EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physics-Guided Neural ODEs for Building Thermal Prediction and Model Predictive Control

Yagang Wang, Lexue Chang, Enzhan Zhang, Kejun Jia, Xia Zhang, Yonghao Li

Buildings · 29.07.2026 · 16 (15) · 3021

DOI: 10.3390/buildings16153021 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Buildings, online erschienen am 29.07.2026. Verfasst von Yagang Wang und 5 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Gansu Provincial Key Laboratory of Solar Energy Utilization.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Building supervisory control requires accurate, prior-consistent, and computationally tractable thermal models. This study proposes a physics-guided neural ordinary differential equation (PG-NODE) framework for thermal prediction and model predictive control (MPC). It combines a resistance–capacitance (RC)-inspired reference, a neural residual, RC-prior directional regularization, and validation-based checkpoint selection; the selected predictor remains fixed during MPC operation. Using 30 min EnergyPlus data from five zones, the framework was evaluated through held-out prediction, Gaussian noise and control-input-mismatch tests, ablation, and surrogate-based closed-loop experiments. Long short-term memory achieved the lowest prediction root mean square error (RMSE), whereas PG-NODE achieved the lowest RMSE among the evaluated neural ODE models and the lowest directional inconsistency rate among learned nonlinear models. In five-seed × five-window BACK SPACE experiments using independently trained, fixed PG-NODE surrogate environments rather than direct EnergyPlus interaction, fixed-block analysis supported lower reference-tracking RMSE for PG-NODE-MPC than for rule-based control and lower tariff-weighted normalized control effort than for the extended Kalman filter-based RC-MPC benchmark. This benchmark achieved the lowest descriptive mean reference-tracking RMSE and total objective. Mean PG-NODE-MPC optimization time was 0.86 s. Results suggest the potential for low-frequency building management system supervisory decision support, subject to physical command mapping and staged field validation.

Förderung

  • Gansu Provincial Key Laboratory of Solar Energy Utilization

Zitieren

APA

Wang, Y., Chang, L., Zhang, E., Jia, K., Zhang, X., & Li, Y. (2026). Physics-Guided Neural ODEs for Building Thermal Prediction and Model Predictive Control. Buildings, 16(15), 3021. https://doi.org/10.3390/buildings16153021

Vancouver

Wang Y, Chang L, Zhang E, Jia K, Zhang X, Li Y. Physics-Guided Neural ODEs for Building Thermal Prediction and Model Predictive Control. Buildings. 2026;16(15):3021. doi:10.3390/buildings16153021

BibTeX
@article{Wang2026,
  author = {Wang, Yagang and Chang, Lexue and Zhang, Enzhan and Jia, Kejun and Zhang, Xia and Li, Yonghao},
  title = {Physics-Guided Neural ODEs for Building Thermal Prediction and Model Predictive Control},
  journal = {Buildings},
  year = {2026},
  volume = {16},
  number = {15},
  pages = {3021},
  doi = {10.3390/buildings16153021}
}
Kostenlos starten →