Solar Resource Assessment in The Sahara: A Hybrid Dataphysics Framework for Photovoltaic Optimization

Leveraging data-driven prediction models in renewable energy systems (RES) is pivotal for boosting efficiency and optimizing power generation. This study investigates advanced machine learning (ML) techniques for accurate solar irradiance forecasting, a critical factor influencing photovoltaic (PV) system performance. Field measurements of solar irradiation, ambient temperature, and wind speed were collected in the Sahara Desert to develop and validate predictive models. Multiple nonlinear ML algorithms were implemented and rigorously compared in terms of accuracy and robustness. The Gradient Boosting Regressor (GBR) emerged as the most effective model, providing highly reliable solar irradiance predictions. These results demonstrate the potential of data-driven approaches to improve PV energy output assessment, support informed decision-making, and advance efficient management of large-scale renewable energy systems in challenging desert environments.