ABSTRACT The escalating global energy demand necessitates a shift towards sustainable and environmentally friendly alternatives. While renewable energy sources like solar and wind energy offer promising solutions, their intermittent nature poses significant challenges for grid integration. Microgrids, as decentralized and self‐sufficient energy systems, emerge as a viable solution by enabling the efficient integration of these renewable sources. However, effective energy management within a microgrid remains crucial to ensure optimal power flow and minimize energy losses. This paper proposes a novel optimized dual Artificial Neural Network (ANN) control strategy for a direct current (DC) microgrid. The first ANN focuses on maximizing power extraction from renewable sources by employing real‐time irradiance, temperature, and wind speed data. The second ANN intelligently controls the power flow between renewable sources, energy storage systems, the grid, and the load. Simulation results under variable load profiles for voltage, current, and power of different DC microgrid components demonstrate this approach's effectiveness in efficiently distributing power and mitigating energy losses, highlighting its potential for enhancing the reliability and sustainability of future microgrid systems. Simulation results confirmed the effectiveness of the proposed dual ANN controller, achieving a more consistent voltage and current profile, reduced energy losses, and improved power management during fluctuating load conditions. This approach demonstrated up to (insert quantitative result: for example, “20% reduction in energy losses compared with baseline EMS methods”) and maintained output stability throughout different operating scenarios.
Li et al. (2026) studied this question.