Smart and Coordinated Charging Using Ensemble Model for Plug-in Electric Vehicle

This work proposed smart and coordinated Plugin Electric Vehicle (PEV) charging technique, which incorporates the forecasting of electricity prices. This work aims to establish the most cost-effective charging schedule that satisfies both vehicle owners and network performance. This work presents the formulation of the PEV charging and discharging scheduling problem from the user's perspective aiming to minimize charging cost and satisfy the EV charging demand while considering the battery's state of charge (SOC), a narrow arrival time range, and electricity price forecasting (EPF). In this work, the control signal proposed using the ML method based on employing an ensemble classifier, which divides charging time hours into low and high charging zones. The proposed ensemble model showed significant improvement in prediction accuracy compared to other machine learning models namely Neural Networks (NN) and Naïve Bayes (NB).