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 The information provided may be out of date.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.

Defining constraint

Please login with a confirmed email address before reporting spam

Hello friends,
I would really appreciate some help in defining a constraint. I am trying to solve a system of PDEs but the result at the earlier times keep overshooting the expected value. The maximum value for this variable should be 825, which is the initial value of this variable. However, each time I solve the system, I keep having the values of this variable within the first few seconds overshoot the initial value which is very unreasonable. This is a consumption problem, the values of the variable should be steadily decrasing to zero. Could someone help me out?
My objective is to constrain my variable to 825 each time the result obtained by solving the system of PDEs by comsol exceeds 825 but keep all values of solution less than 825 at t>0. Thanks for helping out.

Eneh
N/B: I am using comsol 4.2a

2 Replies Last Post 15 nov. 2012, 09:53 UTC−5
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 15 nov. 2012, 09:08 UTC−5
Hi

I'm not sure that is the correct approach, why does it overshoot ?

is it the numerical method used ?
or not enough mesh elements in the visinity of the steep step ?
or simply the equations that implies inertial terms and the initial values are too far from a steady state situation that you get "physical" and not numerical overshoot

--
Good luck
Ivar
Hi I'm not sure that is the correct approach, why does it overshoot ? is it the numerical method used ? or not enough mesh elements in the visinity of the steep step ? or simply the equations that implies inertial terms and the initial values are too far from a steady state situation that you get "physical" and not numerical overshoot -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 15 nov. 2012, 09:53 UTC−5
Hi Ivar,
Thanks for your response. What you've said about the approach is correct. I do not know which of the numerical methods that comsol uses in solving descritized systems of equation as this can greatly impart the solution. To give you more insight, my PDE is of the form found in the attached file. Whenever I remove the differential term on the right-hand side of the equations, I will obtain the expected solution. Hence, the problem is from the differential term on the right hand side. I did try taking it over to the left hand side, but the problem was not solved. The fact is that the value of u4 is about 100 times lesser than u1 and u2 and this problem seems to be local to these two variables. Thanks for any further guidance on what to do to resolve this.

Eneh
Hi Ivar, Thanks for your response. What you've said about the approach is correct. I do not know which of the numerical methods that comsol uses in solving descritized systems of equation as this can greatly impart the solution. To give you more insight, my PDE is of the form found in the attached file. Whenever I remove the differential term on the right-hand side of the equations, I will obtain the expected solution. Hence, the problem is from the differential term on the right hand side. I did try taking it over to the left hand side, but the problem was not solved. The fact is that the value of u4 is about 100 times lesser than u1 and u2 and this problem seems to be local to these two variables. Thanks for any further guidance on what to do to resolve this. Eneh

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.