EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Hybrid Deep Learning Framework with CEEMDAN, Multi-Scale CNN, and Multi-Head Attention for Building Load Forecasting

Limin Wang, Dezheng Wei, Jumin Zhao, Wei Gao, Dengao Li

Buildings · 21.03.2026 · 16 (6) · 1248

DOI: 10.3390/buildings16061248 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Buildings, online erschienen am 21.03.2026. Verfasst von Limin Wang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanxi Provincial Key Research and Development Program, State Grid Shanxi Electric Power Company Science and Technology Project, Shanxi Province Open Bidding for Selecting the Best Candidates Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate building load forecasting is essential for smart grid and energy management, yet nonlinearity, non-stationarity, and multi-scale characteristics of load data challenge traditional methods. To address these issues, we propose a hybrid deep learning framework, CEEMDAN-MultiScale-CNN-BiLSTM-MultiAttention. First, Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) decomposes the load sequence into intrinsic mode functions (IMFs), mitigating mode mixing and complexity. Then, a MultiScale Convolutional Neural Network extracts multi-scale local features from each IMF. A Bidirectional Long Short-Term Memory network captures bidirectional temporal dependencies, and a Multi-Attention mechanism dynamically emphasizes critical time steps and feature channels, enhancing interpretability and prediction. The framework is validated on the Building Data Genome Project 2 dataset, achieving a Mean Absolute Percentage Error (MAPE) of 2.6464% and a coefficient of determination R2 of 0.8999, outperforming mainstream methods across multiple metrics. The main contributions are: (1) a hybrid framework integrating CEEMDAN, multi-scale feature extraction, and attention mechanisms to handle nonlinearity and non-stationarity; (2) a MultiScale-CNN to capture multi-scale temporal features and adapt to multi-frequency components; (3) a Multi-Attention mechanism to dynamically focus on key time steps and channels, improving accuracy and robustness. This work provides an effective solution for building load forecasting in complex energy systems.

Förderung

  • Shanxi Provincial Key Research and Development Program
  • State Grid Shanxi Electric Power Company Science and Technology Project
  • Shanxi Province Open Bidding for Selecting the Best Candidates Project
  • National Natural Science Foundation of China

Zitieren

APA

Wang, L., Wei, D., Zhao, J., Gao, W., & Li, D. (2026). A Hybrid Deep Learning Framework with CEEMDAN, Multi-Scale CNN, and Multi-Head Attention for Building Load Forecasting. Buildings, 16(6), 1248. https://doi.org/10.3390/buildings16061248

Vancouver

Wang L, Wei D, Zhao J, Gao W, Li D. A Hybrid Deep Learning Framework with CEEMDAN, Multi-Scale CNN, and Multi-Head Attention for Building Load Forecasting. Buildings. 2026;16(6):1248. doi:10.3390/buildings16061248

BibTeX
@article{Wang2026,
  author = {Wang, Limin and Wei, Dezheng and Zhao, Jumin and Gao, Wei and Li, Dengao},
  title = {A Hybrid Deep Learning Framework with CEEMDAN, Multi-Scale CNN, and Multi-Head Attention for Building Load Forecasting},
  journal = {Buildings},
  year = {2026},
  volume = {16},
  number = {6},
  pages = {1248},
  doi = {10.3390/buildings16061248}
}
Kostenlos starten →