Physics-informed neural networks for solving forward and inverse flow problems via the Boltzmann-BGK formulation

Lou, Qin; Meng, Xuhui; Karniadakis, George Em

doi:10.1016/j.jcp.2021.110676

Physics-informed neural networks for solving forward and inverse flow problems via the Boltzmann-BGK formulation

Journal Article · Mon Sep 06 00:00:00 EDT 2021 · Journal of Computational Physics

DOI:https://doi.org/10.1016/j.jcp.2021.110676· OSTI ID:1897196

Lou, Qin ^[1]; Meng, Xuhui ^[2]; Karniadakis, George Em ^[3]

University of Shanghai for Science & Technology (China)
Brown University, Providence, RI (United States)
Brown University, Providence, RI (United States); Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)

We report the Boltzmann equation with the Bhatnagar-Gross-Krook collision model (Boltzmann-BGK equation) has been employed to describe multiscale flows, i.e., from the hydrodynamic limit to free molecular flow. In this study we employ physics-informed neural networks (PINNs) to solve forward and inverse problems via the Boltzmann-BGK formulation (PINN-BGK), enabling PINNs to model flows in both the continuum and rarefied regimes. In particular, the PINN-BGK is composed of three sub-networks, i.e., the first for approximating the equilibrium distribution function, the second for approximating the non-equilibrium distribution function, and the third one for encoding the Boltzmann-BGK equation as well as the corresponding boundary/initial conditions. By minimizing the residuals of the governing equations and the mismatch between the predicted and provided boundary/initial conditions, we can approximate the Boltzmann-BGK equation for both continuous and rarefied flows. For forward problems, the PINN-BGK is utilized to solve various benchmark flows given boundary/initial conditions, e.g. Kovasznay flow, Taylor-Green flow, cavity flow, and micro Couette flow for Knudsen number up to 5. For inverse problems, we focus on rarefied flows in which accurate boundary conditions are difficult to obtain. We employ the PINN-BGK to infer the flow field in the entire computational domain given a limited number of interior scattered measurements on the velocity without using the (unknown) boundary conditions. Results for the two-dimensional micro Couette and micro cavity flows with Knudsen numbers ranging from 0.1 to 10 indicate that the PINN-BGK can infer the velocity field in the entire domain with good accuracy. Finally, we also present some results on using transfer learning to accelerate the training process. Specifically, we can obtain a three-fold speedup comparted to the standard training process (e.g., Adam plus L-BFG-B) for the two-dimensional flow problems considered in our work.

View Accepted Manuscript (DOE)

Research Organization:: Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)

Sponsoring Organization:: USDOE; National Natural Science Foundation of China; Natural Science Foundation of Shanghai

Grant/Contract Number:: AC05-76RL01830; SC0019453

OSTI ID:: 1897196

Alternate ID(s):: OSTI ID: 1819973
OSTI ID: 23206060

Report Number(s):: PNNL-SA-178766

Journal Information:: Journal of Computational Physics, Journal Name: Journal of Computational Physics Vol. 447; ISSN 0021-9991

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (69)

