Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Modelling 2D and 3D porous periodic/random microstructure geometry

Please login with a confirmed email address before reporting spam

Hello!

I am working on modelling 2D and 3D microstructure geometry in Comsol. I would have to model both periodic and random micro structures. I intend to utilize the Voronoi algorithm to impleemnt the same. As of now, I have no clue as to how I could go about implementing the same in Comsol?

Can any one provide me suggestions on how this could be accomplished?

8 Replies Last Post 23 févr. 2011, 02:59 UTC−5

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17 févr. 2011, 08:44 UTC−5
You may want to have a look here: A paper on modelling of fiber composites www.springerlink.com/content/wx185g5301300521/fulltext.pdf

We used COMSOL to model a fiber reinforced unit cell (specifically, it was a hexagonal unit cell). Indeed with periodic boundary conditions and in 3-D. A good thing is that we have an analytical condition for randomly distributed microstructures...

Hope it will help.

Stephan
You may want to have a look here: A paper on modelling of fiber composites http://www.springerlink.com/content/wx185g5301300521/fulltext.pdf We used COMSOL to model a fiber reinforced unit cell (specifically, it was a hexagonal unit cell). Indeed with periodic boundary conditions and in 3-D. A good thing is that we have an analytical condition for randomly distributed microstructures... Hope it will help. Stephan

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17 févr. 2011, 09:30 UTC−5
Hello Stephan.

Thank you very much for your response.

But I must inform you that I am an amateur user of Comsol posed with the challenge of preparing 2D/3D periodic and random microstructures in a very short duration. My initial idea has been to utilize the Voronoi algorithm for generating random micro structure and also a similar means to generate a more complicated but nevertheless periodic geometry.

Thus my struggle is with determining how I go about building the model i.e. do I use the MATLAB interface to write the algorithm and generate the geometries in COMSOL, since it seems practically impossible to do it just by using the draw mode in COMSOL.

Since I have so little familiarity with COMSOL, it's rather hard to decode the COMSOL commands from the geometry file of sample models. Thus I intend to know if I am on the right track in thinking that the way forward is to use MATLAB interface to do the desired geometry? If yes, do you have any hints on this?

Thanks for your time.
Regards,
Hello Stephan. Thank you very much for your response. But I must inform you that I am an amateur user of Comsol posed with the challenge of preparing 2D/3D periodic and random microstructures in a very short duration. My initial idea has been to utilize the Voronoi algorithm for generating random micro structure and also a similar means to generate a more complicated but nevertheless periodic geometry. Thus my struggle is with determining how I go about building the model i.e. do I use the MATLAB interface to write the algorithm and generate the geometries in COMSOL, since it seems practically impossible to do it just by using the draw mode in COMSOL. Since I have so little familiarity with COMSOL, it's rather hard to decode the COMSOL commands from the geometry file of sample models. Thus I intend to know if I am on the right track in thinking that the way forward is to use MATLAB interface to do the desired geometry? If yes, do you have any hints on this? Thanks for your time. Regards,

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17 févr. 2011, 09:37 UTC−5
I used Matlab with COMSOL to parametrize the geometry. So I warmly recommend you to do so, especially if you are going to make use of the complicated algorithms that are not included in the commercial code.

BTW, COMSOL does have parametrization of geometry in GUI so you may try to use it. It may help with boundary(edge) numbering at some stage.

Good luck!
Stephan
I used Matlab with COMSOL to parametrize the geometry. So I warmly recommend you to do so, especially if you are going to make use of the complicated algorithms that are not included in the commercial code. BTW, COMSOL does have parametrization of geometry in GUI so you may try to use it. It may help with boundary(edge) numbering at some stage. Good luck! Stephan

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17 févr. 2011, 09:54 UTC−5
Hello Stephan

Thanks for your immediate response.
I am glad that atleast I am on the right track. What remains is the big struggle to pull it off. Will strat work on it immediately.

Regards,
Hello Stephan Thanks for your immediate response. I am glad that atleast I am on the right track. What remains is the big struggle to pull it off. Will strat work on it immediately. Regards,

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22 févr. 2011, 23:22 UTC−5

Hello!

I am working on modelling 2D and 3D microstructure geometry in Comsol. I would have to model both periodic and random micro structures. I intend to utilize the Voronoi algorithm to impleemnt the same. As of now, I have no clue as to how I could go about implementing the same in Comsol?

Can any one provide me suggestions on how this could be accomplished?


You might be interested to view the below youtube video and the paper:
www.youtube.com/watch?v=12PahzCUrAI
portal.acm.org/citation.cfm?id=1841992.1842341

After designing the geometry in SolidWorks, it will be trivial to use COMSOL to apply loads and constraints, and finally the FEA results. Hope this is useful.
[QUOTE] Hello! I am working on modelling 2D and 3D microstructure geometry in Comsol. I would have to model both periodic and random micro structures. I intend to utilize the Voronoi algorithm to impleemnt the same. As of now, I have no clue as to how I could go about implementing the same in Comsol? Can any one provide me suggestions on how this could be accomplished? [/QUOTE] You might be interested to view the below youtube video and the paper: http://www.youtube.com/watch?v=12PahzCUrAI http://portal.acm.org/citation.cfm?id=1841992.1842341 After designing the geometry in SolidWorks, it will be trivial to use COMSOL to apply loads and constraints, and finally the FEA results. Hope this is useful.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 févr. 2011, 02:39 UTC−5
Hello David!

Thank you very much for the links. it will prove to be mighty helpful.

Regards,
Hello David! Thank you very much for the links. it will prove to be mighty helpful. Regards,

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 févr. 2011, 02:51 UTC−5
Yep, very impressive.

But how would we run these for different parameters. Say material volume fractions. should I go to SolidWorks every time I need to change this and then upload the geometry to a FE and then new PBC etc...?

I still think that this should be done in Matlab.

Stephan
Yep, very impressive. But how would we run these for different parameters. Say material volume fractions. should I go to SolidWorks every time I need to change this and then upload the geometry to a FE and then new PBC etc...? I still think that this should be done in Matlab. Stephan

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 févr. 2011, 02:59 UTC−5
Hello Stephan.

If one is modelling with solid works, the only alternative seems to be to return to Solidworks everytime you want to change parameters. It definitely is not convenient and I am not certain how flexible Solidworks is with regards to altering parameters.

I did manage to complete 2D modelling of periodic and random micro structures using Voronoi tesselations in Matlab and adapted it for Comsol but as of now it isn't a very elegant solution that I have. I haven't been able to parametrize it to the extent that I would have wanted to.

Do you have any suggestions on the choice of parameters that could be used for modelling in Matlab?

Regards,
Hello Stephan. If one is modelling with solid works, the only alternative seems to be to return to Solidworks everytime you want to change parameters. It definitely is not convenient and I am not certain how flexible Solidworks is with regards to altering parameters. I did manage to complete 2D modelling of periodic and random micro structures using Voronoi tesselations in Matlab and adapted it for Comsol but as of now it isn't a very elegant solution that I have. I haven't been able to parametrize it to the extent that I would have wanted to. Do you have any suggestions on the choice of parameters that could be used for modelling in Matlab? Regards,

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.