Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Need some advise for Boundary Condition and Iterative method
Posted 9 sept. 2010, 11:12 UTC−4 3 Replies
Please login with a confirmed email address before reporting spam
Hi everyone. I have started using COMSOL for sometime. I have some questions about its boundary condition and solver.
1. My test case is 2D internal fin Fluid flow+Heat transfer+ FSI+transient , and the Direct solver (UMFPACK) is used.
When I am defining the inlet velocity of 0.001m/s, the it is converged and shows nice results.
However, if I increase the velocity inlet up to 0.1 m/s, the monitor shows that " Fail to find consistent initial condition_Last time step is not converged."
2. Next problem, I am defining the inlet velocity of 0.001m/s
,and I wanna switch the solver from Direct method to Iterative method.
This time I used the GMRES and FGMRES.
Therefore, the monitor shows that "Fail to find consistent initial condition_ Matrix_has_zero_on_diagonal and Fail to find consistent initial condition _Last time step is not converged."
Does anyone has suggestion for me? Thank you very much in advance.
Suabsakul
1. My test case is 2D internal fin Fluid flow+Heat transfer+ FSI+transient , and the Direct solver (UMFPACK) is used.
When I am defining the inlet velocity of 0.001m/s, the it is converged and shows nice results.
However, if I increase the velocity inlet up to 0.1 m/s, the monitor shows that " Fail to find consistent initial condition_Last time step is not converged."
2. Next problem, I am defining the inlet velocity of 0.001m/s
,and I wanna switch the solver from Direct method to Iterative method.
This time I used the GMRES and FGMRES.
Therefore, the monitor shows that "Fail to find consistent initial condition_ Matrix_has_zero_on_diagonal and Fail to find consistent initial condition _Last time step is not converged."
Does anyone has suggestion for me? Thank you very much in advance.
Suabsakul
Attachments:
3 Replies Last Post 11 sept. 2010, 15:42 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
have you tried to increase gently the velocity in a parametrical sweep, and then perhpas save the model to use it as a starting point/initial conditions for next analysis ?
and if you start with 0.1 m/s velocity, perhaps you could define initial values more realistic as the default "0"
--
Good luck
Ivar
have you tried to increase gently the velocity in a parametrical sweep, and then perhpas save the model to use it as a starting point/initial conditions for next analysis ?
and if you start with 0.1 m/s velocity, perhaps you could define initial values more realistic as the default "0"
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
To Ivar . Thanks for ur sugguestion. I will try to increase the velocity slightly. Also, I am gonna change the initial condition.
Best Regards,
Suabsakul
Best Regards,
Suabsakul
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
an then there is the issue of the pressure, I have notived that I get often better convergenve with an outlet rather in "pressure no viscuous stress", than just "non viscous stress", the question is then what realistic pressure to use. (all for constant velocity input "inlet")
--
Good luck
Ivar
an then there is the issue of the pressure, I have notived that I get often better convergenve with an outlet rather in "pressure no viscuous stress", than just "non viscous stress", the question is then what realistic pressure to use. (all for constant velocity input "inlet")
--
Good luck
Ivar