Physics-Informed Neural Networks for Predicting Hydration Degree in Cementitious Systems Blended with Slag and Fly Ash
Materials · 10.07.2026 · 19 (14) · 2978
DOI: 10.3390/ma19142978 ↗Kurz gefasst
Fachartikel in der Zeitschrift Materials, online erschienen am 10.07.2026. Verfasst von Xiaoyi Hu und 4 weiteren Autor:innen. Beteiligte Einrichtung: Key Laboratory of Geological Hazards on Three Gorges Reservoir Area, Ministry of Education, China Three Gorges University, Yichang 443002, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Chongqing Municipality Education Commission, Chongqing Municipal Education Commission.
Abstract (Originalsprache)
Accurately predicting the degree of hydration in cement systems blended with slag and fly ash is difficult when experimental data are limited. Empirical kinetic models typically need recalibration for each specific blend, while purely data-driven approaches often fail to maintain physical consistency. In this work, we present a physics-informed neural network for predicting isothermal hydration by incorporating the Avrami–Erofeev–Arrhenius ordinary differential equation into the training loss and using a composition-focused sub-network to relate blend proportions to blend-specific kinetic parameters. The model was trained and tested on 29,379 calorimetry data points from 77 multi-component cement systems gathered from two open-access datasets. For previously unseen test systems, it reached an R2 of 0.9864 and a symmetric mean absolute percentage error (sMAPE) of 25.3%. Its overall sMAPE was lower than that of a neural network with the same architecture but without physics-based constraints (34.9%); stage-resolved analysis showed that the largest difference occurred during early hydration. Across eight random seeds, the physics constraint did not confer a data-efficiency or training-stability advantage over the same-architecture network. The distinguishing features of the PINN are instead its explicit ODE and monotonicity regularization and its composition-conditioned effective parameters. Although random forest regression produced lower pointwise error, it failed to maintain physically consistent hydration-rate behavior. Overall, the framework provides a composition-conditioned surrogate that incorporates kinetic regularization within the composition range investigated.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Key Laboratory of Geological Hazards on Three Gorges Reservoir Area, Ministry of Education, China Three Gorges University, Yichang 443002, China.
- School of Urban Construction Engineering, Chongqing Technology and Business Institute, Chongqing 400085, China.
- College of Civil Engineering & Architecture, China Three Gorges University, Yichang 443002, China.
- Chongqing Zhuyun Science and Technology Limited Corporation, Chongqing 400045, China.
- School of Civil Engineering, Chongqing University, Chongqing 400045, China.
- State Key Laboratory of Safety and Resilience of Civil Engineering in Mountain Area, Chongqing University, Chongqing 400045, China.
ORCID-Profile: Xiaofeng Liao
Förderung
- Chongqing Municipality Education Commission
- Chongqing Municipal Education Commission
Zitieren
APA
Hu, X., Liao, X., Liao, Z., Wang, Z., & Gan, M. (2026). Physics-Informed Neural Networks for Predicting Hydration Degree in Cementitious Systems Blended with Slag and Fly Ash. Materials, 19(14), 2978. https://doi.org/10.3390/ma19142978
Vancouver
Hu X, Liao X, Liao Z, Wang Z, Gan M. Physics-Informed Neural Networks for Predicting Hydration Degree in Cementitious Systems Blended with Slag and Fly Ash. Materials. 2026;19(14):2978. doi:10.3390/ma19142978
BibTeX
@article{Hu2026,
author = {Hu, Xiaoyi and Liao, Xiaofeng and Liao, Zaiyi and Wang, Zhizhi and Gan, Min},
title = {Physics-Informed Neural Networks for Predicting Hydration Degree in Cementitious Systems Blended with Slag and Fly Ash},
journal = {Materials},
year = {2026},
volume = {19},
number = {14},
pages = {2978},
doi = {10.3390/ma19142978}
}
EUVIMED