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.
How to assign solution from steady state to a time dependent problem
Posted 6 nov. 2010, 23:59 UTC−4 3 Replies
Please login with a confirmed email address before reporting spam
Hi,
I would like to solve time dependent FSI in the following approach:
1. Solve steady state with BC as defined at t=0
2. Use the solution as the initial condition for time dependent problem.
My problem: I do not know how to apply the solution to step 1 as the initial condition in step 2. Any pointer in this regard will be appreciate.
Regards,
Susant
I would like to solve time dependent FSI in the following approach:
1. Solve steady state with BC as defined at t=0
2. Use the solution as the initial condition for time dependent problem.
My problem: I do not know how to apply the solution to step 1 as the initial condition in step 2. Any pointer in this regard will be appreciate.
Regards,
Susant
3 Replies Last Post 7 nov. 2010, 18:25 UTC−5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
I have added the model to explain my problem.
1. I solved the stationary prolem and assigned the solution to initial value for transient solution
During transient solution, it still shows that the initial value is inconsistant.
Regards,
Susant
I have added the model to explain my problem.
1. I solved the stationary prolem and assigned the solution to initial value for transient solution
During transient solution, it still shows that the initial value is inconsistant.
Regards,
Susant
Attachments:
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I suspect you are actually doing it right. But in your case the problem is not Comsol, but your model definition. The solution you get from steady-state solver is actually wrong. Try as follow to see it:
- use triangular mesh, and decrease the finest mesh even 3 times. So you have a very fine mesh. Then you will see that the problem is your outlet boundary condition which stays fixed, while the adjacent boundary keeps going up. When using large mesh you dont see the problem of inverted mesh in steady-state solver, but with fine mesh you will see.
@edit: and also increase your relative tolerance to 1e-5
Just for the prove of concept I made the outlet boundary condition movable too, and then transient solver was able to continue where st.st. solver left.
To sum up, your initial solution for transient solver is not correct, so transient does not start from final state but before that, and hence starts creating inverted meshes and finally crashes.
--
Comsol 4.0a
Ubuntu 10.04.1
- use triangular mesh, and decrease the finest mesh even 3 times. So you have a very fine mesh. Then you will see that the problem is your outlet boundary condition which stays fixed, while the adjacent boundary keeps going up. When using large mesh you dont see the problem of inverted mesh in steady-state solver, but with fine mesh you will see.
@edit: and also increase your relative tolerance to 1e-5
Just for the prove of concept I made the outlet boundary condition movable too, and then transient solver was able to continue where st.st. solver left.
To sum up, your initial solution for transient solver is not correct, so transient does not start from final state but before that, and hence starts creating inverted meshes and finally crashes.
--
Comsol 4.0a
Ubuntu 10.04.1
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Danial,
Thank you very much for your observation. I would appreciate if you can send me back the corrected program. I would like to study in light of your comments. Perticularly regarding making output also movable.
Regards,
Susant
Thank you very much for your observation. I would appreciate if you can send me back the corrected program. I would like to study in light of your comments. Perticularly regarding making output also movable.
Regards,
Susant