General question about multiphysics (coupled heat/fluid analysis)

[bjorn_fallqvist]

Hello,

I have a general question regarding the coupled transient heat and fluid analysis. If I have understood this correctly, the solver first solves the temperature problem, and then proceeds to solve the fluid problem with the results acquired. After this, the next iteration is started.

My question is: How come there are settings for controlling the convergence tolerance, relaxation parameters etc. both in the general "Coupled analysis settings" dialog box and the fluid and heat transfer settings dialogue box also? If I set the relaxation parameter in the general settings dialog box, does this override what is currently used in the dialogue box specific for the analysis type?

[Joey.X]

Sound like you are talking about two-way coupling fluid/thermal analysis, you are right that the thermal starts first and then multiple nonlinear iterations needed until convergence.

The convergence tolerance is under the “analysis parameter” window/ check custom load-stepping setting/temp Tol and velocity Tol  and the relaxation is there as well,  for two-way coupled fluid/thermal analysis, there are also “fluid analysis parameter windows” and “thermal analysis parameter windows”, but there are no temperature and velocity tolerance and relaxation inputs.

Jianhui

[bjorn_fallqvist]

What about the "relaxation parameter for nonlinear iteration" under the "Unsteady fluid flow window"? What does that parameter control, then? Likewise, there is also a relaxation parameter under the Transient Heat Transfer analysis window within the two-way heat/fluid analysis.

[Joey.X]

bjorn,

- The "relaxation parameter for nonlinear iteration" under the "Unsteady fluid flow window"controls the inner ineration in fluid part;

- The The convergence tolerance is under the “analysis parameter” window controls the overall fluid and thermal nonlinear iterations;

-  There are no relaxation parameter under the Transient Heat Transfer analysis window

I agree that this sounds not consistency and it needs to be improved.

Thanks,

Jianhui

[bjorn_fallqvist]

Hi, Joey.

Thanks for the answer. However, there is a relaxation parameter for the thermal analysis under "advanced", but since the check box is not checked by default, I guess it doesn't really matter unless you account for the non-linear effects?

Björn