[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Getfem-users] Assembly of 1D Poiseuille term
From: |
Yves Renard |
Subject: |
Re: [Getfem-users] Assembly of 1D Poiseuille term |
Date: |
Mon, 27 Jul 2015 16:23:28 +0200 (CEST) |
Dear Domenico,
In the two assembly languages, the gradient has indeed the dimension of the
mesh, i.e. a 3D one in this case. It is along the tangent direction of the line
in this case.
One way to obtain the tangent vector is to use "element_K" from the high-level
generic assembly (gradient of the geometric transformation).
The assembly string
" -p*Grad_u.element_K "
should work. May be "element_K" have to be transposed.
Additionally, if you need to interpolate some field from the 3D mesh to the 1D
mesh and perform some matrix assembly, the use of the high-level generic
assembly is far more simple (see
http://download.gna.org/getfem/html/homepage/userdoc/gasm_high.html#interpolate-transformations).
Yves.
----- Original Message -----
From: "Domenico Notaro" <address@hidden>
To: "getfem-users" <address@hidden>
Sent: Sunday, July 26, 2015 7:30:42 PM
Subject: [Getfem-users] Assembly of 1D Poiseuille term
Dear Users,
I am studying a coupled 3D-1D problem, for the sake of simplicity let's imagine
a straight line immersed in a box.
I am wondering how to assemble the matrix corresponding to the weak formulation
of the Poiseuille term in the 1D problem, -(p, du/ds), being p the pressure of
the fluid inside the segment, u the velocity and s the arc length along the
segment. I have to idea how to write the derivative inside the generic_assembly
since the only way I know to compute derivate is "Gradient(#..)" that is a
vector.
Do I need to project the gradient along the direction of the segment? If this
is the case, I don't know how to compute the tangent versor. Thank you in
advance.
Regards,
domenico_notaro
_______________________________________________
Getfem-users mailing list
address@hidden
https://mail.gna.org/listinfo/getfem-users