Optimal Allocation And Placement Of Faster Ev Charging Station Using Machine Learning: A Review And Recent Developments.

Abstract

Electric Vehicles (EVs) demand is higher over the years. For addressing the associated fossil fuel automobile issues owing to climate change, the utilization of EVs has become more famous. When more EVs are introduced, concerns about the accessibility as well as convenience of Charging Stations (CSs) for consumers occur. Inadequate CSs are associated with consumers’ decisions to buy EVs. The critical component of the EVs’ accessibility along with future success is to place CSs in regions with developing charging infrastructure. The distribution system is affected by the charging of a large number of EVs. Therefore, voltage fluctuations, power losses, augmented transformer overloads, et cetera may occur. Since ML approaches like Random Forest (RF), K-Nearest Neighbours (KNN), Deep Neural Networks (DNN), Decision Tree (DT), Support Vector Machine (SVM), and Long Short-Term Memory (LSTM) can create accurate future decisions grounded on historical data, they are contrasted concerning their performances in optimization. Therefore, this paper explores the overview of the Electric Vehicle Charging Station (EVCS) infrastructure, optimal allocation, and placement of fast EVCS in distribution and hybrid energy sources using ML. By analyzing the outcomes, the reliability of the usage of ML for the management of EVCS could be verified.