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.
Confusion with Boundary Conditions for PDE Coefficient Form
Posted Jan 11, 2010, 2:41 p.m. EST 1 Reply
Please login with a confirmed email address before reporting spam
Hello,
I have 6 PDEs with 6 unknowns and have successfully put them into COMSOL using the general PDE mode in coefficient form, but I am getting confused with setting the boundary conditions. Some of the 6 dependent variables appear in multiple equations, so I am wondering, if I set a variable on a boundary equal to a constant for one equation, will I again need to set it for the other equations on that boundary?
For example, two of my equations look like:
[1] dxi(c*p*u_i)=L
[2] dxi(c*(1-p)*h_i)= -L
where dxi=d/dxi, and is an index, i=1,2,3
So if I set p=0 on a boundary for equation 1, is it considered 0 for all the equations in the system, or do I need to specify it again for equation 2?
Also, my problem was easier to formulate in coefficient form, but my user manual mentions that the general form should be used for nonlinear problems (which mine is). It does mention that you can choose a different solution form in the solver parameters window, so if you formulate a problem in coefficient it can be solved in the general form, but I was wondering if it's really as simple as selecting "general" in the solution form box before solving. Should I be worried about anything else?
Thanks, and please let me know if you need more information or clarification of my confusion.
I have 6 PDEs with 6 unknowns and have successfully put them into COMSOL using the general PDE mode in coefficient form, but I am getting confused with setting the boundary conditions. Some of the 6 dependent variables appear in multiple equations, so I am wondering, if I set a variable on a boundary equal to a constant for one equation, will I again need to set it for the other equations on that boundary?
For example, two of my equations look like:
[1] dxi(c*p*u_i)=L
[2] dxi(c*(1-p)*h_i)= -L
where dxi=d/dxi, and is an index, i=1,2,3
So if I set p=0 on a boundary for equation 1, is it considered 0 for all the equations in the system, or do I need to specify it again for equation 2?
Also, my problem was easier to formulate in coefficient form, but my user manual mentions that the general form should be used for nonlinear problems (which mine is). It does mention that you can choose a different solution form in the solver parameters window, so if you formulate a problem in coefficient it can be solved in the general form, but I was wondering if it's really as simple as selecting "general" in the solution form box before solving. Should I be worried about anything else?
Thanks, and please let me know if you need more information or clarification of my confusion.
1 Reply Last Post Jan 11, 2010, 2:50 p.m. EST
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
One more question, can I use constants with the same name as the coefficients in the PDE coefficient form? Or will this confuse comsol? For example, there are c, q, a and g coefficients that are specified in the boundary settings and subdomain settings. If I then name a constant named c,q, a or g, would this overwrite the coefficient?