Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Drag01:23

Drag

301
Drag is a resistive force opposing an object’s motion through a fluid, resulting from surface pressure and shear forces. It comprises two components: a perpendicular one from pressure and a tangential one from shear stress. Accurate drag calculations use pressure and wall shear stress distributions, often determined through Computational Fluid Dynamics (CFD) or wind tunnel testing. The drag coefficient, a dimensionless measure, depends on factors like shape, Reynolds number, Mach number,...
301
Newtonian Fluid: Problem Solving01:18

Newtonian Fluid: Problem Solving

709
Newtonian fluids exhibit a constant viscosity, meaning their shear stress and shear strain rate are directly proportional. This property ensures a predictable and stable response to applied forces, maintaining a linear relationship between force and flow. Examples include water, air, and light oils, consistently demonstrating this proportional behavior regardless of external conditions.
A velocity gradient forms within the fluid when a Newtonian fluid is placed between two parallel plates, with...
709
Accelerating Fluids01:17

Accelerating Fluids

1.9K
When a fluid is in constant acceleration, the pressure and buoyant force equations are modified. Suppose a beaker is placed in an elevator accelerating upward with a constant acceleration, a. In the beaker, assume there is a thin cylinder of height h with an infinitesimal cross-sectional area, ΔS.
The motion of the liquid within this infinitesimal cylinder is considered to obtain the pressure difference. Three vertical forces act on this liquid:
1.9K
Drag Force and Terminal Speed01:18

Drag Force and Terminal Speed

3.0K
An interesting force in everyday life is the force of drag on an object when it is moving in a fluid. Like friction, the drag force always opposes the motion of an object. Unlike simple friction, the drag force is proportional to some function of the velocity of the object in that fluid. This functionality is complicated and depends upon the shape of the object, its size, its velocity, and the fluid it is in. For most large objects, such as cyclists, cars, and baseballs, that are not moving too...
3.0K
Pressure of Fluids01:14

Pressure of Fluids

19.7K
There are many examples of pressure in fluids in everyday life, such as in relation to blood (high or low blood pressure) and in relation to weather (high- and low-pressure weather systems). A given force can have a significantly different effect, depending on the area over which the force is exerted. For instance, a force applied to an area of 1 mm2 has a pressure that is 100 times greater than the same force applied to an area of 1 cm2. That's why a sharp needle is able to poke through...
19.7K
Stokes' Law01:20

Stokes' Law

2.3K
Viscous forces, like friction, are intermolecular forces that resist the relative motion of molecules over each other. When a solid body moves through a liquid, viscous forces drag it in the opposite direction. The force's magnitude depends on the solid's shape and size, as well as its speed and the liquid's coefficient of viscosity, density and temperature.
The expression for the force on a solid spherical object in a fluid is called Stokes' law. Stokes' law is valid only...
2.3K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Proteomics Analysis of Changes in the Water-Holding Capacity of Yak Meat During Postmortem Aging.

Foods (Basel, Switzerland)·2026
Same author

Diff-MomentFormer: Generative Diffusion-Augmented Transformer for End-to-End Joint Moment Estimation.

Sensors (Basel, Switzerland)·2026
Same author

A novel mucosal lymphatic vessel classification correlating with histopathology of early esophageal squamous cell carcinoma: A proposed system from China.

Chinese medical journal·2026
Same author

Survival benefit of adjuvant radiotherapy for gallbladder cancer patients with stage IIIA: a population based, propensity score-matched study.

Translational cancer research·2026
Same author

Effects of defatting treatments on flavor components and lipid characteristics in bowl-steamed Tan lamb: A comprehensive investigation based on lipidomics and multi-platform flavor analysis.

Food chemistry·2026
Same author

Development of a secretable frataxin for enhanced efficacy in treating Friedreich's Ataxia.

Molecular therapy. Advances·2026

Related Experiment Video

Updated: Dec 4, 2025

Experimental Measurement of Settling Velocity of Spherical Particles in Unconfined and Confined Surfactant-based Shear Thinning Viscoelastic Fluids
10:28

Experimental Measurement of Settling Velocity of Spherical Particles in Unconfined and Confined Surfactant-based Shear Thinning Viscoelastic Fluids

Published on: January 3, 2014

15.0K

Physics-Guided Deep Learning for Drag Force Prediction in Dense Fluid-Particulate Systems.

