Machine Learning-Aided Optimization Approach for Sustainable Energy Systems

Energy derived from renewable sources is expanding quickly. Renewable energy systems meet off-grid and on-grid load requirements, often connected to conventional sources. Geothermal, solar, wind, and other renewable energy sources can be harnessed economically at the current rate of technological development. Renewable-based electricity production is an alternative form of green power generation that reduces issues with energy consumption. Though this raises numerous uncertainties and difficulties, it may jeopardize potential economic, technological, and environmental gains. Consequently, a Machine Learning aided Optimization Approach (MLOA) for sustainable energy systems has been suggested in this study. The Enhanced Long Short-Term Memory (E-LSTM) model of recurrent neural networks trained by Adam optimizer is employed in MLOA due to its ability to predict solar and wind generation to provide optimal energy consumption control. The numerical outcomes demonstrate that the suggested MLOA method improves prediction and accuracy rates by 98.22% compared to other current approaches.