Flow-Field Prediction in Periodic Domains Using a Convolution Neural Network with Hypernetwork Parametrization

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F23%3A43968210" target="_blank" >RIV/49777513:23520/23:43968210 - isvavai.cz</a>

Result on the web

<a href="https://www.worldscientific.com/doi/abs/10.1142/S1758825123500187" target="_blank" >https://www.worldscientific.com/doi/abs/10.1142/S1758825123500187</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1142/S1758825123500187" target="_blank" >10.1142/S1758825123500187</a>

Result language

angličtina

Original language name

Flow-Field Prediction in Periodic Domains Using a Convolution Neural Network with Hypernetwork Parametrization

Original language description

This paper deals with flow field prediction in a blade cascade using the convolution neural network. The convolutional neural network (CNN) predicts density, pressure and velocity fields based on the given geometry. The blade cascade is modeled as a single interblade channel with periodic boundary conditions. In this paper, an algorithm that enforces periodic boundary conditions onto the CNN is presented. The main target of this study is to parametrize the CNN model depending on the Reynolds number. A new parametrization approach based on a so-called hypernetwork is employed for this purpose. The idea of this approach is that when the Reynolds number is modified, the hypernetwork modifies the weights of the CNN in such a way that it produces flow fields corresponding to that particular Reynolds number. The concept of the hypernetwork-based parametrization is tested on the problem of a compressible fluid flow through a blade cascade with variable blade profiles and Reynolds numbers.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

20302 - Applied mechanics

Project

<a href="/en/project/GA21-31457S" target="_blank" >GA21-31457S: Fast flow-field prediction using deep neural networks for solving fluid-structure interaction problems</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2023

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Name of the periodical

International Journal of Applied Mechanics

ISSN

1758-8251

e-ISSN

1758-826X

Volume of the periodical

15

Issue of the periodical within the volume

2

Country of publishing house

SG - SINGAPORE

Number of pages

20

Pages from-to

1-20

UT code for WoS article

000923377600001

EID of the result in the Scopus database

2-s2.0-85147495139