Spatio-Temporal Prediction of Solar Irradiance and Wind Speed Using Machine Learning: A Case Study in Tunisia
Authors
Laboratoire Optimisation, Logistique et Informatique Décisionnelle (OLID), ISGIS, University of Sfax, Sfax 3021, Tunisia (Tunisie)
Laboratory of Electromechanical Systems (LASEM), National Engineering School of Sfax, University of Sfax, Sfax 3038, Tunisia|Higher Institute of Industrial Management of Sfax, University of Sfax, Sfax 3021, Tunisia (Tunisie)
Higher Institute of Industrial Management of Sfax, University of Sfax, Sfax 3021, Tunisia|OASIS Laboratory, ENIT, Tunisia (Tunisie)
Article Information
DOI: 10.51244/IJRSI.2026.1310000170
Subject Category: Renewable energy
Volume/Issue: 13/10 | Page No: 4585-4596
Publication Timeline
Submitted: 2026-10-20
Accepted: 2026-10-25
Published: 2026-11-12
Abstract
Accurate prediction of solar irradiance and wind speed is essential for renewable energy resource assessment and the planning of photovoltaic (PV), wind, and hybrid PV-wind systems.
This study investigates a spatio-temporal machine learning framework for predicting meteorological variables using observations reported as collected by the National Institute of Meteorology (INM) of Tunisia from 1 September 2023 to 1 September 2024. The dataset covers five Tunisian locations and includes global solar irradiation, wind speed, air temperature, and geographic coordinates. Three regression algorithms , Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) are considered to model the nonlinear relationships between meteorological conditions and renewable energy resources. The comparative results reported in the source manuscript show that XGBoost achieves the highest coefficient of determination ((R² = 0.812)), followed by Random Forest ((R² = 0.803)) and Decision Tree ((R² = 0.732)). XGBoost also yields the lowest reported mean absolute error (MAE = 168.146) and root mean square error (RMSE = 429.362).
However, the available results do not clearly specify the target variable, measurement units, or validation protocol, preventing these performance metrics from being attributed independently to solar irradiance and wind speed. Accordingly, the study establishes a methodological framework for evaluating both prediction tasks and highlights the need for target-specific, temporally consistent validation. Once validated, the forecasts can support renewable resource assessment and preliminary PV-wind system design, providing a basis for subsequent energy yield estimation and techno-economic optimization
Keywords
Solar irradiance, wind speed forecasting, machine learning, XGBoost, Random Forest, Decision Tree , spatio-temporal modelling, renewable energy.
Downloads
References
1. C. Voyant, G. Notton, S. A. Kalogirou, M.-L. Nivet, C. Paoli, F. Motte, and A. Fouilloy, “Machine learning methods for solar radiation forecasting: A review,” Renewable Energy, vol. 105, pp. 569–582, 2017, doi: 10.1016/j.renene.2016.12.095. [Google Scholar] [Crossref]
2. M. A. Inman, H. T. C. Pedro, and C. F. M. Coimbra, “Solar forecasting methods for renewable energy integration,” Progress in Energy and Combustion Science, vol. 39, no. 6, pp. 535–576, 2013, doi: 10.1016/j.pecs.2013.06.002. [Google Scholar] [Crossref]
3. K. Barhmi, C. Heynen, S. Golroodbari, and W. van Sark, “A review of solar forecasting techniques and the role of artificial intelligence,” Solar, vol. 4, no. 1, pp. 99–135, 2024, doi: 10.3390/solar4010005. [Google Scholar] [Crossref]
4. “Machine learning algorithms for solar irradiance prediction: A recent comparative study,” e-Prime—Advances in Electrical Engineering, Electronics and Energy, vol. 7, Art. no. 100453, 2024, doi: 10.1016/j.prime.2024.100453. [Google Scholar] [Crossref]
