Fourier neural operators for spatiotemporal dynamics in two-dimensional turbulence

被引:1
作者
Atif, Mohammad [1 ]
Dubey, Pulkit [2 ]
Aghor, Pratik P. [3 ]
Lopez-Marrero, Vanessa [1 ]
Zhang, Tao [1 ]
Sharfuddin, Abdullah [4 ]
Yu, Kwangmin [1 ]
Yang, Fan [1 ]
Ladeinde, Foluso [4 ]
Liu, Yangang [1 ]
Lin, Meifeng [1 ]
Li, Lingda [1 ]
机构
[1] Brookhaven Natl Lab, Upton, NY 11973 USA
[2] Univ New Hampshire, Durham, NH USA
[3] Georgia Inst Technol, Atlanta, GA USA
[4] SUNY Stony Brook, Stony Brook, NY USA
来源
PROCEEDINGS OF SC24-W: WORKSHOPS OF THE INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS | 2024年
关键词
Machine Learning; Turbulence; Fourier Neural Operator; UNIVERSAL APPROXIMATION; NONLINEAR OPERATORS; NETWORKS;
D O I
10.1109/SCW63240.2024.00013
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
High-fidelity direct numerical simulation of turbulent flows for most real-world applications remains an outstanding computational challenge. Several machine learning approaches have recently been proposed to alleviate the computational cost even though they become unstable or unphysical for long time predictions. We identify that the Fourier neural operator (FNO) based models combined with a partial differential equation (PDE) solver can accelerate fluid dynamic simulations and thus address computational expense of large-scale turbulence simulations. We treat the FNO model on the same footing as a PDE solver and answer important questions about the volume and temporal resolution of data required to build pre-trained models for turbulence. We also discuss the pitfalls of purely data-driven approaches that need to be avoided by the machine learning models to become viable and competitive tools for long time simulations of turbulence.
引用
收藏
页码:41 / 48
页数:8
相关论文
共 36 条
[1]  
Atif M., 2024, arXiv
[2]   Towards accelerating particle-resolved direct numerical simulation with neural operators [J].
Atif, Mohammad ;
Lopez-Marrero, Vanessa ;
Zhang, Tao ;
Sharfuddin, Abdullah Al Muti ;
Yu, Kwangmin ;
Yang, Jiaqi ;
Yang, Fan ;
Ladeinde, Foluso ;
Liu, Yangang ;
Lin, Meifeng ;
Li, Lingda .
STATISTICAL ANALYSIS AND DATA MINING-AN ASA DATA SCIENCE JOURNAL, 2024, 17 (03)
[3]   Essentially entropic lattice Boltzmann model: Theory and simulations [J].
Atif, Mohammad ;
Kolluru, Praveen Kumar ;
Ansumali, Santosh .
PHYSICAL REVIEW E, 2022, 106 (05)
[4]   Essentially Entropic Lattice Boltzmann Model [J].
Atif, Mohammad ;
Kolluru, Praveen Kumar ;
Thantanapally, Chakradhar ;
Ansumali, Santosh .
PHYSICAL REVIEW LETTERS, 2017, 119 (24)
[5]  
Bhushan S., 2020, FLUIDS ENG DIVISION, V83730
[6]  
Brandstetter J, 2023, Arxiv, DOI arXiv:2209.04934
[7]   Laplace neural operator for solving differential equations [J].
Cao, Qianying ;
Goswami, Somdatta ;
Karniadakis, George Em .
NATURE MACHINE INTELLIGENCE, 2024, :631-640
[8]  
Chattopadhyay A, 2024, Arxiv, DOI [arXiv:2304.07029, 10.48550/arXiv.2304.07029]
[9]   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
[10]  
Cvitanovic P., 2005, ChaosBook. org