Neural-network-based filtered drag model for gas-particle flows

Yundi Jiang, Jari Kolehmainen, Yile Gu, Yannis G. Kevrekidis, Ali Ozel, Sankaran Sundaresan

Research output: Contribution to journalArticle

Abstract

Filtered two-fluid model (fTFM) for gas-particle flows require closures for the sub-filter scale corrections to interphase drag force and stresses, the former being more significant. In this study, we have formulated a neural-network-based model to predict the sub-grid drift velocity, which is then used to estimate the drag correction. As a part of the neural network model development effort, we derived a transport equation for drift velocity and then performed a budget analysis to conclude that an algebraic model for drift velocity in terms of the filtered variables that are resolved in a fTFM simulation is adequate, and the model should include the filtered gas-phase pressure gradient as a marker in addition to the filtered particle volume fraction and the filtered gas-solid slip velocity. Both a priori and a posteriori analyses reveal that the present model for drift velocity when used in a fTFM simulation is able to capture the fine-grid simulation results quite well.
Original languageEnglish
Pages (from-to)403-413
Number of pages11
JournalPowder Technology
Volume346
Early online date4 Dec 2018
DOIs
Publication statusPublished - 15 Mar 2019

Fingerprint

Drag
Neural networks
Gases
Fluids
Pressure gradient
Volume fraction

Cite this

Jiang, Y., Kolehmainen, J., Gu, Y., Kevrekidis, Y. G., Ozel, A., & Sundaresan, S. (2019). Neural-network-based filtered drag model for gas-particle flows. Powder Technology, 346, 403-413. https://doi.org/10.1016/j.powtec.2018.11.092
Jiang, Yundi ; Kolehmainen, Jari ; Gu, Yile ; Kevrekidis, Yannis G. ; Ozel, Ali ; Sundaresan, Sankaran. / Neural-network-based filtered drag model for gas-particle flows. In: Powder Technology. 2019 ; Vol. 346. pp. 403-413.
@article{1fd3c09411b4422aa5bbeb667754e355,
title = "Neural-network-based filtered drag model for gas-particle flows",
abstract = "Filtered two-fluid model (fTFM) for gas-particle flows require closures for the sub-filter scale corrections to interphase drag force and stresses, the former being more significant. In this study, we have formulated a neural-network-based model to predict the sub-grid drift velocity, which is then used to estimate the drag correction. As a part of the neural network model development effort, we derived a transport equation for drift velocity and then performed a budget analysis to conclude that an algebraic model for drift velocity in terms of the filtered variables that are resolved in a fTFM simulation is adequate, and the model should include the filtered gas-phase pressure gradient as a marker in addition to the filtered particle volume fraction and the filtered gas-solid slip velocity. Both a priori and a posteriori analyses reveal that the present model for drift velocity when used in a fTFM simulation is able to capture the fine-grid simulation results quite well.",
author = "Yundi Jiang and Jari Kolehmainen and Yile Gu and Kevrekidis, {Yannis G.} and Ali Ozel and Sankaran Sundaresan",
year = "2019",
month = "3",
day = "15",
doi = "10.1016/j.powtec.2018.11.092",
language = "English",
volume = "346",
pages = "403--413",
journal = "Powder Technology",
issn = "0032-5910",
publisher = "Elsevier",

}

Jiang, Y, Kolehmainen, J, Gu, Y, Kevrekidis, YG, Ozel, A & Sundaresan, S 2019, 'Neural-network-based filtered drag model for gas-particle flows', Powder Technology, vol. 346, pp. 403-413. https://doi.org/10.1016/j.powtec.2018.11.092

Neural-network-based filtered drag model for gas-particle flows. / Jiang, Yundi; Kolehmainen, Jari; Gu, Yile; Kevrekidis, Yannis G.; Ozel, Ali; Sundaresan, Sankaran.

In: Powder Technology, Vol. 346, 15.03.2019, p. 403-413.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Neural-network-based filtered drag model for gas-particle flows

AU - Jiang, Yundi

AU - Kolehmainen, Jari

AU - Gu, Yile

AU - Kevrekidis, Yannis G.

AU - Ozel, Ali

AU - Sundaresan, Sankaran

PY - 2019/3/15

Y1 - 2019/3/15

N2 - Filtered two-fluid model (fTFM) for gas-particle flows require closures for the sub-filter scale corrections to interphase drag force and stresses, the former being more significant. In this study, we have formulated a neural-network-based model to predict the sub-grid drift velocity, which is then used to estimate the drag correction. As a part of the neural network model development effort, we derived a transport equation for drift velocity and then performed a budget analysis to conclude that an algebraic model for drift velocity in terms of the filtered variables that are resolved in a fTFM simulation is adequate, and the model should include the filtered gas-phase pressure gradient as a marker in addition to the filtered particle volume fraction and the filtered gas-solid slip velocity. Both a priori and a posteriori analyses reveal that the present model for drift velocity when used in a fTFM simulation is able to capture the fine-grid simulation results quite well.

AB - Filtered two-fluid model (fTFM) for gas-particle flows require closures for the sub-filter scale corrections to interphase drag force and stresses, the former being more significant. In this study, we have formulated a neural-network-based model to predict the sub-grid drift velocity, which is then used to estimate the drag correction. As a part of the neural network model development effort, we derived a transport equation for drift velocity and then performed a budget analysis to conclude that an algebraic model for drift velocity in terms of the filtered variables that are resolved in a fTFM simulation is adequate, and the model should include the filtered gas-phase pressure gradient as a marker in addition to the filtered particle volume fraction and the filtered gas-solid slip velocity. Both a priori and a posteriori analyses reveal that the present model for drift velocity when used in a fTFM simulation is able to capture the fine-grid simulation results quite well.

U2 - 10.1016/j.powtec.2018.11.092

DO - 10.1016/j.powtec.2018.11.092

M3 - Article

VL - 346

SP - 403

EP - 413

JO - Powder Technology

JF - Powder Technology

SN - 0032-5910

ER -

Jiang Y, Kolehmainen J, Gu Y, Kevrekidis YG, Ozel A, Sundaresan S. Neural-network-based filtered drag model for gas-particle flows. Powder Technology. 2019 Mar 15;346:403-413. https://doi.org/10.1016/j.powtec.2018.11.092