skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations

Abstract

For centuries, flow visualization has been the art of making fluid motion visible in physical and biological systems. Although such flow patterns can be, in principle, described by the Navier-Stokes equations, extracting the velocity and pressure fields directly from the images is challenging. We addressed this problem by developing hidden fluid mechanics (HFM), a physics-informed deep-learning framework capable of encoding the Navier-Stokes equations into the neural networks while being agnostic to the geometry or the initial and boundary conditions. We demonstrate HFM for several physical and biomedical problems by extracting quantitative information for which direct measurements may not be possible. HFM is robust to low resolution and substantial noise in the observation data, which is important for potential applications.

Authors:
ORCiD logo [1]; ORCiD logo [2]; ORCiD logo [2]
  1. Division of Applied Mathematics, Brown University, Providence, RI 02906, USA., NVIDIA Corporation, Santa Clara, CA 95051, USA.
  2. Division of Applied Mathematics, Brown University, Providence, RI 02906, USA.
Publication Date:
Sponsoring Org.:
USDOE
OSTI Identifier:
1601918
Grant/Contract Number:  
[AC05-76RL01830]
Resource Type:
Published Article
Journal Name:
Science
Additional Journal Information:
[Journal Name: Science Journal Volume: 367 Journal Issue: 6481]; Journal ID: ISSN 0036-8075
Publisher:
American Association for the Advancement of Science (AAAS)
Country of Publication:
United States
Language:
English

Citation Formats

Raissi, Maziar, Yazdani, Alireza, and Karniadakis, George Em. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. United States: N. p., 2020. Web. doi:10.1126/science.aaw4741.
Raissi, Maziar, Yazdani, Alireza, & Karniadakis, George Em. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. United States. doi:10.1126/science.aaw4741.
Raissi, Maziar, Yazdani, Alireza, and Karniadakis, George Em. Thu . "Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations". United States. doi:10.1126/science.aaw4741.
@article{osti_1601918,
title = {Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations},
author = {Raissi, Maziar and Yazdani, Alireza and Karniadakis, George Em},
abstractNote = {For centuries, flow visualization has been the art of making fluid motion visible in physical and biological systems. Although such flow patterns can be, in principle, described by the Navier-Stokes equations, extracting the velocity and pressure fields directly from the images is challenging. We addressed this problem by developing hidden fluid mechanics (HFM), a physics-informed deep-learning framework capable of encoding the Navier-Stokes equations into the neural networks while being agnostic to the geometry or the initial and boundary conditions. We demonstrate HFM for several physical and biomedical problems by extracting quantitative information for which direct measurements may not be possible. HFM is robust to low resolution and substantial noise in the observation data, which is important for potential applications.},
doi = {10.1126/science.aaw4741},
journal = {Science},
number = [6481],
volume = [367],
place = {United States},
year = {2020},
month = {1}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
DOI: 10.1126/science.aaw4741

Save / Share:

Works referenced in this record:

Wall Shear Stress and Early Atherosclerosis: A Review
journal, June 2000

  • Shaaban, Akram M.; Duerinckx, André J.
  • American Journal of Roentgenology, Vol. 174, Issue 6
  • DOI: 10.2214/ajr.174.6.1741657

The effect of a discrete window offset on the accuracy of cross-correlation analysis of digital PIV recordings
journal, May 1997

  • Westerweel, J.; Dabiri, D.; Gharib, M.
  • Experiments in Fluids, Vol. 23, Issue 1
  • DOI: 10.1007/s003480050082

A Survey on Transfer Learning
journal, October 2010

  • Pan, Sinno Jialin; Yang, Qiang
  • IEEE Transactions on Knowledge and Data Engineering, Vol. 22, Issue 10
  • DOI: 10.1109/TKDE.2009.191

Outflow Boundary Conditions for Arterial Networks with Multiple Outlets
journal, July 2008

  • Grinberg, Leopold; Karniadakis, George Em
  • Annals of Biomedical Engineering, Vol. 36, Issue 9
  • DOI: 10.1007/s10439-008-9527-7

Inflammation, Atherosclerosis, and Coronary Artery Disease
journal, April 2005


A convergence study of a new partitioned fluid–structure interaction algorithm based on fictitious mass and damping
journal, January 2012


Laser-induced fluorescence measurements of mixed fluid concentrationin a liquid plane shear layer
journal, November 1985

  • Koochesfahani, M. M.; Dimotakis, P. E.
  • AIAA Journal, Vol. 23, Issue 11
  • DOI: 10.2514/3.9154

Coronary Atherosclerosis Imaging by Coronary CT Angiography
journal, May 2011


Three-dimensional dynamics and transition to turbulence in the wake of bluff objects
journal, May 1992


Microscale 3D flow mapping with μDDPIV
journal, February 2007

  • Pereira, Francisco; Lu, Jian; Castaño-Graff, Emilio
  • Experiments in Fluids, Vol. 42, Issue 4
  • DOI: 10.1007/s00348-007-0267-5

Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
journal, February 2019


Planar laser induced fluorescence in aqueous flows
journal, April 2008


An Album of Fluid Motion
journal, December 1982

  • van Dyke, Milton; White, Frank M.
  • Journal of Fluids Engineering, Vol. 104, Issue 4
  • DOI: 10.1115/1.3241909

Convection and flow in porous media. Part 1. Visualization by magnetic resonance imaging
journal, February 1997


Prognostic accuracy of cerebral blood flow measurement by perfusion computed tomography, at the time of emergency room admission, in acute stroke patients
journal, March 2002

  • Wintermark, Max; Reichhart, Marc; Thiran, Jean-Philippe
  • Annals of Neurology, Vol. 51, Issue 4
  • DOI: 10.1002/ana.10136

Triangulating a Cognitive Control Network Using Diffusion-Weighted Magnetic Resonance Imaging (MRI) and Functional MRI
journal, April 2007


A low-dimensional model for simulating three-dimensional cylinder flow
journal, May 2002


On the Correlation of Parity and Small-Depth Circuits
journal, January 2014


Approximation by superpositions of a sigmoidal function
journal, December 1989

  • Cybenko, G.
  • Mathematics of Control, Signals, and Systems, Vol. 2, Issue 4
  • DOI: 10.1007/BF02551274

Carotid bifurcation atherosclerosis. Quantitative correlation of plaque localization with flow velocity profiles and wall shear stress.
journal, October 1983

  • Zarins, C. K.; Giddens, D. P.; Bharadvaj, B. K.
  • Circulation Research, Vol. 53, Issue 4
  • DOI: 10.1161/01.RES.53.4.502

Data-driven discovery of partial differential equations
journal, April 2017

  • Rudy, Samuel H.; Brunton, Steven L.; Proctor, Joshua L.
  • Science Advances, Vol. 3, Issue 4
  • DOI: 10.1126/sciadv.1602614