AirfRANS: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier–Stokes Solutions

Florent Bonnet, Jocelyn Ahmed Mazari, Paola Cinnella, patrick gallinari

Keywords: Computational Fluid Dynamics, Navier–Stokes Equations, Partial Differential Equations, Physical Metrics, Geometric Deep Learning, Graph Neural Networks, Point Clouds, Surrogate Models, Reduced Order Models, Meshes, Physically Constrained Deep Learning, Numerical Simulation, Fluid Mechanics

TL;DR: We propose a high fidelity aerodynamic dataset of Reynolds-Averaged Navier–Stokes (RANS) simulations over airfoils

Abstract: Surrogate models are necessary to optimize meaningful quantities in physical dynamics as their recursive numerical resolutions are often prohibitively expensive. It is mainly the case for fluid dynamics and the resolution of Navier–Stokes equations. However, despite the fast-growing field of data-driven models for physical systems, reference datasets representing real-world phenomena are lacking. In this work, we develop \textsc{AirfRANS}, a dataset for studying the two-dimensional incompressible steady-state Reynolds-Averaged Navier–Stokes equations over airfoils at a subsonic regime and for different angles of attacks. We also introduce metrics on the stress forces at the surface of geometries and visualization of boundary layers to assess the capabilities of models to accurately predict the meaningful information of the problem. Finally, we propose deep learning baselines on four machine learning tasks to study \textsc{AirfRANS} under different constraints for generalization considerations: big and scarce data regime, Reynolds number, and angle of attack extrapolation.

License:

Open Data Commons Open Database License v1.0. The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use a database while maintaining this same freedom for others.
https://opendatacommons.org/licenses/odbl/1-0/

Author Statement: Yes

Dataset Url:

Root folder: https://data.isir.upmc.fr/extrality/NeurIPS_2022/
Machine learning ready dataset: https://data.isir.upmc.fr/extrality/NeurIPS_2022/Dataset.zip
Full OpenFOAM dataset: https://data.isir.upmc.fr/extrality/NeurIPS_2022/OF_dataset.zip
Scores and visualizations for the proposed models: https://data.isir.upmc.fr/extrality/NeurIPS_2022/scores.zip

AirfRANS Python library: https://airfrans.readthedocs.io/en/latest/index.html
GitHub repository to reproduce the experiments: https://github.com/Extrality/AirfRANS
GitHub repository to run new simulations: https://github.com/Extrality/NACA_simulation
GitHub repository for AirfRANS library: https://github.com/Extrality/airfrans_lib

URL: https://data.isir.upmc.fr/extrality/NeurIPS_2022/Dataset.zip

Contribution Process Agreement: Yes

In Person Attendance: Yes

Community Implementations: [ 3 code implementations](https://www.catalyzex.com/paper/airfrans-high-fidelity-computational-fluid/code)