Menu

Discussion Forum

New Discussion
Filtra

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 The information provided may be out of date.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.

weak constraint or pointwise constraint

Posted 6 mag 2012, 23:21 GMT-4 Fluid & Heat, Studies & Solvers Version 4.2a 7 Replies

Abbie
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hello,

I would like to have ur opinion for this situation:

I wish to constraint the (fluctuating) pressure to an average value of 20 [constraint expression: aveop1(p)-20] at the inlet boundary while pulsatile flow velocity is imposed at the inlet too.

I am not sure if weak constraint or pointwise constraint option is more suitable for this condition?
I read that weak constraint and pointwise constraint are similar but there is no Lagrange multipliers created in pointwise constraint. Anyone can please tell me what the importance of Lagrange multipliers in this case?

Thank you for your kindness.
7 Replies Last Post 8 mar 2017, 07:06 GMT-5
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 7 mag 2012, 02:10 GMT-4
Hi

I'm not sure I fully follow you, normally one drive one parameter with an equations i.e. keeping a value at an average. That is to add an ODE or a weak expression, define a variabele as you write it, for example the average of something like the pressure, and let COMSOL estimate the best value of the driving object such that this global average value is equal to your goal. This is an optimisation process or a feedback loop all depening on how you look at it.

But you are talking about having a given inlet condition and then "just" ensure the average value is constrained, you need to define what is providing this constraint. It could be the input pressure or flow that you constrain such that the average is something known, but the distribution of your field flow would be given still by the physics around, no-slip boundary etc

--
Good luck
Ivar
Hi I'm not sure I fully follow you, normally one drive one parameter with an equations i.e. keeping a value at an average. That is to add an ODE or a weak expression, define a variabele as you write it, for example the average of something like the pressure, and let COMSOL estimate the best value of the driving object such that this global average value is equal to your goal. This is an optimisation process or a feedback loop all depening on how you look at it. But you are talking about having a given inlet condition and then "just" ensure the average value is constrained, you need to define what is providing this constraint. It could be the input pressure or flow that you constrain such that the average is something known, but the distribution of your field flow would be given still by the physics around, no-slip boundary etc -- Good luck Ivar
Abbie
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 7 mag 2012, 10:50 GMT-4
Thanks Ivar!

I attached cylinder_inlet.jpg and my .mph file for more detail description. In the model, the fluid flows into cylinder and its volume expansion was prescribed through ODE equation.

In this fluid flow case, the pressure at the inlet boundary will be calculated automatically from the imposed inlet velocity through built-in Navier stokes equation. So in order to constraint the average pressure at the inlet boundary to 20, i add a weak constraint to the inlet boundary however i get this error:

Failed to find consistent initial values.
Singular matrix.
There are 1 degrees of freedom that do not occur in any equation (empty columns in matrix) for the variable mod1.lm.
at coordinates: (0,0), ...
Last time step is not converged.

Can u please take a look the attached .mph and give me some guidance?

Thanks again.
Thanks Ivar! I attached cylinder_inlet.jpg and my .mph file for more detail description. In the model, the fluid flows into cylinder and its volume expansion was prescribed through ODE equation. In this fluid flow case, the pressure at the inlet boundary will be calculated automatically from the imposed inlet velocity through built-in Navier stokes equation. So in order to constraint the average pressure at the inlet boundary to 20, i add a weak constraint to the inlet boundary however i get this error: Failed to find consistent initial values. Singular matrix. There are 1 degrees of freedom that do not occur in any equation (empty columns in matrix) for the variable mod1.lm. at coordinates: (0,0), ... Last time step is not converged. Can u please take a look the attached .mph and give me some guidance? Thanks again.
Abbie
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 7 mag 2012, 10:53 GMT-4
The attached cylinder_inlet.jpg.
The attached cylinder_inlet.jpg.

Attachments:

Yoav Green
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 mag 2012, 08:19 GMT-4
Hi
what is the difference between choosing a pointwise constraint versus the weak constraint?
Lets say I need a no-slip velocity on a certain wall, which of the above do I chose? What is preferable? What will give me faster convergence?
Thanks
Hi what is the difference between choosing a pointwise constraint versus the weak constraint? Lets say I need a no-slip velocity on a certain wall, which of the above do I chose? What is preferable? What will give me faster convergence? Thanks
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28 mag 2012, 03:35 GMT-4
Hi

in all generality, any BC defined at a level 2* lower than the domain level: 2D Surface => Line => (Point/Vertex), or 3D volume => Surface => ( Edge => Point/Vertex) tend to give singularities.

Then in CFD an "inlet", as the "outlet" are "names" you can get localy flow "out" of an "inlet" boundary, so long the overall flow condition is respected as a mean over the boundary (slightly BC setings dependent), and the opposite for the "outlet". So respecting and average inflow and positive pressure increase over the full inlet BC gives a different result as the single point restriction

--
Good luck
Ivar
Hi in all generality, any BC defined at a level 2* lower than the domain level: 2D Surface => Line => (Point/Vertex), or 3D volume => Surface => ( Edge => Point/Vertex) tend to give singularities. Then in CFD an "inlet", as the "outlet" are "names" you can get localy flow "out" of an "inlet" boundary, so long the overall flow condition is respected as a mean over the boundary (slightly BC setings dependent), and the opposite for the "outlet". So respecting and average inflow and positive pressure increase over the full inlet BC gives a different result as the single point restriction -- Good luck Ivar
Fernando Jorge Monteiro Dias
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 18 nov 2016, 17:21 GMT-5
Hello

I need some help, please.

How do I impose a current density on a copper cable in COMSOL?


thanks in advance
Hello I need some help, please. How do I impose a current density on a copper cable in COMSOL? thanks in advance
Francis Ridgeon
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 8 mar 2017, 07:06 GMT-5
pointwise constraint.
Define a integral across the surface you want current in.
Then define the function of the current, how it varies with time.
Then pointwise constraint allows you to define

Current function - integral over surface (with variable) = 0
pointwise constraint. Define a integral across the surface you want current in. Then define the function of the current, how it varies with time. Then pointwise constraint allows you to define Current function - integral over surface (with variable) = 0

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.

Suggested Content