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

Meaning of PISO Corrections

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   February 26, 2017, 14:02
Default Meaning of PISO Corrections
  #1
Member
 
Join Date: Sep 2016
Posts: 43
Rep Power: 9
diggee is on a distinguished road
Hello. I am doing a transient simulation and want to use PISO as the pressure velocity coupling scheme. After reading the theory guide, I understand that the skewness correction is for highly skewed meshes and neighbor correction is to get a more accurate solution. But I couldn't find as to what the values (default 1) of both these fields mean? My mesh has almost little to none skewness. So do I make the skewness correction value as 0? What about the neighbor correction value?
diggee is offline   Reply With Quote

Old   February 26, 2017, 18:37
Default
  #2
Senior Member
 
Lucky
Join Date: Apr 2011
Location: Orlando, FL USA
Posts: 5,668
Rep Power: 65
LuckyTran has a spectacular aura aboutLuckyTran has a spectacular aura aboutLuckyTran has a spectacular aura about
Leave them as the defaults.

I actually can't discern the details either and it's not well documented. The problem is that there are too many things going on and these options are implemented in the style of under-relaxation whereas they are discrete steps in the PISO algorithm. I.e. you do an extra neighbor correction step and an extra skewness correction step so the values should be integers like 0,1,2,3, etc... But that's not the way the menu is set up. It probably has something to do with the AMG solver.

My guess is the corrections are always calculated and you apply an under-relaxation factor. If that is so, then setting neighborCorrection to 0 would turn PISO into SIMPLE. You could probably set the skewness correction to 0, but if it is always calculated then you would not get any speedup, and it would slow down your convergence. So unless someone knows better, I would leave them as they are because they don't hurt.
LuckyTran is offline   Reply With Quote

Old   November 22, 2017, 08:49
Default
  #3
New Member
 
Malthe Haahr Hvelplund
Join Date: Oct 2017
Location: Aalborg, Denmark
Posts: 12
Rep Power: 8
Malthehh is on a distinguished road
Good question diggee - I am in the exact same position as you are.
From that I have read in the user guide. I think LuckyTran is right in what he have described.

Did you experience any difference between SIMPLE or PISO in the convergence time or did you experience any problems when using the default values?

Best Malthe
Malthehh is offline   Reply With Quote

Old   November 22, 2017, 11:21
Default
  #4
Member
 
Join Date: Sep 2016
Posts: 43
Rep Power: 9
diggee is on a distinguished road
Quote:
Originally Posted by Malthehh View Post
Good question diggee - I am in the exact same position as you are.
From that I have read in the user guide. I think LuckyTran is right in what he have described.

Did you experience any difference between SIMPLE or PISO in the convergence time or did you experience any problems when using the default values?

Best Malthe
Hey. I asked this question a long long time ago and I now know what they mean. For PISO, those numbers indicate how many iterations are done for the mesh skewness and neighbor correction (hence only integers are accepted as arguments). If you have very good mesh that has little to no skewness (eg pipe meshed with only hex elements) , you can make the skewness correction 0. If your mesh is highly skewed, you can increment the number as you want. Leave the neighbor correction to 1. Obviously, if you increment these numbers, computational time increases.

SIMPLE is always lighter computationally than PISO but at the same time SIMPLE can run into problems with a skewed mesh. Hence the convergence time with each method really depends on the problem setup. If you have a simple geometry and simple flow physics, go ahead with SIMPLE.
diggee is offline   Reply With Quote

Old   November 22, 2017, 12:11
Default
  #5
Senior Member
 
Lucky
Join Date: Apr 2011
Location: Orlando, FL USA
Posts: 5,668
Rep Power: 65
LuckyTran has a spectacular aura aboutLuckyTran has a spectacular aura aboutLuckyTran has a spectacular aura about
Cool! That clears up a lot! Yes, they should be integers. So the interface is just oddly setup and documentation not so good.

Quote:
Originally Posted by diggee View Post
SIMPLE is always lighter computationally than PISO but at the same time SIMPLE can run into problems with a skewed mesh. Hence the convergence time with each method really depends on the problem setup. If you have a simple geometry and simple flow physics, go ahead with SIMPLE.
For complex (transient) problems, SIMPLE can also be better than PISO for the same reason. SIMPLE is faster per iteration than PISO because it doesn't solve the final pressure correction which takes a long time. If you have strong coupling between momentum and other equations, say energy equation due to strongly varying properties or other transport equations, running SIMPLE can save you a lot of time over PISO. The thing is PISO is good for solving the P-V coupling problem in 1 super-expensive step. But if you are coupling with other equations then you need to do many sweeps anyway.

Last edited by LuckyTran; November 22, 2017 at 13:22.
LuckyTran is offline   Reply With Quote

Old   November 28, 2017, 09:07
Default
  #6
New Member
 
Malthe Haahr Hvelplund
Join Date: Oct 2017
Location: Aalborg, Denmark
Posts: 12
Rep Power: 8
Malthehh is on a distinguished road
Thanks a lot for your quick response LuckyTran and Diggee. It helped on the understanding.

Best Malthe
Malthehh is offline   Reply With Quote

Reply

Tags
piso


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
SIMPLE vs. PISO Kappa OpenFOAM Running, Solving & CFD 1 October 17, 2016 08:08
Understanding the PISO loop of icoFoam ooo OpenFOAM Programming & Development 0 March 14, 2014 09:28
Non-linearity Pressure Equation -- PISO algorithm gdeneyer OpenFOAM Programming & Development 1 August 23, 2012 05:19
SIMPLE and PISO Aditya Main CFD Forum 5 April 1, 2006 18:52
PISO vs. SIMPLE benedikt flurl Main CFD Forum 2 April 14, 2005 06:54


All times are GMT -4. The time now is 23:38.