Design of a predictive model for electric vehicle smart charging based on cloud data

Abstract

The integration of distributed energy resources (DERs), such as electric vehicle charging stations (EVCSs), photovoltaic units (PVs), and battery energy storage systems (BESSs) is a key component towards decarbonization. In this context, the development of smart charging control strategies to encourage the usage of renewable energy sources (RESs) for electric vehicle (EV) charging has become a topical research subject. This research project proposes the development of demand forecasting modules for EV charging and photovoltaic generation. Once the proposed modules are finalized, the integration with a smart charging executed through an internet of things (IoT) platform to control DERs, locally integrated through an aggregated system. The objective is to diversify the source of energy for EV charging, in which BESSs and PVs are used to complement the energy from the main grid. The development of the forecasting modules will provide intelligent management of the BESSs in order to coordinate nighttime charging in situations where, the following day, PV generation is low and demand tends to be high. In addition, power injection from the BESSs to the main grid can be managed to compensate for the times when power is more expensive. Moreover, the possibility of anticipating the next day's load and PV generation levels allows techniques for reducing BESS degradation to be applied, as well as solutions for managing the loading of EV fleets considering the injection of power to the main grid through EVCSs with V2G technology. (AU)