CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT > Fluent Multiphase

Approach to simulate millions of particles in a room

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree4Likes
  • 1 Post By vinerm
  • 1 Post By vinerm
  • 1 Post By vinerm
  • 1 Post By vinerm

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   January 29, 2020, 03:59
Default Approach to simulate millions of particles in a room
  #1
Member
 
MMatt
Join Date: May 2013
Posts: 59
Rep Power: 9
MMatt is on a distinguished road
I cannot decide which approach to use in order to simulate particles in a room. I've tried an Eulerian-Eulerian approach, but due to the number of particles (millions), it is not really computer efficient. I wanted to know if there was an other method I could use. Something maybe more statistical than deterministic. Such as in the video below (from 1min17):

https://www.youtube.com/watch?v=qULwdb277Mc

My aim is to see the particles concentration such as on the video. I don't need to see each particles, but their concentration in the fluid domain.

Thank you!
MMatt is offline   Reply With Quote

Old   January 29, 2020, 04:22
Default Euler-Euler is statistical
  #2
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 31
vinerm will become famous soon enough
Hi

Euler-Euler or Mixture models are statistical. Lagrangian on the other hand is a mix; it uses deterministic approach to predict the positions of parcels, however, each parcel represents multiple particles. There are at least two thing that you need to look at to decide if it is Euler-Euler (including PBM) or DPM that is more suitable.

1. Volume fraction of particles in the room. Less than 10-12%, you can use DPM and you should use DPM. More than that, you may use both but DPM would require additional collision model, such as DEM.
2. What's the objective of the simulation?
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   January 29, 2020, 04:30
Default
  #3
Member
 
MMatt
Join Date: May 2013
Posts: 59
Rep Power: 9
MMatt is on a distinguished road
Hi,

Thank you for your answer. Volume fraction is very very small (less than 1%), and what I'm interested to see is the evolution of the particles concentration in the fluid domain relative to time.

Thanks!
MMatt is offline   Reply With Quote

Old   January 29, 2020, 04:31
Default DPM is the way to go
  #4
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 31
vinerm will become famous soon enough
Then use DPM. And calculate Stokes number. If it is less than 1, just use one-way coupling with fluid.
MMatt likes this.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   January 29, 2020, 04:50
Default
  #5
Member
 
MMatt
Join Date: May 2013
Posts: 59
Rep Power: 9
MMatt is on a distinguished road
Thank you! However, two-way coupling is enabled by default and I cannot disable it (interaction with continuous phase).
MMatt is offline   Reply With Quote

Old   January 29, 2020, 04:54
Default
  #6
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 31
vinerm will become famous soon enough
Two-way coupling is always user's decision. You might have some other model enabled that requires two-way coupling, such as unsteady particle tracking. Do note that particle tracking is always unsteady because the equation has only one independent variable, time. Unsteady particle tracking is required when there is some unsteady phenomenon, such as coalscence or break-up, that need to be modeled. Otherwise, you can disable the Interaction. But do not disable it if St number is close to or greater than 1.
MMatt likes this.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   January 30, 2020, 04:22
Default
  #7
Member
 
MMatt
Join Date: May 2013
Posts: 59
Rep Power: 9
MMatt is on a distinguished road
After more reading, I have noticed the Mixture model. However I am not sure if it would be appropriate to my case (my initial feeling is that it is not). I need to see particles accumulation in my fluid domain, therefore some of the particles could have a velocity of 0 or at least a very low velocity. Does that mean the Mixture model cannot be used? I know the question may sound stupid tho

EDIT: Moreover, am I right in saying DPM would be more appropriate to my case than DDPM? The particle conccentration is very low (<1%).

EDIT2: I do not need to simulate collision between particles, as their diameter is very small (sub 1um in diameter), therefore the particules could be considered as points, but they are NOT massless. If that makes any sense.
MMatt is offline   Reply With Quote

Old   January 30, 2020, 05:44
Default Mixture can be used but do not use
  #8
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 31
vinerm will become famous soon enough
Mixture model can be used but not very applicable, particularly when DPM is applicable. Since you wish to observe the accumulation, you have to simulate two-way coupled, Unsteady particle tracking otherwise you will not observe it. Use DPM; DDPM not required.
MMatt likes this.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   January 30, 2020, 07:08
Default
  #9
Member
 
MMatt
Join Date: May 2013
Posts: 59
Rep Power: 9
MMatt is on a distinguished road
Thank you!

One last question (didn't want to open a new thread), I cannot find the function to initialize the fluid domain with a defined number of particles. My plan was to implement them through an injection (a random concentration in the fluid domain is fine in my case, I just need the correct number of particles), but I've read there is a "volume injection" function (last message):

https://studentcommunity.ansys.com/t...esidence-time/

However I cannot find it.
MMatt is offline   Reply With Quote

Old   January 30, 2020, 08:34
Default Beta feature
  #10
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 31
vinerm will become famous soon enough
Volume Injection is a beta feature. Enable it using

def bfa yes

Recommended would be to use File Injection.
MMatt likes this.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Problem with DPM simulation with particles injection and EXECUTE_AT_THE_END UDF. Ari Fluent UDF and Scheme Programming 4 May 31, 2016 08:51
Add lagrangian particles to OpenFoam solver luchen2408 OpenFOAM 0 June 2, 2015 03:10
How to define particles attachement to the wall in eulerian-eulerian approach? mahdim CFX 0 April 8, 2015 18:00
particles model ati_ros61 FLOW-3D 3 December 6, 2009 16:03
how to determine the number of particles injected. welch FLUENT 1 November 1, 2005 05:11


All times are GMT -4. The time now is 16:05.