Lattice Boltzmann equation with Overset method for moving objects in two-dimensional flows Lallemand, Pierre; Luo, Li-Shi Journal of Computational Physics, Vol. 407 https://doi.org/10.1016/j.jcp.2019.109223	journal	April 2020
Kinetic multiscale scheme based on the discrete-velocity and lattice-Boltzmann methods Aristov, V. V.; Ilyin, O. V.; Rogozin, O. A. Journal of Computational Science, Vol. 40 https://doi.org/10.1016/j.jocs.2019.101064	journal	February 2020
Laminar flow behind a two-dimensional grid Kovasznay, L. I. G. Mathematical Proceedings of the Cambridge Philosophical Society, Vol. 44, Issue 1 https://doi.org/10.1017/s0305004100023999	journal	January 1948
Discrete Velocity Model and Implicit Scheme for the bgk Equation of Rarefied gas Dynamics Mieussens, Luc Mathematical Models and Methods in Applied Sciences, Vol. 10, Issue 08 https://doi.org/10.1142/s0218202500000562	journal	November 2000
Neural-network-based approximations for solving partial differential equations Dissanayake, M. W. M. G.; Phan-Thien, N. Communications in Numerical Methods in Engineering, Vol. 10, Issue 3 https://doi.org/10.1002/cnm.1640100303	journal	March 1994
Lattice BGK Model for Incompressible Navier–Stokes Equation Guo, Zhaoli; Shi, Baochang; Wang, Nengchao Journal of Computational Physics, Vol. 165, Issue 1 https://doi.org/10.1006/jcph.2000.6616	journal	November 2000
A Characteristic Galerkin Method for Discrete Boltzmann Equation Lee, Taehun; Lin, Ching-Long Journal of Computational Physics, Vol. 171, Issue 1 https://doi.org/10.1006/jcph.2001.6791	journal	July 2001
Under-resolved and large eddy simulations of a decaying Taylor–Green vortex with the cumulant lattice Boltzmann method Geier, Martin; Lenz, Stephan; Schönherr, Martin Theoretical and Computational Fluid Dynamics, Vol. 35, Issue 2 https://doi.org/10.1007/s00162-020-00555-7	journal	November 2020
A relevant and robust vacuum-drying model applied to hardwoods Redman, Adam L.; Bailleres, Henri; Perré, Patrick Wood Science and Technology, Vol. 51, Issue 4 https://doi.org/10.1007/s00226-017-0908-7	journal	April 2017
Surface–gas interaction effects on nanoscale gas flows Barisik, Murat; Beskok, Ali Microfluidics and Nanofluidics, Vol. 13, Issue 5 https://doi.org/10.1007/s10404-012-1000-0	journal	May 2012
A new curved boundary treatment for LBM modeling of thermal gaseous microflow in the slip regime Liu, Zhenyu; Mu, Zhiyu; Wu, Huiying Microfluidics and Nanofluidics, Vol. 23, Issue 2 https://doi.org/10.1007/s10404-019-2192-3	journal	January 2019
A theory of transfer learning with applications to active learning Yang, Liu; Hanneke, Steve; Carbonell, Jaime Machine Learning, Vol. 90, Issue 2 https://doi.org/10.1007/s10994-012-5310-y	journal	July 2012
Hopf bifurcation of the unsteady regularized driven cavity flow Shen, Jie Journal of Computational Physics, Vol. 95, Issue 1 https://doi.org/10.1016/0021-9991(91)90261-I	journal	July 1991
Dynamics of regularized cavity flow at high Reynolds numbers Shen, Jie Applied Mathematics Letters, Vol. 2, Issue 4 https://doi.org/10.1016/0893-9659(89)90093-1	journal	January 1989
Some recent results on discrete velocity models and ramifications for lattice Boltzmann equation Luo, Li-Shi Computer Physics Communications, Vol. 129, Issue 1-3 https://doi.org/10.1016/S0010-4655(00)00093-X	journal	July 2000
Universal Approximation Using Feedforward Neural Networks: A Survey of Some Existing Methods, and Some New Results Scarselli, Franco; Chung Tsoi, Ah Neural Networks, Vol. 11, Issue 1 https://doi.org/10.1016/S0893-6080(97)00097-X	journal	January 1998
A multiscale study of density-driven flow with dissolution in porous media Meng, Xuhui; Sun, Haoran; Guo, Zhaoli Advances in Water Resources, Vol. 142 https://doi.org/10.1016/j.advwatres.2020.103640	journal	August 2020
Atomic layer deposition in porous electrodes: A pore-scale modeling study Fang, Wen-Zhen; Tang, Yu-Qing; Ban, Chunmei Chemical Engineering Journal, Vol. 378 https://doi.org/10.1016/j.cej.2019.122099	journal	December 2019
