Influence of adversarial training on super-resolution turbulence reconstruction

TL;DR

This paper demonstrates that adversarial training with GANs improves the super-resolution reconstruction of turbulent flows, especially for out-of-sample data, by enhancing small-scale feature recovery and robustness.

Contribution

It provides a comprehensive assessment of GAN-based adversarial training for turbulence super-resolution, highlighting its advantages over supervised CNNs and proposing methods to improve out-of-sample performance.

Findings

GANs outperform CNNs in in-sample turbulence reconstruction

Discriminator training enhances out-of-sample robustness

GANs better reconstruct small-scale turbulence features

Abstract

Supervised super-resolution deep convolutional neural networks (CNNs) have gained significant attention for their potential in reconstructing velocity and scalar fields in turbulent flows. Despite their popularity, CNNs currently lack the ability to accurately produce high-frequency and small-scale features, and tests of their generalizability to out-of-sample flows are not widespread. Generative adversarial networks (GANs), which consist of two distinct neural networks (NNs), a generator and discriminator, are a promising alternative, allowing for both semi-supervised and unsupervised training. The difference in the flow fields produced by these two NN architectures has not been thoroughly investigated, and a comprehensive understanding of the discriminator's role has yet to be developed. This study assesses the effectiveness of the unsupervised adversarial training in GANs for turbulence reconstruction in forced homogeneous isotropic turbulence. GAN-based architectures are found to outperform supervised CNNs for turbulent flow reconstruction for in-sample cases. The reconstruction accuracy of both architectures diminishes for out-of-sample cases, though the GAN's discriminator network significantly improves the generator's out-of-sample robustness using either an additional unsupervised training step with large eddy simulation input fields and a dynamic selection of the most suitable upsampling factor. These enhance the generator's ability to reconstruct small-scale gradients, turbulence intermittency, and velocity-gradient probability density functions. The extrapolation capability of the GAN-based model is demonstrated for out-of-sample flows at higher Reynolds numbers. Based on these findings, incorporating discriminator-based training is recommended to enhance the reconstruction capability of super-resolution CNNs.