Geometry aware neural model improves propeller flow prediction with limited data
FG-ConvNO: A Geometry-Aware Neural Operator for Propeller CFD Prediction
Machine Learning
Summary
Simulating airflow around complex objects like propellers is often very slow and costly. The authors present a new machine learning method that understands the shape of the propeller better, allowing it to predict pressure and thrust with less data and fewer errors. Their approach uses special math tricks to handle irregular shapes and data points, making predictions more accurate and stable. This helps engineers get faster results without running expensive simulations every time.
What this means in practice
- •For aerospace engineers: Speed up propeller design by accurately predicting flow and thrust with fewer CFD simulations and limited training data.
- •For marine propulsion designers: Predict pressure distributions on ship propellers more quickly to optimize designs using geometry-aware neural networks.
Authors
Yichen Di, Valentin Duruisseaux, Di Zhou, Xinyi Li, Daniel Leibovici, Jean Kossaifi, Anima Anandkumar
Abstract
Computational Fluid Dynamics (CFD) is essential in contemporary engineering design, yet it remains expensive to compute and can take thousands of CPU/GPU hours or more for a single high-fidelity simulation. Recent developments in machine learning for scientific modeling have led to powerful surrogate frameworks that can approximate flow fields and physical quantities by using a fraction of the computational cost. However, most of these methods require largescale homogeneous datasets or simple geometries, exhibiting severe generalization gaps when applied to complex and data-scarce configurations. To address this, we propose the Factorized Geometry Convolutional Neural Operator (FG-ConvNO) for robust geometry-aware propeller pressure prediction under limited training data. We introduce geometryaware encoding, efficient interpolation-based decoding, and auxiliary global quantity supervision. In addition, we incorporate DISCO blocks that use discrete continuous convolutions to make the model discretization agnostic. In our experiments, FG-ConvNO achieves the highest accuracy for both surface pressure and global thrust prediction on a procedurally generated propeller CFD dataset. Geometry-aware encoding and architectural simplification improve stability, while random vertex subsampling enhances robustness. Overall, adding geometric priors to grid-based convolutional operators reduces overfitting on complex propeller geometries and improves generalization in low-data regimes.