Application of Video-to-Video Translation Networks to Computational Fluid Dynamics

TL;DR

This paper demonstrates that video-to-video translation GANs, combining pix2pix and LSTM, can accurately estimate high-resolution CFD simulation results from low-resolution data, reducing computational costs.

Contribution

It introduces a novel GAN architecture integrating pix2pix and LSTM for efficient CFD simulation approximation from low-resolution inputs.

Findings

High-resolution density distributions are accurately reproduced from low-resolution data.

GAN-based method outperforms several super-resolution algorithms in CFD applications.

The approach significantly reduces computational costs of CFD simulations.

Abstract

In recent years, the evolution of artificial intelligence, especially deep learning, has been remarkable, and its application to various fields has been growing rapidly. In this paper, I report the results of the application of generative adversarial networks (GANs), specifically video-to-video translation networks, to computational fluid dynamics (CFD) simulations. The purpose of this research is to reduce the computational cost of CFD simulations with GANs. The architecture of GANs in this research is a combination of the image-to-image translation networks (the so-called "pix2pix") and Long Short-Term Memory (LSTM). It is shown that the results of high-cost and high-accuracy simulations (with high-resolution computational grids) can be estimated from those of low-cost and low-accuracy simulations (with low-resolution grids). In particular, the time evolution of density distributions in the cases of a high-resolution grid is reproduced from that in the cases of a low-resolution grid through GANs, and the density inhomogeneity estimated from the image generated by GANs recovers the ground truth with good accuracy. Qualitative and quantitative comparisons of the results of the proposed method with those of several super-resolution algorithms are also presented.