Physics-informed neural networks for high-speed flows Mao, Zhiping; Jagtap, Ameya D.; Karniadakis, George Em Computer Methods in Applied Mechanics and Engineering, Vol. 360 https://doi.org/10.1016/j.cma.2019.112789	journal	March 2020
PPINN: Parareal physics-informed neural network for time-dependent PDEs Meng, Xuhui; Li, Zhen; Zhang, Dongkun Computer Methods in Applied Mechanics and Engineering, Vol. 370 https://doi.org/10.1016/j.cma.2020.113250	journal	October 2020
A multiscale approach for simulation of shale gas transport in organic nanopores Yang, Xu; Zhou, Wenning; Liu, Xunliang Energy, Vol. 210 https://doi.org/10.1016/j.energy.2020.118547	journal	November 2020
Localized lattice Boltzmann equation model for simulating miscible viscous displacement in porous media Meng, Xuhui; Guo, Zhaoli International Journal of Heat and Mass Transfer, Vol. 100 https://doi.org/10.1016/j.ijheatmasstransfer.2016.04.095	journal	September 2016
Multiscale transport mechanism of shale gas in micro/nano-pores Yu, Hao; Chen, Jie; Zhu, YinBo International Journal of Heat and Mass Transfer, Vol. 111 https://doi.org/10.1016/j.ijheatmasstransfer.2017.04.050	journal	August 2017
An implicit kinetic scheme for multiscale heat transfer problem accounting for phonon dispersion and polarization Zhang, Chuang; Guo, Zhaoli; Chen, Songze International Journal of Heat and Mass Transfer, Vol. 130 https://doi.org/10.1016/j.ijheatmasstransfer.2018.10.141	journal	March 2019
Discrete unified gas kinetic scheme for multiscale heat transfer with arbitrary temperature difference Zhang, Chuang; Guo, Zhaoli International Journal of Heat and Mass Transfer, Vol. 134 https://doi.org/10.1016/j.ijheatmasstransfer.2019.02.056	journal	May 2019
A unified gas-kinetic scheme for continuum and rarefied flows Xu, Kun; Huang, Juan-Chen Journal of Computational Physics, Vol. 229, Issue 20 https://doi.org/10.1016/j.jcp.2010.06.032	journal	October 2010
Accuracy analysis of high-order lattice Boltzmann models for rarefied gas flows Meng, Jianping; Zhang, Yonghao Journal of Computational Physics, Vol. 230, Issue 3 https://doi.org/10.1016/j.jcp.2010.10.023	journal	February 2011
A comparative study of an asymptotic preserving scheme and unified gas-kinetic scheme in continuum flow limit Chen, Songze; Xu, Kun Journal of Computational Physics, Vol. 288 https://doi.org/10.1016/j.jcp.2015.02.014	journal	May 2015
Parametrization of the cumulant lattice Boltzmann method for fourth order accurate diffusion part I: Derivation and validation Geier, Martin; Pasquali, Andrea; Schönherr, Martin Journal of Computational Physics, Vol. 348 https://doi.org/10.1016/j.jcp.2017.05.040	journal	November 2017
Parametrization of the cumulant lattice Boltzmann method for fourth order accurate diffusion part II: Application to flow around a sphere at drag crisis Geier, Martin; Pasquali, Andrea; Schönherr, Martin Journal of Computational Physics, Vol. 348 https://doi.org/10.1016/j.jcp.2017.07.004	journal	November 2017
Multiscale model reduction for shale gas transport in poroelastic fractured media Akkutlu, I. Yucel; Efendiev, Yalchin; Vasilyeva, Maria Journal of Computational Physics, Vol. 353 https://doi.org/10.1016/j.jcp.2017.10.023	journal	January 2018
DGM: A deep learning algorithm for solving partial differential equations Sirignano, Justin; Spiliopoulos, Konstantinos Journal of Computational Physics, Vol. 375 https://doi.org/10.1016/j.jcp.2018.08.029	journal	December 2018
Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations Raissi, M.; Perdikaris, P.; Karniadakis, G. E. Journal of Computational Physics, Vol. 378 https://doi.org/10.1016/j.jcp.2018.10.045	journal	February 2019
An improved three-dimensional implicit discrete velocity method on unstructured meshes for all Knudsen number flows Yang, L. M.; Shu, C.; Yang, W. M. Journal of Computational Physics, Vol. 396 https://doi.org/10.1016/j.jcp.2019.07.002	journal	November 2019
A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems Meng, Xuhui; Karniadakis, George Em Journal of Computational Physics, Vol. 401 https://doi.org/10.1016/j.jcp.2019.109020	journal	January 2020
