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.

regarding level set method: evaluation/ calculation of optimum value of gamma (reinitialization factor)

Please login with a confirmed email address before reporting spam

hi,
would anyone of you tell me how to evaluate or calculate the value of gamma (reinitialization factor) before starting the initialization of the level set problem? does comsol has some inbuilt feature or some procedure for this? would somebody tell me some references/ papers to study about this parameter and its importance? if the initial conditions are zero velocity (in my case:- rising bubble model) how to determine the value of gamma if the domain is decreased to e-6 meter scale?

thanks in advance for your reply.

6 Replies Last Post 31 ott 2016, 02:36 GMT-4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 apr 2010, 14:32 GMT-4
Hi

it is very tricky to calculate gamma, but there are some hints in the comsol doc., which are still very little to sort it out!

Regards
Naveed
Hi it is very tricky to calculate gamma, but there are some hints in the comsol doc., which are still very little to sort it out! Regards Naveed

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 apr 2010, 02:32 GMT-4
thanks naved for your reply. i have already read those documentation. they only mention that the maximum velocity in the simulation to be taken as the value of gamma. how to calculate that value before hand? moreover its calculation may be affected by the choice of the value of gamma. so which value of velocity and gamma is correct, can't be said easily without doing so many simulations. it is just like hit and trial method.
if anybody knows any rule of thumb or procedure to calculate gamma, please tell.

anyways, thanks naved.
thanks naved for your reply. i have already read those documentation. they only mention that the maximum velocity in the simulation to be taken as the value of gamma. how to calculate that value before hand? moreover its calculation may be affected by the choice of the value of gamma. so which value of velocity and gamma is correct, can't be said easily without doing so many simulations. it is just like hit and trial method. if anybody knows any rule of thumb or procedure to calculate gamma, please tell. anyways, thanks naved.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 apr 2010, 04:37 GMT-4
Hi Dhiraj

You are absolutely right! I have done many simulations with just hit a trial values of gamma but still my drpolet moves down instead of going forward. I think COMSOL should provide elaborative procedure for epsilon and gamma. There is no way to spend hours on just fixing gamma and epsilon.

Regards
Naveed
Hi Dhiraj You are absolutely right! I have done many simulations with just hit a trial values of gamma but still my drpolet moves down instead of going forward. I think COMSOL should provide elaborative procedure for epsilon and gamma. There is no way to spend hours on just fixing gamma and epsilon. Regards Naveed

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 apr 2010, 05:27 GMT-4
hi naveed,

i have done some experiment to evaluate gamma and i would like to share it with you. you might get benefited.
i have done the first simulation with some guess value of gamma. in the post-processing, i have selected the r or z value of velocity or velocity profile can be selected. this will display the min and max of velocity through out the domain. try to see the results at various times using different time in general menu of post-processing.
from there, after seeing the trend select some good value near to the maximum value of velocity. repeat the procedure.

hope it may help you.
hi naveed, i have done some experiment to evaluate gamma and i would like to share it with you. you might get benefited. i have done the first simulation with some guess value of gamma. in the post-processing, i have selected the r or z value of velocity or velocity profile can be selected. this will display the min and max of velocity through out the domain. try to see the results at various times using different time in general menu of post-processing. from there, after seeing the trend select some good value near to the maximum value of velocity. repeat the procedure. hope it may help you.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 apr 2010, 06:56 GMT-4
Hi Dhiraj

Do we need to re-initialize each time during this iteration? Or shall we go on with the iterative procedure without initialization again and again? Have you ever come across with "inconsient value error" during selection of gamma. By the way is there also some trick for selecting "epsilon"-as far as I know, it should be the maximum mesh size, h.

I worked with droplet movement in two-phase flow (gas-liquid). Thanks for sharing your experience.

Regards
Naveed
Hi Dhiraj Do we need to re-initialize each time during this iteration? Or shall we go on with the iterative procedure without initialization again and again? Have you ever come across with "inconsient value error" during selection of gamma. By the way is there also some trick for selecting "epsilon"-as far as I know, it should be the maximum mesh size, h. I worked with droplet movement in two-phase flow (gas-liquid). Thanks for sharing your experience. Regards Naveed

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 31 ott 2016, 02:36 GMT-4
Hi
I am working on electrowetting technique where droplet has to move from one surface to another. the droplet are taking around 20ms time to reach another surface for gamma=1 but i want that it should take atleast 40 ms. In that case, what changes should i make. should i increase or decrease the reinitialization parameter in level set. Moreover, is it possible to couple phase field and level set ?
Hi I am working on electrowetting technique where droplet has to move from one surface to another. the droplet are taking around 20ms time to reach another surface for gamma=1 but i want that it should take atleast 40 ms. In that case, what changes should i make. should i increase or decrease the reinitialization parameter in level set. Moreover, is it possible to couple phase field and level set ?

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.