AUTHOR=Ghaedrahmati Hadis , Talebi Saeed , Moradi Amirmohammad , Eskandari Aref , Parvin Parviz , Aghaei Mohammadreza 

TITLE=Potential analysis and energy prediction of photovoltaic power plants using satellite-based remote sensing and artificial intelligence techniques

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2025.1611429

DOI=10.3389/fenrg.2025.1611429

ISSN=2296-598X

ABSTRACT=Photovoltaic (PV) systems have seen significant global growth due to their economic and environmental benefits. However, the output of PV systems is subject to uncertainties arising from factors like unpredictable weather conditions. Given the considerable uncertainty in meteorological data, Geographic Information Systems (GIS) have emerged as effective tools for analyzing such data. This study presents a novel method based on satellite-based remote sensing and artificial intelligence techniques to assess the potential of PV power plants and predict energy generation in different locations. We utilize GIS and the Analytic Hierarchy Process (AHP) in ArcGIS software to evaluate suitable sites for PV systems. Satellite data from global sources is used to analyze PV energy production based on specific geographic coordinates. Several machine learning algorithms, including Random Forest (RF), Support Vector Regression (SVR), Decision Tree (DT), and XGBoost, are applied to predict PV energy production from meteorological variables. The evaluation, using various statistical metrics, shows that the XGBoost algorithm outperforms others, achieving up to 91% accuracy in predicting energy production from PV systems.