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 The information provided may be out of date.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.
multiple solutions
Posted 1 sept. 2011, 01:30 UTC−4 Version 4.2 1 Reply
Please login with a confirmed email address before reporting spam
Hi all,
I have a model (electric currents), that I need to solve many times. In my model, everything remains the same, but the source location. So I need multiple solutions for each source location.
From the results I need the following
1) Potential in some nodes
2) The integral of the dot product between the current density, from different source locations ( like int(Jx1*Jx2), where Jx1 is the current density when source is at location 1, and Jx2 when source is at location 2.
The way I am trying to solve this, is one of the following
1) The obvious way to do that is to add many physics in my model, where everything remains the same, but the source location. But this approach requires to define the properties on each physics, the boundary conditions (although it is the same everywhere). It is very time consuming.
2) I read in comsol forums, that although I could create multiple studies, and get results, those results do not interact with each other (like I can’t even many a subtraction of V1-V2, for two different solutions).
3) To create different models, one for each solution. But this way, I still don’t know if I cwill be able to calculate the integral.
4) Parametric solver, where the location of source is a variable. Can I do that?
Which method would you propose?
Thanks in advance.
I have a model (electric currents), that I need to solve many times. In my model, everything remains the same, but the source location. So I need multiple solutions for each source location.
From the results I need the following
1) Potential in some nodes
2) The integral of the dot product between the current density, from different source locations ( like int(Jx1*Jx2), where Jx1 is the current density when source is at location 1, and Jx2 when source is at location 2.
The way I am trying to solve this, is one of the following
1) The obvious way to do that is to add many physics in my model, where everything remains the same, but the source location. But this approach requires to define the properties on each physics, the boundary conditions (although it is the same everywhere). It is very time consuming.
2) I read in comsol forums, that although I could create multiple studies, and get results, those results do not interact with each other (like I can’t even many a subtraction of V1-V2, for two different solutions).
3) To create different models, one for each solution. But this way, I still don’t know if I cwill be able to calculate the integral.
4) Parametric solver, where the location of source is a variable. Can I do that?
Which method would you propose?
Thanks in advance.
1 Reply Last Post 2 sept. 2011, 14:53 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
for me its possile to do a "V1-V2" but first you mus declare the "V" as global, that is you need to map the V1 from slution 1 onto a general valirale MyV1, the same for V2 and then take the differences of the global variables. Then you must not forget that when you write V1 in a field you mean in fact V1(x,y,z) for the domain/boundary where V1 has been defined so when you write MyV1-MyV2 you must also be sure your are doing this over a domain/boundary that defines correctly the x,yz for both variables (OK if you have the same geoemtry).
You obtain the mapping by using the Definition coupling variables, see the doc
--
Good luck
Ivar
for me its possile to do a "V1-V2" but first you mus declare the "V" as global, that is you need to map the V1 from slution 1 onto a general valirale MyV1, the same for V2 and then take the differences of the global variables. Then you must not forget that when you write V1 in a field you mean in fact V1(x,y,z) for the domain/boundary where V1 has been defined so when you write MyV1-MyV2 you must also be sure your are doing this over a domain/boundary that defines correctly the x,yz for both variables (OK if you have the same geoemtry).
You obtain the mapping by using the Definition coupling variables, see the doc
--
Good luck
Ivar