging of large-scale private electric vehicles in residential areas. Firstly, it analyzes the historical travel rules of private electric vehicles in the community and the charging and discharging characteristics of electric vehicles. Secondly, based on the peak-to- valley time-of-use(TOU) electricity price, the owner of the vehicle is fully utilized to make the electric vehicle orderly charging, so as to obtain the total charging load of the electric vehicle in the community; Finally, the actual data is used to predict the change of the load curve of private electric vehicles in the future after large-scale access to the grid. It is found that under the scenario of disorderly charging of private electric vehicles, the larger volume of the grid has the greater difference between the peaks and valleys of the grid. Through further calculations, it is found that the use of valley time charging can alleviate the impact of the electric vehicle charging load on the power grid to a certain extent. That also can effectively reduce the peak-to-valley difference of the grid load and residential area distribution network overload rate.