Login
- Document type:
- Forschungsdaten
- Publication date:
- 29.09.2023
- Responsible:
- Schlichter, Philipp
- Authors:
- Widmann, Sebastian; Schlichter, Philipp; Reck, Michaela; Indinger, Thomas
- Author affiliation:
- TUM
- Publisher:
- TUM
- Title:
- AirfoilMNIST: A Large-Scale Dataset based on Two-Dimensional RANS Simulations of Airfoils
- Identifier:
- doi:10.14459/2023mp1712519
- End date of data production:
- 31.05.2023
- Subject area:
- DAT Datenverarbeitung, Informatik; TEC Technik, Ingenieurwissenschaften (allgemein)
- Resource type:
- Simulationen / simulations
- Data type:
- Bilder / images; Datenbanken / data bases
- Description:
- Studying aerodynamic problems is still dominated by computationally heavy simulations which require substantial resources and knowledge about aerodynamics and numerical methods. While first adaptions of artificial intelligence have reached the field of aerodynamics to accelerate this process, data is still scarce and often lacks the proper validation. To bridge this gap, we propose airfoilMNIST, a comprehensive dataset of two-dimensional, RANS-based flow fields for NACA 4- and 5-series airfoils. Our dataset consists of around 150’000 samples for Mach numbers up to 0.6 and angles of attack −5 <= alpha <= 15 where each sample contains the mean flow fields for density, eddy viscosity, pressure, temperature, turbulent kinetic energy, turbulent thermal diffusivity, specific rate of dissipation and velocity.
The dataset is separated into three subsets:
- airfoilMNIST-raw: Individual CFD samples which are stored in the VTU file format together with all airfoil geometries as STL files. Additionally, each sample has an associated text file with the progression of the force and moment coefficients for the respective iteration number of the CFD simulation. The entire subset has a total size of 3.5TB.
- airfoilMNIST: Preprocessed dataset compatible with the Tensorflow Data API. Flow fields have been resampled onto a uniform grid and stored as two NumPy arrays, designed for encoder-decoder architectures. Encoder array is a field with the initial and boundary conditions while the decoder array are the individual flow fields. Dataset has a predefined, randomised train-test split of (80, 20)% and a total size of 500GB.
- airfoilMNIST-incompressible: Further simplification of the airfoilMNIST subset. Only incompressible simulations (Mach < 0.3) are stored. Instead of storing all flow fields, only the pressure and velocity fields normalised with the freestream conditions are saved. Dataset has a predefined, randomised train-test split of (80, 20)% and a total size of 50GB. - Method of data assessment:
- The data was acquired using open-source tools, namely Python and OpenFOAM. No propiertary software was used for the data acquisition. To generate the CFD data, the server capacities from the chair of aerodynamics and fluid dynamics were used. The simulations were ran on 15 server modules with an average of 12 cores / module and a RAM size of 128GB / module. To postprocess the simulations, a NVIDIA RTX3080Ti GPU was used.
The data was generated during the research thesis of Sebastian Widmann as... » - Links:
Link to code implementation: https://github.com/sebastianwidmann/nacaFOAM
- Key words:
- dataset, airfoilMNIST, CFD, airfoil, RANS, compressible, incompressible, neural networks, aerodynamics
- Technical remarks:
- Representative Dataset: View and download (47 GB total, 81 Files)
The data server also offers downloads with FTP
The data server also offers downloads with rsync (password m1712519.rep):
rsync rsync://m1712519.rep@dataserv.ub.tum.de/m1712519.rep /
Entire Dateset: View and download (3,34 TB total, 311.986 Files)
The data server also offers downloads with FTP
The data server also offers downloads with rsync (password m1712519):
rsync rsync://m1712519@dataserv.ub.tum.de/m1712519/ - Language:
- en
- Rights:
- by-sa, http://creativecommons.org/licenses/by-sa/4.0
BibTeX