A Novel Optimization Approach for Sustainable Renewable Energy Distribution and Multi-Source Resource Integration

The rapid expansion of renewable energy systems has introduced new challenges in ensuring the sustainable, efficient, and reliable distribution of electricity. Variability in generation from sources such as solar, wind, and hydro, combined with fluctuating demand, necessitates the development of intelligent optimization methods for multi-source resource integration. This research presents a novel optimization approach that integrates solar, wind, and hydro resources with energy storage and auxiliary backup systems to achieve efficient and sustainable distribution within microgrids. The proposed framework employs a hybrid machine learning-driven optimization model, combining Ladybug Beetle-driven Weighted Random Forest Regression (LB-Weighted RFR) for accurate power generation forecasting and optimized energy distribution. The forecasts enable dynamic scheduling that balances supply and demand, reduces peak loads, and maximizes renewable energy utilization. Forecasting incorporates historical production data, weather patterns, and grid conditions, achieving low error metrics such as MSE (0.00008) for energy sources. Simulation results demonstrate significant improvements of the proposed model over conventional approaches, including reductions in operating costs, decreases in peak demand, enhanced supply-demand balance, and increased renewable energy utilization.