Physics Informed Neural Network for Multi-dimensional Turbulent Self-gravitating Flows
Abstract
Molecular clouds exhibit complex, multi-scale dynamics driven by the interplay of turbulence, self-gravity, and gas pressure. Modelling these systems with conventional grid-based methods becomes increasingly expensive as spatial resolution and dimensionality increase. Physics-Informed Neural Networks (PINNs) offer a mesh-free alternative by incorporating the governing equations directly into the training objective, but their ability to model realistic turbulent self-gravitating flows remains largely unexplored. We apply PINNs to isothermal molecular clouds initialized with astrophysically motivated turbulent velocity fields, considering subsonic and supersonic evolution in two and three dimensions. We benchmark the predicted density and velocity fields against the finite-volume code Athena++. PINNs accurately recover the global evolution of the flow over multiple free-fall times across turbulent realizations, Mach numbers, and dimensionalities. Their training cost also exhibits weak dependence on spatial resolution, increasing by less than 25% from 128^3 to 1024^3 in our experiments. These results demonstrate that PINNs can model substantially more realistic self-gravitating turbulent flows than previously explored and provide a promising complementary framework for studying these systems.
Related
- Cosmo-PINN: A Physics-Informed Neural Network for Cosmological Reconstruction
- A Physics Informed Neural Network Approach to Decipher Environmental Confinement of Molecular Clouds in the Third Galactic Quadrant
- Metaheuristic physics-informed extreme-learning machine for initial orbit determination of near-Earth asteroids from too-short arcs
- Physics-Informed Multi-Task Surrogate Model for the Martian Nightside Thermosphere
- Reconstructing the generalized Barrow holographic dark energy with physics-informed neural networks
BibTeX
@misc{de2026physics,
title = {{Physics Informed Neural Network for Multi-dimensional Turbulent Self-gravitating Flows}},
author = {Tirthankar De and Sayantan Auddy and Ramit Dey and Shantanu Basu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.32840},
eprint = {2610.32840},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.32840},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Physics Informed Neural Network for Multi-dimensional Turbulent Self-gravitating Flows
AU - Tirthankar De
AU - Sayantan Auddy
AU - Ramit Dey
AU - Shantanu Basu
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.32840
UR - https://arxiv.org/abs/2610.32840
AB - Molecular clouds exhibit complex, multi-scale dynamics driven by the interplay of turbulence, self-gravity, and gas pressure. Modelling these systems with conventional grid-based methods becomes increasingly expensive as spatial resolution and dimensionality increase. Physics-Informed Neural Networks (PINNs) offer a mesh-free alternative by incorporating the governing equations directly into the training objective, but their ability to model realistic turbulent self-gravitating flows remains largely unexplored. We apply PINNs to isothermal molecular clouds initialized with astrophysically motivated turbulent velocity fields, considering subsonic and supersonic evolution in two and three dimensions. We benchmark the predicted density and velocity fields against the finite-volume code Athena++. PINNs accurately recover the global evolution of the flow over multiple free-fall times across turbulent realizations, Mach numbers, and dimensionalities. Their training cost also exhibits weak dependence on spatial resolution, increasing by less than 25% from 128^3 to 1024^3 in our experiments. These results demonstrate that PINNs can model substantially more realistic self-gravitating turbulent flows than previously explored and provide a promising complementary framework for studying these systems.
ER -