Hello,
Is there any keyword for coupling any given degrees of freedom over a selected group of nodes?
Abhishek
Coupled degrees of freedom
-
- Site Admin
- Posts: 4832
- Joined: 22 Aug 2009, 11:57
- Antispam: Yes
- Location: Espoo, Finland
- Contact:
Re: Coupled degrees of freedom
Hi, there are many ways by which dofs may be coupled. Could you be more specific what you mean. -Peter
Re: Coupled degrees of freedom
Hello,
Suppose that I am solving a thermal conduction problem, and I have to ensure that the temperature of all nodes on a given region are the same, so I can couple the temperature degrees of freedom for these group of nodes. Is it possible?
In the Elmer Solver Manual, there is an article on Linear Constraints, but no example.
Regards
Abhishek
Suppose that I am solving a thermal conduction problem, and I have to ensure that the temperature of all nodes on a given region are the same, so I can couple the temperature degrees of freedom for these group of nodes. Is it possible?
In the Elmer Solver Manual, there is an article on Linear Constraints, but no example.
Regards
Abhishek
-
- Site Admin
- Posts: 4832
- Joined: 22 Aug 2009, 11:57
- Antispam: Yes
- Location: Espoo, Finland
- Contact:
Re: Coupled degrees of freedom
Hi
I'm not either familiar with using the linear constraint solver. I think tutorial 25 deals with an example. There is also an alternative "coupled solver" approach which allows for all kinds of constraints and strong coupling of different equations in different regions. This is alternative to the main weak coupling loop and hasn't been much pursued but you can find CoupledSolver* examples in the tests directory. Unfortunately the examples are related to just a simple Poisson equation.
Perhaps it would be an option to describe just a really high thermal conductivity to the region where these nodes are, to ensure that they get roughly the same value.
-Peter
I'm not either familiar with using the linear constraint solver. I think tutorial 25 deals with an example. There is also an alternative "coupled solver" approach which allows for all kinds of constraints and strong coupling of different equations in different regions. This is alternative to the main weak coupling loop and hasn't been much pursued but you can find CoupledSolver* examples in the tests directory. Unfortunately the examples are related to just a simple Poisson equation.
Perhaps it would be an option to describe just a really high thermal conductivity to the region where these nodes are, to ensure that they get roughly the same value.
-Peter
Re: Coupled degrees of freedom
Hello,
Is there any example problem for tutorial 25??? ..... I don't think giving large thermal conductivity will solve the issue, as it might lead to stiff matrices...
Abhishek
Is there any example problem for tutorial 25??? ..... I don't think giving large thermal conductivity will solve the issue, as it might lead to stiff matrices...
Abhishek