EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Enhancing Photovoltaic Power Forecasting Using Soft Physical Constraint‐Based Deep Learning Under Limited Data Conditions

Da-Hun You, Kun-Yik Jo, Hyun-Ho Park, Seon-Ju Ahn, Sang-Yun Yun, Joon-Ho Choi

International Journal of Energy Research · 2026-01 · 2026 (1)

DOI: 10.1155/er/5328962 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Energy Research, online erschienen am 14.11.2026. Verfasst von Da-Hun You und 5 weiteren Autor:innen. Die Arbeit nennt 95 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea Institute of Energy Technology Evaluation and Planning.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate photovoltaic (PV) power forecasting (PPF) is essential for reliable grid operation and efficient utilization of renewable energy resources. However, conventional deep learning‐based forecasting models often fail to adequately reflect the physical characteristics of PV generation and tend to suffer from performance degradation in multistep forecasting. To address this issue, this article proposes a soft physical constraint‐based deep learning framework for PPF, referred to as SPC‐DL‐PPF. The proposed approach incorporates physical constraints into baseline deep learning models, including long short‐term memory (LSTM), attention mechanism (AM), and Transformer architectures, through a loss function to guide the training process. Experimental results under various input configurations and single‐step and multistep forecasting scenarios demonstrate that the proposed model improves forecasting accuracy compared with baseline models and effectively mitigates performance degradation in multistep forecasting. Furthermore, the proposed approach maintains stable forecasting performance under limited data conditions and shows the potential for application to nearby PV sites by using sensor measurements from a reference site together with site‐specific physical characteristics.

Förderung

  • Korea Institute of Energy Technology Evaluation and Planning

Zitieren

APA

You, D., Jo, K., Park, H., Ahn, S., Yun, S., & Choi, J. (2026). Enhancing Photovoltaic Power Forecasting Using Soft Physical Constraint‐Based Deep Learning Under Limited Data Conditions. International Journal of Energy Research, 2026(1). https://doi.org/10.1155/er/5328962

Vancouver

You D, Jo K, Park H, Ahn S, Yun S, Choi J. Enhancing Photovoltaic Power Forecasting Using Soft Physical Constraint‐Based Deep Learning Under Limited Data Conditions. International Journal of Energy Research. 2026;2026(1). doi:10.1155/er/5328962

BibTeX
@article{You2026,
  author = {You, Da-Hun and Jo, Kun-Yik and Park, Hyun-Ho and Ahn, Seon-Ju and Yun, Sang-Yun and Choi, Joon-Ho},
  title = {Enhancing Photovoltaic Power Forecasting Using Soft Physical Constraint‐Based Deep Learning Under Limited Data Conditions},
  journal = {International Journal of Energy Research},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/er/5328962}
}
Kostenlos starten →