Forecasting electric vehicles recharge impact on distribution network

Abstract

The expansion of the electric vehicles market has resulted in the major emergence of new models with increasingly efficient charging systems, designed to meet drivers' everyday needs. This emergence will undoubtedly result in the proliferation of electric vehicle charging stations that will be connected to the electrical distribution network for medium voltage power. In this paper, we have demonstrated the impact of fast charging of electric vehicles during peak periods exceeding the power limit of the distribution network compared to random charging. Additionally, two artificial intelligence models, Artificial Neural Network (ANN) and Random Forest (RF), were used to predict potential power surges resulting from electric vehicle fast charging. The prediction results show high performance for the RF model compared to those obtained by ANN model.