kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Recurrent neural networks and Koopman-based frameworks for temporal predictions in a low-order model of turbulence
Univ Tehran, Fac New Sci & Technol, Tehran, Iran..
KTH, School of Engineering Sciences (SCI), Centres, Linné Flow Center, FLOW. KTH, Centres, SeRC - Swedish e-Science Research Centre.
KTH, School of Engineering Sciences (SCI), Centres, Linné Flow Center, FLOW. KTH, Centres, SeRC - Swedish e-Science Research Centre. KTH, School of Engineering Sciences (SCI), Engineering Mechanics, Fluid Mechanics and Engineering Acoustics.ORCID iD: 0000-0001-9627-5903
KTH, Centres, Science for Life Laboratory, SciLifeLab. KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0001-5211-6388
Show others and affiliations
2021 (English)In: International Journal of Heat and Fluid Flow, ISSN 0142-727X, E-ISSN 1879-2278, Vol. 90, article id 108816Article in journal (Refereed) Published
Abstract [en]

The capabilities of recurrent neural networks and Koopman-based frameworks are assessed in the prediction of temporal dynamics of the low-order model of near-wall turbulence by Moehlis et al. (New J. Phys. 6, 56, 2004). Our results show that it is possible to obtain excellent reproductions of the long-term statistics and the dynamic behavior of the chaotic system with properly trained long-short-term memory (LSTM) networks, leading to relative errors in the mean and the fluctuations below 1%. Besides, a newly developed Koopman-based framework, called Koopman with nonlinear forcing (KNF), leads to the same level of accuracy in the statistics at a significantly lower computational expense. Furthermore, the KNF framework outperforms the LSTM network when it comes to short-term predictions. We also observe that using a loss function based only on the instantaneous predictions of the chaotic system can lead to suboptimal reproductions in terms of long-term statistics. Thus, we propose a model-selection criterion based on the computed statistics which allows to achieve excellent statistical reconstruction even on small datasets, with minimal loss of accuracy in the instantaneous predictions.

Place, publisher, year, edition, pages
Elsevier BV , 2021. Vol. 90, article id 108816
Keywords [en]
Dynamical systems, Machine learning, Data-driven modeling, Recurrent neural networks, Koopman operator
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:kth:diva-300856DOI: 10.1016/j.ijheatfluidflow.2021.108816ISI: 000687251300005Scopus ID: 2-s2.0-85106289306OAI: oai:DiVA.org:kth-300856DiVA, id: diva2:1598046
Note

QC 20210928

Available from: 2021-09-28 Created: 2021-09-28 Last updated: 2023-05-15Bibliographically approved
In thesis
1. Time, space and control: deep-learning applications to turbulent flows
Open this publication in new window or tab >>Time, space and control: deep-learning applications to turbulent flows
2023 (English)Doctoral thesis, comprehensive summary (Other academic)
Alternative title[sv]
Tid, rum och kontroll: djupinlärningsapplikationer för turbulenta flöden
Abstract [en]

In the present thesis, the application of deep learning and deep reinforcement learning to turbulent-flow simulations is investigated. Deep-learning models are trained to perform temporal and spatial predictions, while deep reinforcement learning is applied to a flow-control problem, namely the reduction of drag in an open channel flow. Long short-term memory (LSTM, Hochreiter & Schmidhuber 1997) networks and Koopman non-linear forcing (KNF) models are optimized to perform temporal predictions in two reduced-order-models of turbulence, namely the nine-equations model proposed by Moehlis et al. (2004) and a truncated proper orthogonal decomposition (POD) of a minimal channel flow (Jiménez & Moin 1991). In the first application, both models are able to produce accurate short-term predictions. Furthermore, the predicted system trajectories are statistically correct. KNF models outperform LSTM networks in short-term predictions, with a much lower training computational cost. In the second task, only LSTMs can be trained successfully, predicting trajectories that are statistically accurate. Spatial predictions are performed in two turbulent flows: an open channel flow and a boundary-layer flow. Fully-convolutional networks (FCNs) are used to predict two-dimensional velocity-fluctuation fields at a given wall-normal location using wall measurements (and vice versa). Thanks to the non-linear nature of these models, they provide a better reconstruction performance than optimal linear methods like extended POD (Borée 2003). Finally, we show the potential of deep reinforcement learning in discovering new control strategies for turbulent flows. By framing the fluid-dynamics problem as a multi-agent reinforcement-learning environment and by training the agents using a location-invariant deep deterministic policy-gradient (DDPG) algorithm, we are able to learn a control strategy that achieves a remarkable 30% drag reduction, improving over existing strategies by about 10 percentage points.

