Flow reconstruction in time-varying geometries using graph neural networks

TL;DR

This paper introduces a Graph Attention Convolutional Network (GACN) for reconstructing flow fields from sparse, time-varying data, demonstrating superior performance over traditional methods on DNS and PIV datasets, especially for larger domains.

Contribution

The paper develops a novel GACN model with feature propagation and validity masking, enabling accurate flow reconstruction from extremely sparse, unstructured, and variable-sized data in dynamic geometries.

Findings

GACN outperforms CNN and cubic interpolation in reconstruction accuracy.

The model effectively handles unstructured data and larger domains.

GACN maintains robustness across different resolutions and unseen datasets.

Abstract

The paper presents a Graph Attention Convolutional Network (GACN) for flow reconstruction from very sparse data in time-varying geometries. The model incorporates a feature propagation algorithm as a preprocessing step to handle extremely sparse inputs, leveraging information from neighboring nodes to initialize missing features. In addition, a binary indicator is introduced as a validity mask to distinguish between the original and propagated data points, enabling more effective learning from sparse inputs. Trained on a unique data set of Direct Numerical Simulations (DNS) of a motored engine at a technically relevant operating condition, the GACN shows robust performance across different resolutions and domain sizes and can effectively handle unstructured data and variable input sizes. The model is tested on previously unseen DNS data as well as on an experimental data set from Particle Image Velocimetry (PIV) measurements that were not considered during training. A comparative analysis shows that the GACN consistently outperforms both a conventional Convolutional Neural Network (CNN) and cubic interpolation methods on the DNS and PIV test sets by achieving lower reconstruction errors and better capturing fine-scale turbulent structures. In particular, the GACN effectively reconstructs flow fields from domains up to 14 times larger than those observed during training, with the performance advantage increasing for larger domains.