Time Series Foundation Model for Improved Transformer Load Forecasting and Overload Detection
Energies · 31.01.2025 · 18 (3) · 660
DOI: 10.3390/en18030660 ↗Kurz gefasst
Fachartikel in der Zeitschrift Energies, online erschienen am 31.01.2025. Verfasst von Yikai Hou und 5 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 79 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Heilongjiang Provincial Natural Science Foundation of China, National Natural Science Foundation, China, Key Research and Development Program of Heilongjiang.
Abstract (Originalsprache)
Simple load forecasting and overload prediction models, such as LSTM and XGBoost, are unable to handle the increasing amount of data in power systems. Recently, various foundation models (FMs) for time series analysis have been proposed, which can be scaled up for large time series variables and datasets across domains. However, the simple pre-training setting makes FMs unsuitable for complex downstream tasks. Effectively handling real-world tasks depends on additional data, i.e., covariates, and prior knowledge. Incorporating these through structural modifications to FMs is not feasible, as it would disrupt the pre-trained weights. To address this issue, this paper proposes a frequency domain mixer, i.e., FreqMixer, framework for enhancing the task-specific analytical capabilities of FMs. FreqMixer is an auxiliary network for the backbone FMs that takes covariates as input. It has the same number of layers as the backbone and communicates with it at each layer, allowing the incorporation of prior knowledge without altering the backbone’s structure. Through experiments, FreqMixer demonstrates high efficiency and performance, reducing MAPE by 23.65%, recall by 87%, and precision by 72% in transformer load forecasting during the Spring Festival while improving precision by 192.09% and accuracy by 14% in corresponding overload prediction, all while processing data from over 160 transformers with just 1M additional parameters.
Förderung
- Heilongjiang Provincial Natural Science Foundation of China
- National Natural Science Foundation, China
- Key Research and Development Program of Heilongjiang
- Special projects for the central government to guide the development of local science and technology, China
- Harbin Manufacturing Technology Innovation Talent Project
- Natural Science Foundation of Heilongjiang Province of China
Zitieren
APA
Hou, Y., Ma, C., Li, X., Sun, Y., Yu, H., & Fang, Z. (2025). Time Series Foundation Model for Improved Transformer Load Forecasting and Overload Detection. Energies, 18(3), 660. https://doi.org/10.3390/en18030660
Vancouver
Hou Y, Ma C, Li X, Sun Y, Yu H, Fang Z. Time Series Foundation Model for Improved Transformer Load Forecasting and Overload Detection. Energies. 2025;18(3):660. doi:10.3390/en18030660
BibTeX
@article{Hou2025,
author = {Hou, Yikai and Ma, Chao and Li, Xiang and Sun, Yinggang and Yu, Haining and Fang, Zhou},
title = {Time Series Foundation Model for Improved Transformer Load Forecasting and Overload Detection},
journal = {Energies},
year = {2025},
volume = {18},
number = {3},
pages = {660},
doi = {10.3390/en18030660}
}
EUVIMED