Data-driven deep learning of partial differential equations in modal space Wu, Kailiang; Xiu, Dongbin Journal of Computational Physics, Vol. 408 https://doi.org/10.1016/j.jcp.2020.109307	journal	May 2020
A velocity-space adaptive unified gas kinetic scheme for continuum and rarefied flows Xiao, Tianbai; Liu, Chang; Xu, Kun Journal of Computational Physics, Vol. 415 https://doi.org/10.1016/j.jcp.2020.109535	journal	August 2020
The lattice Boltzmann method for nearly incompressible flows Lallemand, Pierre; Luo, Li-Shi; Krafczyk, Manfred Journal of Computational Physics, Vol. 431 https://doi.org/10.1016/j.jcp.2020.109713	journal	April 2021
NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations Jin, Xiaowei; Cai, Shengze; Li, Hui Journal of Computational Physics, Vol. 426 https://doi.org/10.1016/j.jcp.2020.109951	journal	February 2021
Transfer learning using computational intelligence: A survey Lu, Jie; Behbood, Vahid; Hao, Peng Knowledge-Based Systems, Vol. 80 https://doi.org/10.1016/j.knosys.2015.01.010	journal	May 2015
A unified implicit scheme for kinetic model equations. Part I. Memory reduction technique Chen, Songze; Zhang, Chuang; Zhu, Lianhua Science Bulletin, Vol. 62, Issue 2 https://doi.org/10.1016/j.scib.2016.12.010	journal	January 2017
Kinetic theory representation of hydrodynamics: a way beyond the Navier–Stokes equation Shan, Xiaowen; Yuan, Xue-Feng; Chen, Hudong Journal of Fluid Mechanics, Vol. 550, Issue -1 https://doi.org/10.1017/S0022112005008153	journal	February 2006
Macroscopic description of arbitrary Knudsen number flow using Boltzmann–BGK kinetic theory Chen, Hudong; Orszag, Steven A.; Staroselsky, Ilya Journal of Fluid Mechanics, Vol. 574 https://doi.org/10.1017/S0022112006004241	journal	February 2007
A high-order moment approach for capturing non-equilibrium phenomena in the transition regime Gu, Xiao-Jun; Emerson, David R. Journal of Fluid Mechanics, Vol. 636 https://doi.org/10.1017/S002211200900768X	journal	September 2009
Rough-wall turbulent boundary layers in the transition regime Bandyopadhyay, Promode R. Journal of Fluid Mechanics, Vol. 180, Issue -1 https://doi.org/10.1017/S0022112087001794	journal	July 1987
Laminar flow behind a two-dimensional grid Kovasznay, L. I. G. Mathematical Proceedings of the Cambridge Philosophical Society, Vol. 44, Issue 1 https://doi.org/10.1017/S0305004100023999	journal	January 1948
Molecular and Continuum Transport Perspectives on Electroosmotic Slip Flows Celebi, Alper Tunga; Beskok, Ali The Journal of Physical Chemistry C, Vol. 122, Issue 17 https://doi.org/10.1021/acs.jpcc.8b02519	journal	April 2018
Modulation of Electroosmotic Flow by Neutral Polymers Qiao, Rui; He, Ping Langmuir, Vol. 23, Issue 10 https://doi.org/10.1021/la063042v	journal	May 2007
An extrapolation method for boundary conditions in lattice Boltzmann method Guo, Zhaoli; Zheng, Chuguang; Shi, Baochang Physics of Fluids, Vol. 14, Issue 6 https://doi.org/10.1063/1.1471914	journal	June 2002
Flow of methane in shale nanopores at low and high pressure by molecular dynamics simulations Jin, Zhehui; Firoozabadi, Abbas The Journal of Chemical Physics, Vol. 143, Issue 10 https://doi.org/10.1063/1.4930006	journal	September 2015
Uniformly accurate machine learning-based hydrodynamic models for kinetic equations Han, Jiequn; Ma, Chao; Ma, Zheng Proceedings of the National Academy of Sciences, Vol. 116, Issue 44 https://doi.org/10.1073/pnas.1909854116	journal	October 2019
Gauss quadrature rules for the evaluation of $2\pi^{-1/2} \int_0^\infty \exp(-x^2) f(x) dx$ Galant, David Mathematics of Computation, Vol. 23, Issue 107 https://doi.org/10.1090/S0025-5718-69-99859-7	journal	September 1969
An improved unified gas-kinetic scheme and the study of shock structures Xu, K.; Huang, J. -C. IMA Journal of Applied Mathematics, Vol. 76, Issue 5 https://doi.org/10.1093/imamat/hxr002	journal	March 2011
A Model for Collision Processes in Gases. I. Small Amplitude Processes in Charged and Neutral One-Component Systems Bhatnagar, P. L.; Gross, E. P.; Krook, M. Physical Review, Vol. 94, Issue 3 https://doi.org/10.1103/PhysRev.94.511	journal	May 1954
