EQDEHO: An enhanced Elk herd optimizer for adaptive energy management in grid-connected microgrids with renewable and EV integration

Abstract

This research suggests EQDEHO, an enhanced version of Elk herd optimizer (EHO), for optimally managing the energy of a microgrid (MG) powered by renewable and traditional energy resources. The proposed approach includes three essential enhancements: dynamic elite mutation strategy (DEMS), enhanced solution quality strategy (ESQ), and adaptive bull rate strategy (ABRS). The DEMS is employed to achieve a gradual transition from broad exploration to focused exploitation to prevent falling into premature convergence. The ESQ is utilized to enhance exploration and exploitation capabilities, helping the algorithm to avoid local optima and progressively refine solution quality. Moreover, ABRS introduces a dynamic bull rate that evolves using exponential and cosine functions. It begins with a higTwo versions have been provided for Author Contributions. We have followed the .json as per style. Please check and correct if necessary. h value to encourage exploration and gradually decreases to strengthen exploitation, thereby improving search adaptability over time. The MG under consideration is a grid-connected and includes photovoltaic (PV) generating unit, wind turbine (WT), fuel cell (FC), micro-turbine (MT), battery storage system, and electric vehicles (EVs). The primary goals are to reduce the overall operational costs and environmental pollutant emissions while keeping the generation and dTwo versions have been provided for Author Contributions. We have followed the .json as per style. Please check and correct if necessary. emand balancing, generation restrictions, and storage limitations. Two situations are investigated: the first excludes EVs, while the second tests the EVs in three different charging modes: uncontrolled, controlled, and smart. The recommended EQDEHO is evaluated in contrast to the published Fuzzy self-adaptive particle swarm optimizer (FSAPSO) and other programmed algorithms. The proposed approach outperformed FSAPSO in terms of reducing MG operating costs and emissions by 11.642% and 56.856%, respectively, while the EVs are detached from the MG. Furthermore, the largest reductions when the EVs are plugged into the MG attained with the suggested EQDEHO are 23.018% compared to AOA, 32.840% compared to SWO, and 60.765% compared to AOA during uncontrolled, controlled, and smart modes, respectively. The obtained findings demonstrated the strength and competence of the proposed approach as an effective energy management solution for MG.