← Papers

Airfoil RANS CNN

CNN prediction of RANS flow fields around airfoils

Paper title Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networks

Bhatnagar et al. (University of Michigan, General Motors) · Aero & CFD

Paper arXiv 1905.13166PDF

An encoder-decoder CNN takes the airfoil as a signed distance function on a 150 by 150 grid, with the Reynolds number and angle of attack, and outputs the velocity and pressure fields of a RANS solution on the same grid. A single shared decoder for all three fields replaces separate decoders, and a gradient-sharpening term in the loss reduces blurring in the wake.

Published
Computational Mechanics

Share on LinkedIn, X or Bluesky, or or .

Know related work that belongs here?

Something wrong on this page?

More papers

All papers
  1. PhysicsNeMo-Mesh — GPU-Native Meshes and Simulation Cases for Physics AI
  2. Multi-fidelity scaling — Compute-optimal data scaling via multi-fidelity training
  3. CAE-UQ — Predictive Uncertainty for Neural CAE Surrogates