Enhanced Energy Management Scheme for Secure, Emission-Conscious, and Optimized Operation of Networked Microgrids

The growing imperative to address environmental concerns and reduce dependence on fossil fuels is accelerating the global transition to sustainable energy systems. In this context, microgrids (MGs), which integrate diverse distributed energy resources (DERs) such as solar photovoltaics and wind turbines, are emerging as a robust and eco-friendly energy solution. Despite their advantages, the inherent intermittency of renewable sources and the dynamic nature of grid conditions introduce complexities in maintaining stable supply-demand balance, reducing emissions, and ensuring cost-efficient MG operation. To this end, this paper proposes an intelligent energy management system (EMS) scheme integrated with machine learning (ML) to enhance forecasting accuracy, optimize energy scheduling, and support reliable and low-emission operation of networked MGs (NMGs). The proposed system uses ML models trained on historical generation profiles, weather patterns, and grid parameters to predict renewables generation. These forecasts are integrated into an optimization model, solved using the Hippopotamus Optimizer (HO), to obtain optimal DER dispatch and energy transactions among NMGs. Detailed case studies highlight the effectiveness of the proposed scheme in maintaining energy balance, reducing emissions, and improving system reliability, highlighting its potential to support the deployment of sustainable and resilient networks.