Forecasting Renewable Energy Employing A Regularized Deep Learning Approach

Renewable energy prediction has become an important research area due to the increasing demand for clean energy, the depletion of conventional fossil fuels, and the need to reduce greenhouse gas emissions. Among the various renewable energy sources, solar energy has gained considerable attention because of its wide availability, environmental sustainability, and potential for large-scale electricity generation. However, the amount of solar energy generated by photovoltaic (PV) systems depends on several environmental and meteorological factors, including solar irradiance, ambient temperature, humidity, cloud cover, wind speed, and atmospheric conditions. These factors exhibit significant temporal variations, making solar energy prediction a challenging task. The proposed approach combines the DWT and a Bayesian Optimized Deep Neural Network for forecasting renewable energy using solar irradiation dataset. It has been shown that the model attains a MAE of just 3.0 and an MAPE value of 0.03% which clearly outperforms existing baseline techniques in the domain.