EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Structure-Guided Dual-Timescale Learning for Heterogeneous Short-Horizon Time-Series Forecasting

Tian Peng, Junrong Ye, Min Qi, Yuqiang Bai, Honghao Wu

Eng · 01.07.2026 · 7 (7) · 318

DOI: 10.3390/eng7070318 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Eng, online erschienen am 01.07.2026. Verfasst von Tian Peng und 4 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Beijing Municipal Education Commission.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Many real-world forecasting tasks involve heterogeneous temporal sources, where high-frequency dynamic observations coexist with low-frequency semantic or contextual information. Existing methods often treat these inputs through simple broadcasting or direct concatenation, which weakens cross-scale dependency modeling and makes the prediction process vulnerable to unstable contextual correlations. To address this issue, this paper proposes a structure-guided dual-timescale learning framework for heterogeneous short-horizon time-series forecasting. A fast-scale temporal encoder is used to extract fine-grained dynamic patterns, while a slow-scale semantic encoder is introduced to characterize slowly varying contextual information. On this basis, a structure-guided semantic refinement module is designed to suppress unstable semantic components before fusion, and a structure-aware cross-scale attention mechanism is developed to adaptively align fast dynamics with the most relevant slow-varying context. In addition, a coarse-to-fine prediction structure is employed to separate background tendency modeling from short-term fluctuation correction, and intervention-based consistency regularization is incorporated to improve robustness under changing conditions. Feature contribution analysis further confirms the importance of selecting effective contextual variables to form a compact slow-timescale representation. Experiments on a real-world ultra-short-term aggregate load forecasting task demonstrate that the proposed framework achieves superior accuracy and robustness, indicating its potential as a general solution for forecasting problems with heterogeneous temporal resolutions.

Förderung

  • Beijing Municipal Education Commission

Zitieren

APA

Peng, T., Ye, J., Qi, M., Bai, Y., & Wu, H. (2026). Structure-Guided Dual-Timescale Learning for Heterogeneous Short-Horizon Time-Series Forecasting. Eng, 7(7), 318. https://doi.org/10.3390/eng7070318

Vancouver

Peng T, Ye J, Qi M, Bai Y, Wu H. Structure-Guided Dual-Timescale Learning for Heterogeneous Short-Horizon Time-Series Forecasting. Eng. 2026;7(7):318. doi:10.3390/eng7070318

BibTeX
@article{Peng2026,
  author = {Peng, Tian and Ye, Junrong and Qi, Min and Bai, Yuqiang and Wu, Honghao},
  title = {Structure-Guided Dual-Timescale Learning for Heterogeneous Short-Horizon Time-Series Forecasting},
  journal = {Eng},
  year = {2026},
  volume = {7},
  number = {7},
  pages = {318},
  doi = {10.3390/eng7070318}
}
Kostenlos starten →