Comparison of Hybrid DC-AC Microgrid Energy Management System Using Neural Network and Vector-Decoupled Algorithm Driven by Horse Herd Optimization

This research paper presents a comparative analysis of Energy Management Systems (EMS) in hybrid DC-AC microgrids, focusing on the application of a neural network vector-decoupled algorithm driven by Horse Herd Optimization (HHO). The objective is to assess the performance and competence of this innovative approach compared to traditional methods in controlling power flow and optimizing energy utilization within microgrid environments. The study begins by outlining the structure and components of a hybrid DC-AC microgrid scheme, emphasizing the incorporation of non-conventional resources, Battery, and the grid. Key components such as converters, inverters, and control mechanisms are discussed to provide a complete empathetic of the scheme. The core focus of this research lies in the comparison between conventional energy management techniques and the proposed neural network vector-decoupled algorithm enhanced by HHO. The neural network algorithm facilitates real-time decision-making and adaptive control, optimizing power flow and enhancing system stability. HHO further enhances the algorithm's efficiency by leveraging the collective intelligence of a simulated horse herd, mimicking their behavior of communication and collaboration for optimization. Simulation studies are done using MATLAB/Simulink to justify the efficacy of the suggested approach. Performance metrics such as system stability, power quality, energy losses, and economic considerations are analyzed and compared against baseline models and existing methodologies. The results demonstrate significant improvements in energy management efficiency, grid stability, and cost-effectiveness with the neural network vector-decoupled algorithm driven by HHO. The paper concludes with insights into the practical implications and future research directions for advancing energy management systems in hybrid DC-AC microgrids.