Datasets for "Enhanced physics-informed neural networks with adaptive weighting for incompressible fluid flow modeling"

Published: 4 November 2026| Version 1 | DOI: 10.17632/3wpkb6ybzw.1
Contributor:
Pan Qin

Description

This dataset contains numerical data used for training and evaluating the models in our research. It covers three classic fluid mechanics benchmark cases: 2D cavity flow, flow around cylinder, and flow around an oscillating airfoil. Cavity Flow & Flow Around Cylinder: The datasets for these two cases were collected from established literature. They serve as reference cases for steady and unsteady flow evaluation. Oscillating Airfoil (Re = 1000): The dataset for the moving-boundary oscillating airfoil was generated via numerical simulations using OpenFOAM. The setup was modified based on the standard wingMotion2D_pimpleFoam tutorial provided in official OpenFOAM distributions. Data Organization: The uploaded files are structured into separate folders corresponding to each benchmark case: 01_Cavity_Flow/: Velocity and pressure field data for cavity flow. 02_Cylinder_Flow/: Flow field data for flow around cylinder. 03_Airfoil/: Processed flow field data for flow around airfoil.

Files

Institutions

Categories

Fluid Flow, Deep Learning

Licence