Towards accelerating particle-resolved direct numerical simulation with neural operators

被引：1

作者：

Atif, Mohammad ^{[1
]}

Lopez-Marrero, Vanessa ^{[1
,2
]}

Zhang, Tao ^{[3
]}

Sharfuddin, Abdullah Al Muti ^{[4
]}

Yu, Kwangmin ^{[1
]}

Yang, Jiaqi ^{[5
]}

Yang, Fan ^{[3
]}

Ladeinde, Foluso ^{[4
]}

Liu, Yangang ^{[3
]}

Lin, Meifeng ^{[1
]}

Li, Lingda ^{[1
]}

机构：

[1] Brookhaven Natl Lab, Computat Sci Initiat, Upton, NY 11973 USA

[2] SUNY Stony Brook, IACS, Stony Brook, NY USA

[3] Brookhaven Natl Lab, Environm & Climate Sci Dept, Upton, NY USA

[4] SUNY Stony Brook, Dept Mech Engn, Stony Brook, NY USA

[5] Emory Univ, Dept Math, Atlanta, GA USA

来源：

STATISTICAL ANALYSIS AND DATA MINING-AN ASA DATA SCIENCE JOURNAL | 2024年 / 17卷 / 03期

关键词：

fluid dynamics; machine learning; neural operators; particle resolved direct numerical simulation; UNIVERSAL APPROXIMATION; NONLINEAR OPERATORS; NETWORKS;

D O I：

10.1002/sam.11690

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present our ongoing work aimed at accelerating a particle-resolved direct numerical simulation model designed to study aerosol-cloud-turbulence interactions. The dynamical model consists of two main components-a set of fluid dynamics equations for air velocity, temperature, and humidity, coupled with a set of equations for particle (i.e., cloud droplet) tracing. Rather than attempting to replace the original numerical solution method in its entirety with a machine learning (ML) method, we consider developing a hybrid approach. We exploit the potential of neural operator learning to yield fast and accurate surrogate models and, in this study, develop such surrogates for the velocity and vorticity fields. We discuss results from numerical experiments designed to assess the performance of ML architectures under consideration as well as their suitability for capturing the behavior of relevant dynamical systems.

引用

页数：11

共 29 条

[1] Essentially entropic lattice Boltzmann model: Theory and simulations [J].

Atif, Mohammad ;

Kolluru, Praveen Kumar ;

Ansumali, Santosh .

PHYSICAL REVIEW E, 2022, 106 (05)

[2] Essentially Entropic Lattice Boltzmann Model [J].

Atif, Mohammad ;

Kolluru, Praveen Kumar ;

Thantanapally, Chakradhar ;

Ansumali, Santosh .

PHYSICAL REVIEW LETTERS, 2017, 119 (24)

[3]

Brandstetter J, 2023, Arxiv, DOI arXiv:2209.04934

[4] PERFORMANCE OF UNDER-RESOLVED 2-DIMENSIONAL INCOMPRESSIBLE-FLOW SIMULATIONS [J].

BROWN, DL ;

MINION, ML .

JOURNAL OF COMPUTATIONAL PHYSICS, 1995, 122 (01) :165-183

[5] Discovering governing equations from data by sparse identification of nonlinear dynamical systems [J].

Brunton, Steven L. ;

Proctor, Joshua L. ;

Kutz, J. Nathan .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2016, 113 (15) :3932-3937

[6] UNIVERSAL APPROXIMATION TO NONLINEAR OPERATORS BY NEURAL NETWORKS WITH ARBITRARY ACTIVATION FUNCTIONS AND ITS APPLICATION TO DYNAMICAL-SYSTEMS [J].

CHEN, TP ;

CHEN, H .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1995, 6 (04) :911-917

[7] Turbulence Modeling in the Age of Data [J].

Duraisamy, Karthik ;

Iaccarino, Gianluca ;

Xiao, Heng .

ANNUAL REVIEW OF FLUID MECHANICS, VOL 51, 2019, 51 :357-377

[8] Investigation of Turbulent Entrainment-Mixing Processes With a New Particle-Resolved Direct Numerical Simulation Model [J].

Gao, Zheng ;

Liu, Yangang ;

Li, Xiaolin ;

Lu, Chunsong .

JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 2018, 123 (04) :2194-2214

[9] Model-parallel Fourier neural operators as learned surrogates for large-scale parametric PDEs [J].

Grady, Thomas J. ;

Khan, Rishi ;

Louboutin, Mathias ;

Yin, Ziyi ;

Witte, Philipp A. ;

Chandra, Ranveer ;

Hewett, Russell J. ;

Herrmann, Felix J. .

COMPUTERS & GEOSCIENCES, 2023, 178

[10] Fourier Neural Operator Network for Fast Photoacoustic Wave Simulations [J].

Guan, Steven ;

Hsu, Ko-Tsung ;

Chitnis, Parag V. .

ALGORITHMS, 2023, 16 (02)

← 1 2 3 →