5. D. Alves, F. Mendonça, S. S. Mostafa, and F. Morgado-Dias, “The potential of machine learning for wind speed and direction short-term forecasting: A systematic review,” Computers, vol. 12, no. 10, Art. no. 206, 2023, doi: 10.3390/computers12100206. [Google Scholar] [Crossref]
6. S. M. Valdivia-Bautista, J. A. Domínguez-Navarro, M. Pérez-Cisneros, C. J. Vega-Gómez, and B. Castillo-Téllez, “Artificial intelligence in wind speed forecasting: A review,” Energies, vol. 16, no. 5, Art. no. 2457, 2023, doi: 10.3390/en16052457. [Google Scholar] [Crossref]
7. Y. Yang et al., “A survey on wind power forecasting with machine learning approaches,” Neural Computing and Applications, vol. 36, pp. 12753–12773, 2024, doi: 10.1007/s00521-024-09923-4. [Google Scholar] [Crossref]
8. R. J. Hyndman and G. Athanasopoulos, Forecasting: Principles and Practice, 3rd ed. Melbourne, Australia: OTexts, 2021. [Online]. Available: [Google Scholar] [Crossref]
9. A. M. Tashman, “Out-of-sample tests of forecasting accuracy: An analysis and review,” International Journal of Forecasting, vol. 16, no. 4, pp. 437–450, 2000, doi: 10.1016/S0169-2070(00)00065-0. [Google Scholar] [Crossref]
10. L. Breiman, “Random forests,” Machine Learning, vol. 45, pp. 5–32, 2001, doi: 10.1023/A:1010933404324. [Google Scholar] [Crossref]
11. J. H. Friedman, “Greedy function approximation: A gradient boosting machine,” The Annals of Statistics, vol. 29, no. 5, pp. 1189–1232, 2001, doi: 10.1214/aos/1013203451. [Google Scholar] [Crossref]
12. T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 2016, pp. 785–794, doi: 10.1145/2939672.2939785. [Google Scholar] [Crossref]
13. V. K. Saini et al., “Learning based short term wind speed forecasting models for smart grid applications: An extensive review and case study,” Electric Power Systems Research, 2023, Art. no. 109502, doi: 10.1016/j.epsr.2023.109502. [Google Scholar] [Crossref]
14. H. Liu and Z. Zhang, “Development and trending of deep learning methods for wind power predictions,” Artificial Intelligence Review, vol. 57, Art. no. 112, 2024, doi: 10.1007/s10462-024-10728-z. [Google Scholar] [Crossref]
15. O. Khouili et al., “Evaluating the impact of deep learning approaches on solar and photovoltaic power forecasting: A systematic review,” Energy Strategy Reviews, vol. 59, Art. no. 101735, 2025, doi: 10.1016/j.esr.2025.101735. [Google Scholar] [Crossref]
16. “A review on data-driven methods for solar energy forecasting,” Applied Energy, vol. 400, Art. no. 126631, 2025, doi: 10.1016/j.apenergy.2025.126631. [Google Scholar] [Crossref]
17. “Artificial intelligence for wind speed forecasting: A review on multi-scale decomposition and intelligent fusion strategies,” Advances in Wind Engineering, vol. 2, no. 2, Art. no. 100055, 2025, doi: 10.1016/j.awe.2025.100055. [Google Scholar] [Crossref]
18. “Forecast evaluation for data scientists: Common pitfalls and best practices,” Data Mining and Knowledge Discovery, 2023, doi: 10.1007/s10618-022-00894-5. [Google Scholar] [Crossref]
Metrics
Views & Downloads
Similar Articles
- Work Function of Metal Back Contact Surface Alloy Molybdenum (Mo) With Tungsten (W) For Copper Indium Gallium Selenide (Cigs) Thin Film Solar Cell: Simulation Using Scaps-1d And Density Functional Theory (Dft) Using Winmostar Quantum Espresso
- Biodegradable Materials and Renewable Materials Innovation with Eggshell Powder in Sustainable Product Design
- Investigation of Zinc Cobaltite (ZnCo₂O₄) as a Hole Transport Layer for Perovskite Solar Cells: Implications for Renewable-Energy Research and Innovation
- Predictive Modeling of Demand Response Impact on Solar-Integrated Power Systems Using Bayesian Optimisation Long Short-Term Memory Neural Networks
- Gis-Based Spatial Classification of Onshore Wind Energy Potential in Nigeria