Reproduce “FG-ConvNO: A Geometry-Aware Neural Operator for Propeller CFD Prediction” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yichen Di, Valentin Duruisseaux, Di Zhou, Xinyi Li, Daniel Leibovici, Jean Kossaifi, Anima Anandkumar
FG-ConvNO predicts propeller-surface pressure and global performance from geometry and operating conditions using geometry-aware features, factorized grids, and interpolation-based decoding. On a procedural dataset of steady-state propeller flows, its surface-only model reports 23.68% relative L2 error, versus 29.16% for GINO. Joint supervision reports 3.43 × 10⁻⁴ global loss but slightly worsens surface error; results do not establish runtime gains or broader generalization.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.