Skip to main content
11th International Symposium on Turbulence and Shear Flow Phenomena, 2019

Full Program »
paper
View File
pdf
7.5MB
one page
View File
pdf
24KB

SUPER-RESOLUTION ANALYSIS WITH MACHINE LEARNING FOR LOW-RESOLUTION TURBULENT FLOW DATA

We present machine learning-based super-resolution techniques to reconstruct subgrid-scale turbulent flow fields from low-resolution (LR) data. Two-dimensional turbulent flow data obtained from direct numerical simulation is con- sidered, with which LR data sets are generated by max- pooling operation. We construct machine learning (ML) models based on the convolutional neural networks (CNN) to predict high-resolution (HR) flow fields from LR data. Resolved flow field by our machine learned method called the Downsampled Skip-Connection Multi-Scale (DSC-MS) model shows agreement between the reconstructed flow field and the reference solution in terms of its flow field and kinetic energy cascade. Towards the end of the paper, we offer discussions on the applicability of the current ap- proach for general flow field data obtained from numerical simulations and experimental measurements.

Kai Fukami
Keio Univeristy / Florida State University / University of California, Los Angeles
United States

Koji Fukagata
Keio University
Japan

Kunihiko Taira
University of California, Los Angeles / Florida State University
United States

 


Powered by OpenConf®
Copyright ©2002-2018 Zakon Group LLC