|
From: | Tetsuo Koyama |
Subject: | Re: Coupling equations |
Date: | Wed, 11 May 2022 23:58:30 +0900 |
The weak formulation is in this case :
Rotationel(A) * rotationel(Test_A) - B.Test_A=0
Rotationel(B) * rotationel(Test_B) - A.Test_B=0
With the right Dirichlet conditions of course
You may have right, Getfem considers the test functions to seperate equations ( Test_A for the first and Test_B for the second)
That will make sense
Best LahoussaineCordialementLahoussaine BOURRICHE
3A, ENSEEIHT , Physique numérique
+33 6 20 76 17 12De: "Tetsuo Koyama" <tkoyama010@gmail.com>
À: "Lahoussaine Bourriche" <lahoussaine.bourriche@etu.toulouse-inp.fr>
Cc: "getfem-users" <getfem-users@nongnu.org>
Envoyé: Mercredi 11 Mai 2022 16:46:23
Objet: Re: Coupling equationsDear LahoussaineYour weak formulation is not correct I guess.Laplacian(A)+B.Test_A=0Laplacian(B)+A.Test_B=0Test function defines which part of equation it is.Laplacian brick includes Test function inside.2022年5月11日(水) 23:31 Bourriche Lahoussaine <lahoussaine.bourriche@etu.toulouse-inp.fr>:Thank you
But what it bothers me is that this implementation is the same as
Laplace (A)+B+Laplace(B)+A=0 implementation's
Two things different with the same implementationCordialementLahoussaine BOURRICHE
3A, ENSEEIHT , Physique numérique
+33 6 20 76 17 12De: "Tetsuo Koyama" <tkoyama010@gmail.com>
À: "LahouDearssaine Bourriche" <lahoussaine.bourriche@etu.toulouse-inp.fr>
Cc: "getfem-users" <getfem-users@nongnu.org>
Envoyé: Mercredi 11 Mai 2022 16:23:59
Objet: Re: Coupling equationsDear BourricheThank you for reply.I also think it works.Best Tetsuo2022年5月11日(水) 23:13 Bourriche Lahoussaine <lahoussaine.bourriche@etu.toulouse-inp.fr>:Hy Tetsu
Thank you for your time.
I saw the link and it's far from what i have, my problem is kind of :
Laplace(A) + B =0
Laplace(B) + A =0
I suggest :
#Equation 1 :
Add_non_lineair_term( " Weak form of ( Laplace(A) ) )")
Add_lineair_term( " Weak form of B )")
#Equation 2 :
Add_non_lineair_term( " Weak form of ( Laplace(B) ) )")
Add_lineair_term( " Weak form of A )")
I disscused this with mr. Renard and he is ok with it. What do you think you
Best LahoussaineCordialementLahoussaine BOURRICHE
3A, ENSEEIHT , Physique numérique
+33 6 20 76 17 12De: "Tetsuo Koyama" <tkoyama010@gmail.com>
À: "Lahoussaine Bourriche" <lahoussaine.bourriche@etu.toulouse-inp.fr>
Cc: "getfem-users" <getfem-users@nongnu.org>
Envoyé: Mercredi 11 Mai 2022 15:58:06
Objet: Coupling equationsDear Bourriche Lahoussaine
> When i have two coupling equation, how getfem knows that I m working on the
> first equation /second equation, in other words
> How getfem will knowz which term to put in which equation
>
> Add_lineair (first equation)
> Add_lineair (second equation)
We have to combine bricks to variables which is define by MeshFem object.
GetFEM builds tangent matrix on each variable.
This tutorial might helps you.
https://getfem.org/tutorial/thermo_coupling.html
Best Tetsuo
[Prev in Thread] | Current Thread | [Next in Thread] |