Deep learning-based photovoltaic energy forecasting with ground-based sky imagery and atmospheric data

Abstract

Due to the rise in the use of renewable energies as an alternative to traditional energies, and especially solar energy, there is increasing interest in studying how to address photovoltaic forecasting in the face of the challenge of variability in photovoltaic energy production, using different techniques and methodologies. This work develops a hybrid methodology for short and long-term forecasting based on two studies with the same purpose. A multimodal approach is proposed that combines images of the sky and photovoltaic energy history with meteorological data. The main objective is to improve the accuracy of ramp event prediction, increase the robustness of forecasts in cloudy conditions, and extend capabilities beyond nowcasting, in order to support more efficient operation of the power grid and better management of solar variability. Deep convolutional neural network architectures are used for nowcasting and forecasting models, incorporating individual and multiple meteorological variables, as well as solar position. The results demonstrate that the inclusion of meteorological data, particularly the surface long-wave (thermal), radiation downwards (strd), and the combination of wind and solar position, significantly improves predictions in both nowcasting and forecasting tasks, especially on cloudy days. This study highlights the importance of integrating diverse data sources to improve the reliability and interpretability of solar energy prediction models.