Power management and control of a DC microgrid with hybrid energy storage systems using hybrid ANN-based model predictive control

This work proposes a novel power management strategy (PMS) by using hybrid artificial neural networks (ANNs) based model predictive control (MPC) for DC microgrids (DCMG) with hybrid energy storage systems (HESS). The study has taken into account a DCMG that includes a Photovoltaic (PV) system, a wind energy system (WES), HESS consisting of a battery and supercapacitor (SC), and load. The proposed control technique is employed to enhance power-sharing between batteries and SC, alleviate demand-generation discrepancies, maintain state-of-charge (SoC) under boundaries, and regulate DC bus voltage. The suggested approach leads to enhanced battery longevity as a consequence of redirecting unutilized battery currents, including high-frequency elements, toward the supercapacitor. Moreover, the effectiveness of the proposed strategy is assessed by comparing it with the conventional control strategies in terms of dynamic and transient performance. Computational and real-time investigations are conducted to evaluate the efficacy of the proposed PMS across various case studies.