Authors: Faiza Mehmood, Muhammad Nabeel Asim, Aamir Mehmood, Waqar Mahmood, Sebastian Vollmer, Mehmood Ul Haq, Faisal Ali, Andreas Dengel
Overview
Solar irradiance forecasting is essential for integrating solar power into existing grids, yet the research on it is highly fragmented. Hundreds of methods, datasets and benchmarks make it hard to see what actually moves the field forward.
This meta-review, carried out with DFKI (the German Research Center for Artificial Intelligence), analyses 72 review articles on solar irradiance forecasting. A structured, reproducible framework is used to identify, screen and analyse the literature, datasets, source code and tools.
The result is a single map of the field: forecasting horizons, model families, benchmark datasets, evaluation practice and publication venues, ending with the research gaps most worth pursuing next.
Key findings
- Synthesises 72 review articles on solar irradiance forecasting.
- Groups forecasting horizons into very short-term, short-term, medium-term and long-term.
- Classifies models into eight families: physical, statistical, machine learning, deep learning, ensemble, hybrid, fuzzy logic-based and persistence.
- Compiles a repository of benchmark datasets and open-source tools to support reproducible, standardised evaluation.
- Summarises methods, evaluation metrics and state-of-the-art results for each model family across widely used datasets.
- Identifies explainable AI and physics-informed neural networks as key directions for future work.
This is our summary of the paper. The full text, with its abstract, figures and data, is on the publisher's site.
Details
- Journal
- Results in Engineering
- Published
- Dec 2026
- Volume
- 32
- Article number
- 112449
- DOI
- 10.1016/j.rineng.2026.112449
Topics
Cite
Faiza Mehmood, Muhammad Nabeel Asim, Aamir Mehmood, Waqar Mahmood, Sebastian Vollmer, Mehmood Ul Haq, Faisal Ali, Andreas Dengel (2026). Solar irradiance forecasting meta-review: An in-depth systematic meta-review on solar irradiance forecasting datasets, predictive pipelines and performance. Results in Engineering, 32, 112449. https://doi.org/10.1016/j.rineng.2026.112449










