Simulation of Steady and Transient 3D Flows via Physics-Informed Deep Learning

Philipp Moser, Wolfgang Fenz, Stefan Thumfart, Isabell Ganitzer, Michael Giretzlehner

2023

Abstract

Physics-Informed deep learning methods are attracting increased attention for modeling physical systems due to their mesh-free approach, their straightforward handling of forward and inverse problems, and the possibility to seamlessly include measurement data. Today, most learning-based flow modeling reports rely on the representational power of fully-connected neural networks, although many different architectures have been introduced into deep learning, each with specific benefits for certain applications. In this paper, we successfully demonstrate the application of physics-informed neural networks for modeling steady and transient flows through 3D geometries. Our work serves as a practical guideline for machine learning practitioners by comparing several popular network architectures in terms of accuracy and computational costs. The steady flow results were in good agreement with finite element-based simulations, while the transient flows proved more challenging for the continuous-time PINN approaches. Overall, our findings suggest that standard fully-connected neural networks offer an efficient balance between training time and accuracy. Although not readily supported by statistical/practical significance, we could identify a few more complex architectures, namely Fourier networks and Deep Galerkin Methods, as attractive options for accurate flow modeling.

Paper Citation

in Harvard Style

Moser P., Fenz W., Thumfart S., Ganitzer I. and Giretzlehner M. (2023). Simulation of Steady and Transient 3D Flows via Physics-Informed Deep Learning. In Proceedings of the 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications - Volume 1: SIMULTECH; ISBN 978-989-758-668-2, SciTePress, pages 243-250. DOI: 10.5220/0012078600003546

in Bibtex Style

@conference{simultech23,
author={Philipp Moser and Wolfgang Fenz and Stefan Thumfart and Isabell Ganitzer and Michael Giretzlehner},
title={Simulation of Steady and Transient 3D Flows via Physics-Informed Deep Learning},
booktitle={Proceedings of the 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications - Volume 1: SIMULTECH},
year={2023},
pages={243-250},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012078600003546},
isbn={978-989-758-668-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Simulation and Modeling Methodologies, Technologies and Applications - Volume 1: SIMULTECH
TI - Simulation of Steady and Transient 3D Flows via Physics-Informed Deep Learning
SN - 978-989-758-668-2
AU - Moser P.
AU - Fenz W.
AU - Thumfart S.
AU - Ganitzer I.
AU - Giretzlehner M.
PY - 2023
SP - 243
EP - 250
DO - 10.5220/0012078600003546
PB - SciTePress