Physics-constrained machine learning for reduced composition space chemical kinetics
Data-Centric Engineering · 2025 · 6
DOI: 10.1017/dce.2025.10012 ↗Kurz gefasst
Fachartikel in der Zeitschrift Data-Centric Engineering, online erschienen am 10.07.2025. Verfasst von Anuj Kumar und 1 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0). Gefördert von: Aramco Americas.
Abstract (Originalsprache)
Modeling detailed chemical kinetics is a primary challenge in combustion simulations. We present a novel framework to enforce physical constraints, specifically total mass and elemental conservation, during the reaction of ML models’ training for the reduced composition space chemical kinetics of large chemical mechanisms in combustion. In these models, the transport equations for a subset of representative species are solved with the ML approaches, while the remaining nonrepresentative species are “recovered” with a separate artificial neural network trained on data. Given the strong correlation between full and reduced solution vectors, our method utilizes a small neural network to establish an accurate and physically consistent mapping. By leveraging this mapping, we enforce physical constraints in the training process of the ML model for reduced composition space chemical kinetics. The framework is demonstrated here for methane, CH 4 , and oxidation. The resulting solution vectors from our deep operator networks (DeepONet)-based approach are accurate and align more consistently with physical laws.
Förderung
- Aramco Americas
Zitieren
APA
Kumar, A., & Echekki, T. (2025). Physics-constrained machine learning for reduced composition space chemical kinetics. Data-Centric Engineering, 6. https://doi.org/10.1017/dce.2025.10012
Vancouver
Kumar A, Echekki T. Physics-constrained machine learning for reduced composition space chemical kinetics. Data-Centric Engineering. 2025;6. doi:10.1017/dce.2025.10012
BibTeX
@article{Kumar2025,
author = {Kumar, Anuj and Echekki, Tarek},
title = {Physics-constrained machine learning for reduced composition space chemical kinetics},
journal = {Data-Centric Engineering},
year = {2025},
volume = {6},
doi = {10.1017/dce.2025.10012}
}
EUVIMED