AI-Driven Smart Energy Systems for Sustainable Renewable Power Generation

This chapter presents the Smart Energy Forecasting and Optimisation System (SEFOS), an intelligent framework based on an Enhanced Long Short-Term Memory Recurrent Neural Network (LSTM-RNN). Aimed at residential microgrids with photovoltaic panels and battery storage, SEFOS addresses the challenges of fluctuating energy demand, variable solar availability, and energy management inefficiencies. By leveraging the predictive strengths of LSTM-RNN, SEFOS significantly improves solar power forecasting accuracy and optimises energy consumption. Simulation results confirm enhanced energy efficiency, reduced operational costs, and increased reliance on renewable sources, underscoring the transformative role of deep learning in sustainable energy systems.