Two-layer optimization approach for Electric Vehicle Charging Station with dynamic reconfiguration of charging points

This paper presents a two-layer optimization of a fast Electric Vehicle (EV) Charging Station powered by the grid, a Photovoltaic (PV) system, and a Battery Energy Storage System (BESS). The paper aims to increase profits by providing an energy schedule of the BESS and the grid, but also dynamically adjusting the power output of every Charging Point (CP). The first layer of optimization gives the daily energy scheduling in thirty-minute intervals considering forecast values of PV production, EV cumulative demand, and electrical price. Meanwhile, the second layer, based on Model Predictive Control, adapts in real time the energy scheduling from the first layer taking into account the actual EV power demand, and the PV power production. Additionally, it dynamically allocates power to each CP depending on the EVs remaining charging time which is estimated using the corresponding EV power curve. The power rate of each CP varies by mechanically changing the internal connection of the Charging Column (CC). We evaluate the proposed methodology by introducing forecast errors regarding the cumulative EV demand and PV power production on sunny and cloudy days. Additionally, we assess the real-time operation with diverse EV arrival times, EV power demand and random EV types. Our findings demonstrate that the optimal dynamic reconfiguration of the CC effectively enables adherence to the daily energy schedule, ensuring increased profit, and EV’s satisfaction without affecting the charging time. •A holistic energy management with demand response and adaptable charging columns.•Two-layer optimization for energy scheduling and dynamic power rate adaptation.•Model Predictive Control optimally and dynamically manages the charging points.•The impact of forecast error is assessed on profit and EV demand satisfaction.•Increment of infrastructure utilization and profit with increasing EV demand.