Nikhil Muralidhar1,2, Jie Bu1,2, Ze Cao3

  • 1Department of Computer Science, Virginia Tech, Arlington, Virginia, USA.

Big Data
|October 22, 2020
PubMed
Summary

This study introduces PhyNet, a deep learning model that integrates physics knowledge to improve drag force prediction in fluid dynamics simulations. PhyNet enhances accuracy by incorporating physical principles, addressing data scarcity in complex simulations.

Keywords:
computational fluid dynamicsdata miningmachine learningphysics-guided learning

More Related Videos

Visually Based Characterization of the Incipient Particle Motion in Regular Substrates: From Laminar to Turbulent Conditions
11:51

Visually Based Characterization of the Incipient Particle Motion in Regular Substrates: From Laminar to Turbulent Conditions

Published on: February 22, 2018

9.0K
An Analog Macroscopic Technique for Studying Molecular Hydrodynamic Processes in Dense Gases and Liquids
11:03

An Analog Macroscopic Technique for Studying Molecular Hydrodynamic Processes in Dense Gases and Liquids

Published on: December 4, 2017

8.9K

Related Experiment Videos

Last Updated: Dec 4, 2025

Experimental Measurement of Settling Velocity of Spherical Particles in Unconfined and Confined Surfactant-based Shear Thinning Viscoelastic Fluids
10:28

Experimental Measurement of Settling Velocity of Spherical Particles in Unconfined and Confined Surfactant-based Shear Thinning Viscoelastic Fluids

Published on: January 3, 2014

15.0K
Visually Based Characterization of the Incipient Particle Motion in Regular Substrates: From Laminar to Turbulent Conditions
11:51

Visually Based Characterization of the Incipient Particle Motion in Regular Substrates: From Laminar to Turbulent Conditions

Published on: February 22, 2018

9.0K
An Analog Macroscopic Technique for Studying Molecular Hydrodynamic Processes in Dense Gases and Liquids
11:03

An Analog Macroscopic Technique for Studying Molecular Hydrodynamic Processes in Dense Gases and Liquids

Published on: December 4, 2017

8.9K

Area of Science:

  • Computational fluid dynamics
  • Discrete element method
  • Machine learning

Background:

  • Physics-based simulations are crucial for modeling complex systems like fluid dynamics but often face accuracy limitations due to computational costs and incomplete physical knowledge.
  • Machine learning (ML) can augment simulations by learning from data, but data generation is expensive, necessitating models that handle data scarcity.
  • Incorporating domain-specific physical knowledge into ML model architecture and training can enhance performance and data efficiency.

Purpose of the Study:

  • To develop a deep learning model, PhyNet, that effectively models drag forces on particles in computational fluid dynamics-discrete element method (CFD-DEM) simulations.
  • To leverage physics-guided structural priors and aggregate supervision to improve the accuracy and efficiency of ML models in data-scarce simulation environments.
  • To demonstrate the benefits of integrating physical knowledge into deep learning formulations for scientific modeling.

Main Methods:

  • Proposed PhyNet, a deep learning architecture incorporating physics-guided structural priors.
  • Implemented physics-guided aggregate supervision to constrain the learning process.
  • Conducted extensive experiments for drag force prediction within CFD-DEM simulations.

Main Results:

  • PhyNet demonstrated significant performance improvements compared to state-of-the-art models in drag force prediction.
  • Achieved an average performance improvement of 7.09% by effectively integrating physics knowledge.
  • Validated the effectiveness of physics-guided priors and aggregate supervision in data-limited scenarios.

Conclusions:

  • Integrating physics knowledge into deep learning models is a viable and effective strategy for enhancing scientific simulations.
  • PhyNet offers a robust approach to modeling complex physical phenomena, particularly in data-constrained settings.
  • The proposed method provides a significant advancement in the accuracy of drag force prediction for CFD-DEM applications.