From Single Buildings to Clusters: A Pre-Trained Large Language Model-Based Framework for Cross-Building and Data-Scarce Energy Consumption Forecasting
Buildings · 18.08.2026 · 16 (16) · 3281
DOI: 10.3390/buildings16163281 ↗Kurz gefasst
Fachartikel in der Zeitschrift Buildings, online erschienen am 18.08.2026. Verfasst von Changhao Wang und 7 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Liaoning Provincial Science and Technology Joint Program Project.
Abstract (Originalsprache)
Accurate short-term building energy consumption forecasting is crucial for energy-system operational efficiency and demand-side management. Existing deep learning models heavily rely on historical data, making them costly for data-scarce buildings. Furthermore, their cross-building generalization ability is limited, requiring building-specific tuning that hinders large-scale deployment. To address these issues, this study proposes a novel framework based on large language models for short-term building energy consumption forecasting. It adapts large language models through parameter-efficient LoRA fine-tuning and incorporates building domain knowledge by using enhanced feature extraction modules and prompt design. Specifically, a prompt template rich in building physical semantics was designed to leverage the abundant pre-training knowledge of the large language model (LLM). This enables the model to avoid blindly fitting the data, directly aligning with building operating rules, providing a reasonable prediction basis even with limited data, and enhancing cross-building generalization. In addition, a cross-feature attention mechanism is designed to analyze the impact of dynamic meteorological features on energy consumption, thereby improving cross-climate scenario adaptability. Finally, to handle non-typical mutations in actual building operations, depthwise separable convolutional layers decompose residual components to filter out noise while preserving key features of anomalous occupancy patterns, thereby enhancing model robustness. Experiments on five real-world datasets have shown that the proposed framework outperforms state-of-the-art baselines, achieving an average improvement of 6.06% in the MAE and 4.80% in the RMSE. More importantly, it exhibits strong few-shot learning capabilities and extends to zero-shot forecasting. By reducing data dependency and enabling cross-building generalization, the framework developed in this work achieves scalable, low adaptation cost energy consumption forecasting from single buildings to clusters.
Förderung
- Liaoning Provincial Science and Technology Joint Program Project
Zitieren
APA
Wang, C., Li, S., Arıcı, M., Yang, R., Xu, X., Wang, Z., A, S., & Liu, S. (2026). From Single Buildings to Clusters: A Pre-Trained Large Language Model-Based Framework for Cross-Building and Data-Scarce Energy Consumption Forecasting. Buildings, 16(16), 3281. https://doi.org/10.3390/buildings16163281
Vancouver
Wang C, Li S, Arıcı M, Yang R, Xu X, Wang Z, et al. From Single Buildings to Clusters: A Pre-Trained Large Language Model-Based Framework for Cross-Building and Data-Scarce Energy Consumption Forecasting. Buildings. 2026;16(16):3281. doi:10.3390/buildings16163281
BibTeX
@article{Wang2026,
author = {Wang, Changhao and Li, Shanshan and Arıcı, Müslüm and Yang, Ruitong and Xu, Xinyue and Wang, Ziyang and A, Sina and Liu, Sichen},
title = {From Single Buildings to Clusters: A Pre-Trained Large Language Model-Based Framework for Cross-Building and Data-Scarce Energy Consumption Forecasting},
journal = {Buildings},
year = {2026},
volume = {16},
number = {16},
pages = {3281},
doi = {10.3390/buildings16163281}
}
EUVIMED