Comparison of day-ahead photovoltaic power forecasting models using site-relative NWP deviation features
Journal of Renewable and Sustainable Energy · 01.07.2026 · 18 (4)
DOI: 10.1063/5.0307354 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Renewable and Sustainable Energy, online erschienen am 31.07.2026. Verfasst von Hongying Bai und 4 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Gefördert von: Ordos Higher Education Institutions Scientific Research Innovation Project.
Abstract (Originalsprache)
Accurate day-ahead forecasting of photovoltaic power is essential for grid scheduling, but forecast skill is limited by meteorological uncertainty and by the mismatch between coarse numerical weather prediction (NWP) outputs and site-level conditions. This study develops a framework that combines physically based capacity-factor analysis with data-driven day-ahead forecasting on the Photovoltaic Power Output Dataset, a multi-site dataset. We first use a theoretical capacity factor model to quantify the gap between ideal and measured generation. We then evaluate the main forecasting models used in this study and analyze their behavior under different weather conditions. To better represent local departures from coarse NWP inputs, we construct site-relative deviation features for temperature and irradiance from each station's difference from the contemporaneous multi-site mean. These variables are treated as lightweight correction features rather than meteorological downscaling. Results show that the Light Gradient Boosting Machine (LightGBM) provides the best overall accuracy for the NWP-enhanced benchmark, while RandomForest is more robust in low-power and outlier-prone periods. The site-relative deviation features produce model-dependent gains: the root mean square error (RMSE) of LightGBM decreases from 0.9852 to 0.9511, and the RMSE of RandomForest decreases from 1.1659 to 1.1544, whereas Extreme Gradient Boosting shows no improvement. The revised findings indicate that simple NWP correction features can be operationally useful, but their benefits are modest and should not be interpreted as a substitute for true NWP downscaling.
Förderung
- Ordos Higher Education Institutions Scientific Research Innovation Project
Zitieren
APA
Bai, H., Wang, S., Zhao, X., Wang, S., & Zhao, X. (2026). Comparison of day-ahead photovoltaic power forecasting models using site-relative NWP deviation features. Journal of Renewable and Sustainable Energy, 18(4). https://doi.org/10.1063/5.0307354
Vancouver
Bai H, Wang S, Zhao X, Wang S, Zhao X. Comparison of day-ahead photovoltaic power forecasting models using site-relative NWP deviation features. Journal of Renewable and Sustainable Energy. 2026;18(4). doi:10.1063/5.0307354
BibTeX
@article{Bai2026,
author = {Bai, Hongying and Wang, Sheng and Zhao, Xiaoyu and Wang, Shaojie and Zhao, Xibo},
title = {Comparison of day-ahead photovoltaic power forecasting models using site-relative NWP deviation features},
journal = {Journal of Renewable and Sustainable Energy},
year = {2026},
volume = {18},
number = {4},
doi = {10.1063/5.0307354}
}
EUVIMED