Advanced power management algorithm for PV-EV charging stations using a real-time model predictive control

In the context of grid-connected PV-EV charging stations, efficient power management is a crucial issue. However, existing approaches often rely on the stability of the electrical grid, which can be disrupted by grid faults, causing EV charging interruptions. Moreover, neglecting real-time adjustments and battery electric vehicle (BEV) state of charge can lead to battery damage due to over-current or overvoltage situations, regardless of weather conditions. To address these limitations, a novel station manager algorithm is proposed, which dynamically adjusts power flow among the PV system, EV power demand, and the grid based on real-time measurements of system powers, grid availability, and BEV state of charge. This dynamic adjustment ensures an uninterrupted power supply to the EV while maintaining its battery safe during the charging operation. The proposed station manager introduces multiple operating modes, including adaptive charging mode and fast charging mode, each integrated with a dedicated model predictive controller (MPC) to achieve its specific control objective. Through a semi-experimental simulation using a process-in-the-loop (PIL) test approach on an embedded board, the eZdsp TMS320F28335, the results demonstrate the effectiveness of the algorithm in balancing power flow between the PV power and the BEV, optimizing energy utilization, and ensuring uninterrupted and reliable power supply to the EV.