|
[Sponsors] |
September 30, 2015, 07:17 |
Alternative way of particle tracking
|
#1 |
Member
Join Date: Jul 2015
Location: Aalborg
Posts: 83
Rep Power: 10 |
Hi,
I want to track particles in an incompressible flow (big system) and try to get an overview about the smartest way of doing that for my case. I know that there are solvers to track particles, but they need time and computer power. My idea is the following: - The Stokes number is very small in my case (0.001) which means in my opinion, the particles follow the flow close to perfect. - I am not interested in the forces on a particle or something like that, only on it's path Following that logic, it should be a fair assumption to use the stream tracker (assuming the particles follow the stream lines), put e.g. 1000 streamlines at the inlet and get x,y and z values of each stream line. Of course I tried that already: After inserting a stream-tracker, one clicks on the stream-tracker and goes to File --> Save Data, as CSV for example... What do you thing about this idea? |
|
October 2, 2015, 03:40 |
|
#2 |
Senior Member
Gerhard Holzinger
Join Date: Feb 2012
Location: Austria
Posts: 339
Rep Power: 28 |
Why don't you track 1000 particles which perfectly follow the fluid?
There are two implementations of Lagrangian particles in OpenFOAM. The more the basic implementation is the solidParticle and the advanced one is in the intermediate library (based on the parcel class). You can create a clone of solidParticle by deriving a class from the class particle (which is the base class of both solidParticle and all the parcel-based classes) and tell your new particle class to follow the fluid. You do this by prescribing the particle's velocity in the move() method. Study the solidParticle class and throw out everything you don't need, e.g. calculation of drag. As long as the number of particles you want to track is low, the additional computational cost will be low. |
|
October 2, 2015, 04:10 |
|
#3 |
Member
Join Date: Jul 2015
Location: Aalborg
Posts: 83
Rep Power: 10 |
Thank you for your answer!
I am doing something similar in small case, but in my case I need 1000000 particles in one m³.... so it might make life a lot easier if, I can directly use streamlines - if that is possible... |
|
October 2, 2015, 04:25 |
|
#4 |
Senior Member
Gerhard Holzinger
Join Date: Feb 2012
Location: Austria
Posts: 339
Rep Power: 28 |
Ok, I did read from your original post that you need to use such a large number of particles. I have also noticed that the larger the number of particles is, the faster computation times explode.
|
|
October 5, 2015, 03:05 |
|
#5 |
Member
Join Date: Jul 2015
Location: Aalborg
Posts: 83
Rep Power: 10 |
Hi,
for everyone who is interested: The columns of x,y and z direction are called "Points:0", "Points:1" and "Points:2". Best Gerrit |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
dispersion model with lagragian particle tracking model for incompressible flows | eelcovv | OpenFOAM Running, Solving & CFD | 54 | April 10, 2018 09:36 |
Lagrangian Particle Tracking in Eulerian-Eulerian Multiphase Flow | DarrenC | CFX | 5 | April 7, 2016 14:50 |
Ubuntu 12.10 + openfoam2.2.0 ==> paraview error message | peteryuan | OpenFOAM Installation | 6 | August 18, 2013 18:00 |
[OpenFOAM] ParaView ErrOr | soheil nazmdeh | ParaView | 1 | August 17, 2013 07:40 |
injection problem | Mark New | FLUENT | 0 | August 4, 2013 01:30 |