Towards prediction of turbulent flows at high Reynolds numbers using high performance computing data and deep learning

TL;DR

This paper evaluates deep learning, specifically Wasserstein GANs, for modeling small-scale turbulence using high-resolution DNS data, demonstrating promising qualitative and quantitative results.

Contribution

It introduces the application of WGANs to generate turbulent flow structures and analyzes the impact of network parameters on modeling accuracy.

Findings

WGANs can generate turbulence structures with good qualitative agreement.

Quantitative assessments show promising statistical similarity to DNS data.

Network parameter tuning affects the quality of turbulence generation.

Abstract

In this paper, deep learning (DL) methods are evaluated in the context of turbulent flows. Various generative adversarial networks (GANs) are discussed with respect to their suitability for understanding and modeling turbulence. Wasserstein GANs (WGANs) are then chosen to generate small-scale turbulence. Highly resolved direct numerical simulation (DNS) turbulent data is used for training the WGANs and the effect of network parameters, such as learning rate and loss function, is studied. Qualitatively good agreement between DNS input data and generated turbulent structures is shown. A quantitative statistical assessment of the predicted turbulent fields is performed.