Abstract [sv]

I den förinställda avhandlingen undersöks tillämpningen av djupinlärning och djupförstärkningsinlärning på turbulenta flödessimuleringar. Modeller för djupinlärning tränas för att utföra tids- och rumsförutsägelser, medan djupförstärkningsinlärning tillämpas på ett flödeskontrollproblem, nämligen minskningen av motståndet i ett öppet kanalflöde. Long short-term memory (LSTM, Hochreiter & Schmidhuber 1997) nätverk och Koopman non-linear forcing (KNF) modeller optimeras för att utföratidsförutsägelser i två turbulensmodeller med reducerad ordning, nämligen nio-ekvationsmodellen som föreslagits av Moehlis et al. (2004) och en trunkerad proper orthogonal decomposition (POD) av ett minimalt kanalflöde (Jiménez & Moin 1991). I den första applikationen kan båda modellerna producera korrekta korttidsförutsägelser, dessutom är de förutsagda trajektorierna statistiskt korrekta. KNF-modeller överträffar LSTM-nätverk i kortsiktiga förutsägelser, med en mycket lägre utbildningsberäkningskostnad. I den andra uppgiften kan endast LSTM nätverken tränas framgångsrikt, med trajektorier som är statistiskt korrekta. Spatiala förutsägelser utförs i två turbulenta flöden, en öppen kanal flöde och en gränsskikt. Fully-convolutional networks (FCN) används för att förutsäga tvådimensionella hastighetsfluktuationsfält vid givet avstånd från väggen med hjälp av väggmätningar (och vice versa). Tack vare deras icke-linjär karaktär ger dessa modeller bättre rekonstruktionsprestanda än optimala linjära metoder som extended POD (Borée 2003). Slutligen visar vi potentialen med djup förstärkningsinlärning för att upptäcka nya kontrollstrategier i turbulenta flöden. Genom att inrama strömningsmekaniska problemet som en förstärknings-inlärningsmiljö med flera agenter och genom att träna agenterna med hjälp av en positionsinvariant deep deterministic policy gradient (DDPG) algoritm, kan vi lära oss en kontrollstrategi som uppnår en anmärkningsvärd 30% minskning av luftmotståndet, vilket jämfört med existerande strategier är en förbättring med cirka 10 procentenheter.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2023. p. 342
Series
TRITA-SCI-FOU ; 2023:27
Keywords
turbulence, deep learning, deep reinforcement learning, flow control, turbulens, djupinlärning, djupförstärkningsinlärning, flödeskontroll
National Category
Fluid Mechanics
Research subject
Engineering Mechanics
Identifiers
urn:nbn:se:kth:diva-326961 (URN)978-91-8040-601-7 (ISBN)
Public defence
2023-06-12, F3, Lindstedtsvägen 26 & 28, Stockholm, 10:00 (English)
Opponent
Supervisors
Funder
EU, European Research Council, 2021-CoG-101043998, DEEPCONTROLSwedish e‐Science Research CenterKnut and Alice Wallenberg Foundation
Note

QC 230516

Available from: 2023-05-16 Created: 2023-05-15 Last updated: 2025-02-09Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Guastoni, LucaSchlatter, PhilippAzizpour, HosseinVinuesa, Ricardo

Search in DiVA

By author/editor
Guastoni, LucaSchlatter, PhilippAzizpour, HosseinVinuesa, Ricardo
By organisation
Linné Flow Center, FLOWSeRC - Swedish e-Science Research CentreFluid Mechanics and Engineering AcousticsScience for Life Laboratory, SciLifeLabRobotics, Perception and Learning, RPL
In the same journal
International Journal of Heat and Fluid Flow
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 239 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf