Deep Neural Networks for Modeling Astrophysical Nuclear Reacting Flows
Open Access
- 2 September 2025
- journal article
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 990 (2) , 105
Abstract
In astrophysical simulations, nuclear reacting flows pose computational challenges due to the stiffness of reaction networks. We introduce neural network-based surrogate models using the DeePODE framework to enhance simulation efficiency while maintaining accuracy and robustness. Our method replaces conventional stiff ordinary differential equation (ODE) solvers with deep learning models trained through evolutionary Monte Carlo sampling from zero-dimensional simulation data, ensuring generalization across varied thermonuclear and hydrodynamic conditions. Tested on 3-species and 13-species reaction networks, the models achieve ≲1% accuracy relative to semi-implicit numerical solutions and deliver a ∼2.6× speedup on CPUs. A temperature-thresholded deployment strategy ensures stability in extreme conditions, sustaining neural network utilization above 75% in multidimensional simulations. These data-driven surrogates effectively mitigate stiffness constraints, offering a scalable approach for high-fidelity modeling of astrophysical nuclear reacting flows.Keywords
All Related Versions
Funding Information
- MOST ∣ National Natural Science Foundation of China (92270001)
- MOST ∣ National Natural Science Foundation of China (12371511)
- MOST ∣ National Natural Science Foundation of China (12422119)
- MOST ∣ National Natural Science Foundation of China (92470127)
- MOST ∣ National Natural Science Foundation of China (12305246)
This publication has 35 references indexed in Scilit:
- Enforcing physical conservation in neural network surrogate models for complex chemical kineticsCombustion and Flame, 2025
- A multi-scale sampling method for accurate and robust deep neural network to predict combustion chemical kineticsCombustion and Flame, 2022
- Combustion machine learning: Principles, progress and prospectsProgress in Energy and Combustion Science, 2022
- Double detonations of sub-MCh CO white dwarfs: variations in Type Ia supernovae due to different core and He shell massesAstronomy & Astrophysics, 2021
- Kelvin–Helmholtz instabilities as the source of inhomogeneous mixing in nova explosionsNature, 2011
- CASTRO: A NEW COMPRESSIBLE ASTROPHYSICAL SOLVER. I. HYDRODYNAMICS AND SELF-GRAVITYThe Astrophysical Journal, 2010
- Three‐dimensional Delayed‐Detonation Model of Type Ia SupernovaeThe Astrophysical Journal, 2005
- FLASH: An Adaptive Mesh Hydrodynamics Code for Modeling Astrophysical Thermonuclear FlashesThe Astrophysical Journal Supplement Series, 2000
- Integration of Nuclear Reaction Networks for Stellar HydrodynamicsThe Astrophysical Journal Supplement Series, 1999
- Neutrino Energy Loss in Stellar Interiors. VII. Pair, Photo-, Plasma, Bremsstrahlung, and Recombination Neutrino ProcessesThe Astrophysical Journal Supplement Series, 1996