EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DHFF: A Dynamic Meta-Learning Framework for Few-Shot Feeder-Level Air-Conditioning Load Forecasting

Chuan Long, Yunche Su, Wenhua Zhang, Xinting Yang, Ling Zhang, Fang Liu, Wei Chen, Haolan Yang

Energies · 20.05.2026 · 19 (10) · 2449

DOI: 10.3390/en19102449 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Energies, online erschienen am 20.05.2026. Verfasst von Chuan Long und 7 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: State Grid Sichuan Electric Power Company.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate prediction of air conditioning (AC) loads at the feeder level is imperative for the effective implementation of targeted demand response and effective peak load management. However, it is challenged by data heterogeneity and scarcity, especially for new feeders. Traditional methods often fail under such conditions. This paper proposes a meta-learning-based Dynamic Hierarchical Forecasting Framework (DHFF) explicitly designed for efficient few-shot load forecasting. A core gating network dynamically fuses predictions from a global unified model and a feeder specific model, adapting based on context. The framework’s effectiveness was validated through dual testing scenarios. Outstanding performance was achieved in data-rich environments, with an RMSE of 0.0135, improving upon a strong LSTM baseline by nearly 80%. Furthermore, rigorous few-shot experiments confirmed its primary design goal: under extreme data scarcity (e.g., 5% data), DHFF demonstrated superior accuracy, improving up to 7.01% over standard approaches by intelligently leveraging generalized knowledge. These results validate DHFF as an adaptive, high-performing solution across both data-scarce and data-rich feeder forecasting scenarios.

Förderung

  • State Grid Sichuan Electric Power Company

Zitieren

APA

Long, C., Su, Y., Zhang, W., Yang, X., Zhang, L., Liu, F., Chen, W., & Yang, H. (2026). DHFF: A Dynamic Meta-Learning Framework for Few-Shot Feeder-Level Air-Conditioning Load Forecasting. Energies, 19(10), 2449. https://doi.org/10.3390/en19102449

Vancouver

Long C, Su Y, Zhang W, Yang X, Zhang L, Liu F, et al. DHFF: A Dynamic Meta-Learning Framework for Few-Shot Feeder-Level Air-Conditioning Load Forecasting. Energies. 2026;19(10):2449. doi:10.3390/en19102449

BibTeX
@article{Long2026,
  author = {Long, Chuan and Su, Yunche and Zhang, Wenhua and Yang, Xinting and Zhang, Ling and Liu, Fang and Chen, Wei and Yang, Haolan},
  title = {DHFF: A Dynamic Meta-Learning Framework for Few-Shot Feeder-Level Air-Conditioning Load Forecasting},
  journal = {Energies},
  year = {2026},
  volume = {19},
  number = {10},
  pages = {2449},
  doi = {10.3390/en19102449}
}
Kostenlos starten →