SuperWing: a comprehensive transonic wing dataset for data-driven aerodynamic design
Yunjia Yang, Weishao Tang, Mengxin Liu, Nils Thuerey, Yufei Zhang, Haixin Chen · 2026
Read SuperWing: a comprehensive transonic wing dataset for data-driven aerodynamic design by Yunjia Yang, Weishao Tang, Mengxin Liu, Nils Thuerey, Yufei Zhang, Haixin Chen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13427721.txt.
Description
Yang and colleagues release SuperWing in 2026, an open dataset built to train machine learning models that predict airflow over three-dimensional aircraft wings. It holds 4,239 wing shapes and 28,856 detailed flow simulations computed with standard engineering solvers. Shapes vary along the span in airfoil profile, twist, and upward angle, rather than being small tweaks of one baseline wing. Conditions span the normal flight range of speeds and attack angles.
Two state-of-the-art transformer models trained on the data predicted surface flow accurately, missing drag by only two and a half drag counts on unseen wings. Models pre-trained on SuperWing also transferred without retraining to complex standard test wings used by German and NASA benchmarks. The authors argue such diverse shared data is the missing ingredient for general-purpose aerodynamic design tools that can answer in seconds instead of hours of simulation.
Catalog record
| CREATOR | Yunjia Yang, Weishao Tang, Mengxin Liu, Nils Thuerey, Yufei Zhang, Haixin Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |