Reinforcement Learning control strategies for Electric Vehicles and Renewable energy sources Virtual Power Plants

Abstract

The increasing demand for direct electric energy in the grid is also tied to the increase of Electric Vehicle (EV) usage in the cities, which eventually will totally substitute combustion engine Vehicles. Nevertheless, this high amount of energy required, which is stored in the EV batteries, is not always used and it can constitute a virtual power plant on its own. Bidirectional EVs equipped with batteries connected to the grid can therefore charge or discharge energy depending on public needs, producing a smart shift of energy where and when needed. EVs employed as mobile storage devices can add resilience and supply/demand balance benefits to specific loads, in many cases as part of a Microgrid (MG). Depending on the direction of the energy transfer, EVs can provide backup power to households through vehicle-to-house (V2H) charging, or storing unused renewable power through renewable-to-vehicle (RE2V) charging. V2H and RE2V solutions can complement renewable power sources like solar photovoltaic (PV) panels and wind turbines (WT), which fluctuate over time, increasing the self-consumption and autarky. The concept of distributed energy resources (DERs) is becoming more and more present and requires new solutions for the integration of multiple complementary resources with variable supply over time. The development of these ideas is coupled with the growth of new AI techniques that will potentially be the managing core of such systems. Machine learning techniques can model the energy grid environment in such a flexible way that constant optimization is possible. This fascinating working principle introduces the wider concept of an interconnected, shared, decentralized grid of energy. This research on Reinforcement Learning control strategies for Electric Vehicles and Renewable energy sources Virtual Power Plants focuses on providing solutions for such energy supply optimization models.