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.

velocity field mistake in two-phase flow simulation with level set method

Please login with a confirmed email address before reporting spam

In two-phase flow simulation, the phase lost is an serious problem in level set method. we found that the velocity field across two-phase interface is mistake when using level set method or phase field method in COMSOL. the example show in COMSOL model library is also wrong. the following picture showed that the velocity is concave across the interface. it should be smooth. why ?



2 Replies Last Post 2021年11月25日 GMT-5 20:36
Mats Nigam COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 3 years ago 2021年10月26日 GMT-4 07:19

Dear Qianlin, If you have problems with conservation, there's most likely an inconsitency in your model setup. Our support team will be happy to help you with the setup of your specific model.

Regarding the capillary filling model, the reason for the variation of the velocity magnitude along the cut line is that the liquid enters the domain radially and is forced around a corner. You can see this in the Arrow Surface 1 plot in the Velocity plot group.

For reference, I have included the corresponding plot of the velocity magnitude along the cut line from a moving-mesh simulation. This is a sharp-interface method, which is more accurate, but you can clearly see the same shape. The magnitude is slightly less because of the no-slip boundary condition at the wall (in the level-set simulation, a Navier-slip condition is applied), but the variation and especially the "concaveness" is larger since the interface is now sharp.

So there is no mistake, there is nothing wrong with the tutorial. This is in a way the beauty with modeling. What you thought is a mistake reveals that even in this simple example, the results are not completely trivial to understand.

Dear Qianlin, If you have problems with conservation, there's most likely an inconsitency in your model setup. Our support team will be happy to help you with the setup of your specific model. Regarding the capillary filling model, the reason for the variation of the velocity magnitude along the cut line is that the liquid enters the domain radially and is forced around a corner. You can see this in the Arrow Surface 1 plot in the Velocity plot group. For reference, I have included the corresponding plot of the velocity magnitude along the cut line from a moving-mesh simulation. This is a sharp-interface method, which is more accurate, but you can clearly see the same shape. The magnitude is slightly less because of the no-slip boundary condition at the wall (in the level-set simulation, a Navier-slip condition is applied), but the variation and especially the "concaveness" is larger since the interface is now sharp. So there is no mistake, there is nothing wrong with the tutorial. This is in a way the beauty with modeling. What you thought is a mistake reveals that even in this simple example, the results are not completely trivial to understand.


Please login with a confirmed email address before reporting spam

Posted: 3 years ago 2021年11月25日 GMT-5 20:36

Dear Mats Thanks for your reply. After your reply, I check the phase conservation of the tutorial during this period. Indeed, the conservation of the model is no problem. However, I found the interface capture method, such as level set and phase field, rely on very fine meshes, small time step and reinitialize parameter, which limit the use in model with large scale, long time simulation or complicate velocity. would you give some guidance or reference about how to use level set in model with large size and long time? Thanks!

Dear Mats Thanks for your reply. After your reply, I check the phase conservation of the tutorial during this period. Indeed, the conservation of the model is no problem. However, I found the interface capture method, such as level set and phase field, rely on very fine meshes, small time step and reinitialize parameter, which limit the use in model with large scale, long time simulation or complicate velocity. would you give some guidance or reference about how to use level set in model with large size and long time? Thanks!

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.