Nonequilibrium gas flow in the transition regime: A molecular-dynamics study Bhattacharya, D. K.; Lie, G. C. Physical Review A, Vol. 43, Issue 2 https://doi.org/10.1103/PhysRevA.43.761	journal	January 1991
Theory of the lattice Boltzmann method: Lattice Boltzmann models for nonideal gases Luo, Li-Shi Physical Review E, Vol. 62, Issue 4 https://doi.org/10.1103/PhysRevE.62.4982	journal	October 2000
Lattice Boltzmann model for incompressible flows through porous media Guo, Zhaoli; Zhao, T. S. Physical Review E, Vol. 66, Issue 3 https://doi.org/10.1103/PhysRevE.66.036304	journal	September 2002
Discrete unified gas kinetic scheme for all Knudsen number flows: Low-speed isothermal case Guo, Zhaoli; Xu, Kun; Wang, Ruijie Physical Review E, Vol. 88, Issue 3 https://doi.org/10.1103/PhysRevE.88.033305	journal	September 2013
Boundary condition for lattice Boltzmann modeling of microscale gas flows with curved walls in the slip regime Tao, Shi; Guo, Zhaoli Physical Review E, Vol. 91, Issue 4 https://doi.org/10.1103/PhysRevE.91.043305	journal	April 2015
Impact of the kinetic boundary condition on porous media flow in the lattice Boltzmann formulation Singh, Shiwani; Jiang, Fei; Tsuji, Takeshi Physical Review E, Vol. 96, Issue 1 https://doi.org/10.1103/PhysRevE.96.013303	journal	July 2017
Discretization of the Velocity Space in the Solution of the Boltzmann Equation Shan, Xiaowen; He, Xiaoyi Physical Review Letters, Vol. 80, Issue 1 https://doi.org/10.1103/PhysRevLett.80.65	journal	January 1998
Transitioning from the art to the science of thin films: 1964 to 2003 Greene, J. E. Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films, Vol. 21, Issue 5 https://doi.org/10.1116/1.1600449	journal	September 2003
Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations Raissi, Maziar; Yazdani, Alireza; Karniadakis, George Em Science, Vol. 367, Issue 6481 https://doi.org/10.1126/science.aaw4741	journal	January 2020
Numerical method for computing two-dimensional unsteady rarefied gas flows in arbitrarily shaped domains Titarev, V. A. Computational Mathematics and Mathematical Physics, Vol. 49, Issue 7 https://doi.org/10.1134/S0965542509070112	journal	July 2009
A Limited Memory Algorithm for Bound Constrained Optimization Byrd, Richard H.; Lu, Peihuang; Nocedal, Jorge SIAM Journal on Scientific Computing, Vol. 16, Issue 5 https://doi.org/10.1137/0916069	journal	September 1995
DeepXDE: A Deep Learning Library for Solving Differential Equations Lu, Lu; Meng, Xuhui; Mao, Zhiping SIAM Review, Vol. 63, Issue 1 https://doi.org/10.1137/19M1274067	journal	January 2021
Discrete Velocity Model and Implicit Scheme for the bgk Equation of Rarefied gas Dynamics Mieussens, Luc Mathematical Models and Methods in Applied Sciences, Vol. 10, Issue 08 https://doi.org/10.1142/S0218202500000562	journal	November 2000
Lattice-Boltzmann Method for Complex Flows Aidun, Cyrus K.; Clausen, Jonathan R. Annual Review of Fluid Mechanics, Vol. 42, Issue 1 https://doi.org/10.1146/annurev-fluid-121108-145519	journal	January 2010
Lattice BGK Models for Navier-Stokes Equation Qian, Y. H.; D'Humières, D.; Lallemand, P. Europhysics Letters (EPL), Vol. 17, Issue 6 https://doi.org/10.1209/0295-5075/17/6/001	journal	February 1992

Similar Records

Derivation of slip boundary conditions for the Navier-Stokes system from the Boltzmann equation

Journal Article · Tue Jan 31 23:00:00 EST 1989 · J. Stat. Phys.; (United States) · OSTI ID:5844069

Momentum transport in rarefied gases

Thesis/Dissertation · Sat Dec 31 23:00:00 EST 1988 · OSTI ID:6064994

A coupled ordinates method for solution acceleration of rarefied gas dynamics simulations

Journal Article · Fri May 15 00:00:00 EDT 2015 · Journal of Computational Physics · OSTI ID:22465625

Related Subjects

71 CLASSICAL AND QUANTUM MECHANICS
GENERAL PHYSICS
Boltzmann-BGK equation
physics-informed neural networks
rarefied flows
velocity slip
inverse problems
transit learning

Physics-informed neural networks for solving forward and inverse flow problems via the Boltzmann-BGK formulation

Citation Formats

References (69)

Similar Records

Related Subjects