|
[Sponsors] |
February 12, 2013, 04:10 |
Convergence/Residuals Ending
|
#1 |
Member
VS
Join Date: Nov 2012
Posts: 86
Rep Power: 13 |
Is there any way to stop automatically the solution when our variables reach convergence?
For example, given only your dT and not endTime, the solution stops when it reaches convergence. Thanks |
|
February 12, 2013, 04:19 |
|
#2 | |
Senior Member
|
Hi,
you can add to your fvSolution file the following lines; Quote:
Regards Marco |
||
February 12, 2013, 04:53 |
|
#3 |
Member
VS
Join Date: Nov 2012
Posts: 86
Rep Power: 13 |
Thanks man!
|
|
March 4, 2013, 10:41 |
|
#4 |
Member
VS
Join Date: Nov 2012
Posts: 86
Rep Power: 13 |
Marco,
i had problems in controlling the residuals using icoFoam. I added the lines you suggested but nothing happens. solvers { p { solver PCG; preconditioner DIC; tolerance 1e-04; relTol 0; } U { solver PBiCG; preconditioner DILU; tolerance 1e-05; relTol 0; } } PISO { nCorrectors 2; nNonOrthogonalCorrectors 0; pRefCell 0; pRefValue 0; residualControl { p 1e-3; U 1e-3; } } I put 1e-03 to check the case and i plot initial residuals on-the-fly. Both u and p pass this limit but the solution continues, it doesn't stop. (I use OpenFoam 2.1.1) |
|
March 4, 2013, 10:47 |
|
#5 |
Senior Member
|
Hi,
do you wanna simulate a steady or a transient simulation? icoFoam solver is inherently transient! So it does not make sense to stop the solution to a steady state! Regards Marco |
|
March 4, 2013, 11:42 |
|
#6 |
Member
VS
Join Date: Nov 2012
Posts: 86
Rep Power: 13 |
I'm studying the flow over a sphere using icoFoam, which has a transient part 'till the distribution around the sphere reaches a "steady-state", meaning it doesn't change no matter how much time passes from there on.
So, now that i know that ,when my initial residuals have reached e.g. 1e-05 for p, my solution has converged and i don't need to have it running any more, i want it to stop automatically when reaching this values. Thanks for all your replies,really |
|
June 3, 2014, 05:53 |
|
#7 | |
Member
Sami
Join Date: Nov 2012
Location: Cap Town, South Africa
Posts: 87
Rep Power: 13 |
Quote:
Did you solve your problem ? I'm trying to do the same thing. Thank you |
||
June 14, 2014, 04:48 |
|
#8 |
Member
VS
Join Date: Nov 2012
Posts: 86
Rep Power: 13 |
Hello Mehrez,
as said above you should add the following lines in fvSolution. residualControl { p 1e-06; U 1e-06; } However, I would recommend this approach only in cases where you know for sure that the flow field is fully developed and no other alteration is going to occur. This comes with experience. Kind regards |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[Other] TurboGrid - problem with trailing edges ending with hub/shroud curve | TTM | ANSYS Meshing & Geometry | 8 | February 16, 2020 07:11 |
Ending condition | Julien Pralong | FLOW-3D | 1 | January 8, 2009 18:09 |
FoamX error aachenBomb case | Ervin Adorean (Adorean) | OpenFOAM Pre-Processing | 13 | March 7, 2005 03:50 |