A Novel Large-Scale Electric Vehicle Charging Load Forecasting Method and Its Application on Regional Power Distribution Networks

Abstract

The growing number of electric vehicles (EVs) will pose a potential threat to the existing residential microgrids and power distribution networks (PDNs). The larg-scale EV charging load will affect the operation of PDNs. In this paper, a daily load curve forecasting method for large-scale EV charging load is proposed by using K-means and long short-term memory neural network (LSTM) algorithms. To highlight the uncertainty of the future amount of EVs, we predict the quantity of EVs based on diverse EVs growth models. Taking into account the large-scale EV charging loads, a systematic methodology includes EV charging profiles and the future EV ownership can estimate the EV charging load. This method is verified by the empirical analyses in Hubei province in China. The simulation results indicate that the maximum value of the predicted EV charging load in 2025 would occur at 18:00 and equal 938.66 MW, which could elevate the existing load peak by 2.01% in 2025.