search

KratosMultiphysics / Kratos

Kratos Multiphysics (A.K.A Kratos) is a framework for building parallel multi-disciplinary simulation software. Modularity, extensibility and HPC are the main objectives. Kratos has BSD license and is written in C++ with extensive Python interface.
https://kratosmultiphysics.github.io/Kratos/
Other
1.04k stars 246 forks source link

[FluidDynamicsApplication] CFD --> REACTION #6009

Closed soudah closed 2 years ago

soudah commented 4 years ago

Dear All,

In collaboration with @rubenzorrilla, we are using the REACTION forces to compute the Wall Shear Stress(WSS) over a cilindrical tube, mainly thinking on Biomedical problems. Next image shows the | REACTION |.

REACTION

Next image shows the NORMAL

imagen

Then, we use " VariableRedistributionUtility::DistributePointValues(rModelPart, REACTION, FACE_LOAD, tolerance, max_it)" to project the reactions over the nodes of the mesh, getting:

imagen

Based on that projection (FACE LOAD), the Wall Shear Stress is computed.

imagen

In order to try to improve the results, @rubenzorrilla told me that I had to decrease the tolerance of the solver. Right now, the parameters of the solver are:

    "relative_velocity_tolerance" : 0.000000001,
    "absolute_velocity_tolerance" : 1e-9,
    "relative_pressure_tolerance" : 0.000000001,
    "absolute_pressure_tolerance" : 1e-9,

The results showed are using these parameters, so my question is:

How we can improve the results? How we can avoid this marbled effect? Any ideas?

Thank you very much,

p.D: The implementation is done in: FluidDynamicalBiomedicalApplication. This application derives from FluidDynamicsApplication.

@rubenzorrilla @RiccardoRossi @jcotela @philbucher @KratosMultiphysics/fluid-dynamics @KratosMultiphysics/fluid-maintainers

RiccardoRossi commented 4 years ago

do you have any sort of regularity in the thickness of the first element layer in the vicinity of the boundary?

soudah commented 4 years ago

Thanks Riccardo.

These are the results using a "strutured and center-strutured mesh".

imagen

Right now I will try with the same mesh + Boundary layer.

soudah commented 4 years ago

These are the results with Boundary Layer. The results improve a lot, but there are some peak values on the WSS. I am reviwing how the WSS is computed. But WSS is just only the norm of projection of the face_Load over the elements.

imagen

rubenzorrilla commented 4 years ago

Considering that the REACTION, which is converted to FACE_LOAD, and the NORMAL are smooth, I dare say that the problem is in the WSS calculation.

philbucher commented 4 years ago

can this be closed?

Virginiasam commented 4 years ago

Hi All, I am also interested in computing WSS. Do you have this i1mplemented to share? I am currently testing Kratos for microfluidics applications. Thks!

rubenzorrilla commented 4 years ago

We plan to retake this during the upcoming weeks. @Virginiasam we will let you know as soon as we merge the changes. 😉

Virginiasam commented 4 years ago

Thank you @rubenzorrilla ! Looking forward

rubenzorrilla commented 2 years ago

Closing as it seems to not be an issue anymore.

As a note for the record, I'd like to add that all the WSS-like capabilities are now collected in the FluidDynamicsBiomedicalApplication.