A two-stage framework for photovoltaic power forecasting based on decomposition
Journal of Renewable and Sustainable Energy · 01.07.2026 · 18 (4)
DOI: 10.1063/5.0331359 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Renewable and Sustainable Energy, online erschienen am 10.07.2026. Verfasst von Yifeng Yin und 3 weiteren Autor:innen. Die Arbeit nennt 22 Literaturangaben. Gefördert von: Natural Science Foundation of Henan Province, Key Foundation of Science and Technology Development of Henan Province, Key Scientific Research Project of Higher Education of Henan Province.
Abstract (Originalsprache)
Photovoltaic power forecasting is vital for power grid stability, but the existing signal decomposition methods face the dual challenges of low computational efficiency and poor feature availability. This paper proposes a two-stage forecasting framework that integrates feature reconstruction and component screening. First, the reconstruction component (RC) features are constructed through rolling variational mode decomposition and terminal-point reorganization. Then screen the key RCs, and design to forecast the key RC characteristics first, and then forecast the end-to-end process of the final power to address the issue of uneven performance in component forecasting. Through evaluation, the efficiency of feature construction of the proposed RC method has been improved by 75% compared with the traditional rolling decomposition method that directly stores all complete intrinsic mode function sequences. In addition, the mean absolute error is reduced by 45.8% compared with the baseline temporal convolutional network method. On the premise that temporal causality is ensured, the proposed approach improves the forecast accuracy and calculation efficiency, and provides a feasible solution for engineering applications.
Förderung
- Natural Science Foundation of Henan Province
- Key Foundation of Science and Technology Development of Henan Province
- Key Scientific Research Project of Higher Education of Henan Province
- The CSC Visiting Fellow Scholarship
- Open Foundation of key laboratory of Cyberspace Security, Ministry of Education of China and Henan Key Laboratory of Network Cryptography
Zitieren
APA
Yin, Y., Su, R., Zhang, Y., & Wang, H. (2026). A two-stage framework for photovoltaic power forecasting based on decomposition. Journal of Renewable and Sustainable Energy, 18(4). https://doi.org/10.1063/5.0331359
Vancouver
Yin Y, Su R, Zhang Y, Wang H. A two-stage framework for photovoltaic power forecasting based on decomposition. Journal of Renewable and Sustainable Energy. 2026;18(4). doi:10.1063/5.0331359
BibTeX
@article{Yin2026,
author = {Yin, Yifeng and Su, Rui and Zhang, Yanhua and Wang, Hui},
title = {A two-stage framework for photovoltaic power forecasting based on decomposition},
journal = {Journal of Renewable and Sustainable Energy},
year = {2026},
volume = {18},
number = {4},
doi = {10.1063/5.0331359}
}
EUVIMED