Convolutional Autoencoder for the Spatiotemporal Latent Representation of Turbulence

Nguyen Anh Khoa Doan, Alberto Racca, Luca Magri

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

18 Downloads (Pure)

Abstract

Turbulence is characterised by chaotic dynamics and a high-dimensional state space, which make this phenomenon challenging to predict. However, turbulent flows are often characterised by coherent spatiotemporal structures, such as vortices or large-scale modes, which can help obtain a latent description of turbulent flows. However, current approaches are often limited by either the need to use some form of thresholding on quantities defining the isosurfaces to which the flow structures are associated or the linearity of traditional modal flow decomposition approaches, such as those based on proper orthogonal decomposition. This problem is exacerbated in flows that exhibit extreme events, which are rare and sudden changes in a turbulent state. The goal of this paper is to obtain an efficient and accurate reduced-order latent representation of a turbulent flow that exhibits extreme events. Specifically, we employ a three-dimensional multiscale convolutional autoencoder (CAE) to obtain such latent representation. We apply it to a three-dimensional turbulent flow. We show that the Multiscale CAE is efficient, requiring less than 10% degrees of freedom than proper orthogonal decomposition for compressing the data and is able to accurately reconstruct flow states related to extreme events. The proposed deep learning architecture opens opportunities for nonlinear reduced-order modeling of turbulent flows from data.
Original languageEnglish
Title of host publicationLecture Notes in Computer Science - ICCS 2023
Subtitle of host publicationComputational Science – ICCS 2023 23rd International Conference, Prague, Czech Republic, July 3–5, 2023, Proceedings, Part IV
EditorsDmitry V. Kozyrev
PublisherSpringer
Pages328-335
Volume10476
ISBN (Electronic)978-3-031-50482-2
DOIs
Publication statusPublished - 2023
Event23rd International Conference on Computational Science, ICCS 2023 - Prague, Czech Republic
Duration: 3 Jul 20235 Jul 2023

Publication series

NameLecture Notes in Computer Science - ICCS2023
PublisherSpringer
Volume10476
ISSN (Electronic)1611-3349

Conference

Conference23rd International Conference on Computational Science, ICCS 2023
Country/TerritoryCzech Republic
CityPrague
Period3/07/235/07/23

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Fingerprint

Dive into the research topics of 'Convolutional Autoencoder for the Spatiotemporal Latent Representation of Turbulence'. Together they form a unique fingerprint.

Cite this