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.

Is COMSOL able to develop User defined elements (zero thickness interface)?

Please login with a confirmed email address before reporting spam

Dear all:

I want to do hydraulic fracturing analysis by COMSOL. At present, a zero thickness interface model was proposed to simulate the fractures explicitly based on FEM. A basic idea is that a pipe like element is embeded between two tranditional volume elements. You can refer to the image given in https://www.google.ca/search?hl=en&biw=1536&bih=734&tbm=isch&sa=1&q=zero+thickness+interface+&oq=zero+thickness+interface+&gs_l=psy-ab.3...5176.5176.0.5443.1.1.0.0.0.0.68.68.1.1.0....0...1.1.64.psy-ab..0.0.0....0.6HdAUDU90qs#imgrc=8PN-9wqwc_P2RM:https://www.google.ca/search?hl=en&biw=1536&bih=734&tbm=isch&sa=1&q=zero+thickness+interface+&oq=zero+thickness+interface+&gs_l=psy-ab.3...5176.5176.0.5443.1.1.0.0.0.0.68.68.1.1.0....0...1.1.64.psy-ab..0.0.0....0.6HdAUDU90qs#imgrc=8PN-9wqwc_P2RM:

This is a really attractive method to simulate fracture propagation. However, COMSOL now cannot combine these two kinds of element into one geometry. Thus, I am wondering whether I could define such a zero thickness interface element by myself, and then incorporate it in COMSOL to do calculation. Would you mind sharing me some ideas about that?

Thank you so much for your help.

Yu

-------------------
Yu Zhang

6 Replies Last Post 14.10.2017, 10:51 GMT-4
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 13.10.2017, 17:19 GMT-4
Updated: 7 years ago 13.10.2017, 17:19 GMT-4

Yes, it is possible to add your own PDEs on lower-dimension entities (such as, in 3D, surfaces, lines and points). Your starting point is to select, in the Model Wizard, Mathematics > PDE Interfaces > Lower dimensions . If you've never used the PDE interfaces before, you'll want to read the chapter on Equation-Based Modeling in the Reference Manual (Chapter 16). A tutorial that demonstrates the use of lower dimension PDE interfaces is this one: https://www.comsol.com/model/transport-and-adsorption-5 .

Best,

Jeff

-------------------
Jeff Hiller
Yes, it is possible to add your own PDEs on lower-dimension entities (such as, in 3D, surfaces, lines and points). Your starting point is to select, in the Model Wizard, Mathematics > PDE Interfaces > Lower dimensions . If you've never used the PDE interfaces before, you'll want to read the chapter on Equation-Based Modeling in the Reference Manual (Chapter 16). A tutorial that demonstrates the use of lower dimension PDE interfaces is this one: https://www.comsol.com/model/transport-and-adsorption-5 . Best, Jeff

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 13.10.2017, 17:35 GMT-4
Updated: 7 years ago 13.10.2017, 17:45 GMT-4

Hi, Dr Jeff:

Thank you so much for your help. I have tried the Low dimensions interface. However, it can only be applied on the lines I drawed in the geometry. What I intended to do is to apply such a PDE on every line of each meshed element (every line used to mesh the whole geometry). It means that after meshing the geometry, lines used for meshing are 1D element, while the area surrounded by these lines are 2D element. By doing so, I can use the 1D element to represent fractures and the 2D element to represent the porous matrix. I am wondering whether COMSOL can achieve that? If not, could I define such a 1D element by my self in COMSOL?

Thank you for your time and help.

Yu

-------------------
Yu Zhang
Hi, Dr Jeff: Thank you so much for your help. I have tried the Low dimensions interface. However, it can only be applied on the lines I drawed in the geometry. What I intended to do is to apply such a PDE on every line of each meshed element (every line used to mesh the whole geometry). It means that after meshing the geometry, lines used for meshing are 1D element, while the area surrounded by these lines are 2D element. By doing so, I can use the 1D element to represent fractures and the 2D element to represent the porous matrix. I am wondering whether COMSOL can achieve that? If not, could I define such a 1D element by my self in COMSOL? Thank you for your time and help. Yu

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 13.10.2017, 17:45 GMT-4

Hi Yu, In COMSOL, equations are always applied to geometric entities. Best, Jeff

-------------------
Jeff Hiller
Hi Yu, In COMSOL, equations are always applied to geometric entities. Best, Jeff

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 13.10.2017, 17:51 GMT-4

Hi Yu, In COMSOL, equations are always applied to geometric entities. Best, Jeff

Thank you for your quick reply. Are there any ways to defined such a 1D element by myself in COMSOL? At present, this has achieved by other codes which are not popular and commercialized. Could I use these codes in COMSOL?

Thank you for your time.

-------------------
Yu Zhang
>Hi Yu, >In COMSOL, equations are always applied to geometric entities. >Best, >Jeff Thank you for your quick reply. Are there any ways to defined such a 1D element by myself in COMSOL? At present, this has achieved by other codes which are not popular and commercialized. Could I use these codes in COMSOL? Thank you for your time.

Jim Freels mechanical side of nuclear engineering, multiphysics analysis, COMSOL specialist

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 14.10.2017, 10:22 GMT-4

Have you considered thin-film approaches similar to what already exists in COMSOL now ? For example in heat trasnfer or structural mechanics there are shell elements that reduce the 3D component down to a surface. You also can program in, or write you own equations directly into the GUI and let COMSOL take care of the element assembly process. To the user, this is similar to specifying a more complex boundary condition. This all works based on FE theory which essentially applies Green's theorem.

-------------------
James D. Freels, Ph.D., P.E.
Have you considered thin-film approaches similar to what already exists in COMSOL now ? For example in heat trasnfer or structural mechanics there are shell elements that reduce the 3D component down to a surface. You also can program in, or write you own equations directly into the GUI and let COMSOL take care of the element assembly process. To the user, this is similar to specifying a more complex boundary condition. This all works based on FE theory which essentially applies Green's theorem.

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 14.10.2017, 10:51 GMT-4

Have you considered thin-film approaches similar to what already exists in COMSOL now ? For example in heat trasnfer or structural mechanics there are shell elements that reduce the 3D component down to a surface. You also can program in, or write you own equations directly into the GUI and let COMSOL take care of the element assembly process. To the user, this is similar to specifying a more complex boundary condition. This all works based on FE theory which essentially applies Green's theorem.

Dear Dr. Freels:

Thank you so much for your reply. The thin-film approach seems to be an attractive method to solve this problem, but I haven't used this functionality yet. Could you please share me more information about it, like where I can find this interface in COMSOL? BTW, is it able to couple the fluid flow process in it?

Thank you for your time and help.

Yu

-------------------
Yu Zhang
>Have you considered thin-film approaches similar to what already exists in COMSOL now ? For example in heat trasnfer or structural mechanics there are shell elements that reduce the 3D component down to a surface. You also can program in, or write you own equations directly into the GUI and let COMSOL take care of the element assembly process. To the user, this is similar to specifying a more complex boundary condition. This all works based on FE theory which essentially applies Green's theorem. Dear Dr. Freels: Thank you so much for your reply. The thin-film approach seems to be an attractive method to solve this problem, but I haven't used this functionality yet. Could you please share me more information about it, like where I can find this interface in COMSOL? BTW, is it able to couple the fluid flow process in it? Thank you for your time and help. Yu

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.