Report

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

See allHide authors and affiliations

Science  30 Jan 2020:
eaaw4741
DOI: 10.1126/science.aaw4741

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

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 address 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. Important for potential applications, HFM is robust to low resolution and significant noise in the observation data.

View Full Text

Stay Connected to Science