PETSc : integrate surface

General Discussion
1

Dear all,

I would like to calculate a surface with “int2d” but I have fews issues to find the good keyword to not take the overlap

real Surf = int2d(Th3,1)( 1.0);

give the surface on each local mesh Th3

I try

mpiAllReduce(Surf,SurfSum,mpiCommWorld, mpiSUM);

but this is sum the element overlap.

Thank you in advance

2

You should either use the partition of unity or generate another mesh without ghost elements.

3

hi, thanks you for your answer.
In fact, I had tested the solution with a global mesh and integrating only with processor 0

if (mpirank == 0)
real SurfG = int2d(Th3Global,1)( 1. );

But I wanted to try with the partition of unity, and I didn’t know how to get started.

For my solution u, I huse the operator .D but for a real it didn’t works.

4

My objective is to calculate flows in a heat equation.

By operating on a global mesh, I get the right surface calculations. However, my flux calculation (function of dx(u)) is processor-dependent.
(moreover even with one processor, I didn’t have exactly the same flux at the output as that imposed in my boundary conditions.)
I’ve attached the code
231026_Exemple.edp (3.4 KB)

Thank you in advance

5

What is the issue in this script?

6

I impose a flow of 500,
for post-processing calculation with 1 processor

FluxGTh1 =int2d(Th3Global,101)(eKGlobal* (dy(sum))
FluxGTh1/SurfG = -495.749 [W]

and with 4 processors

FluxGTh1 =int2d(Th3Global,101)(eKGlobal* (dy(sum))
FluxGTh1/SurfG = -481.749 [W]

I don’t understand why the two calculations give differents results.

7

Use -pc_type lu and you’ll get the proper answer. In the beginning, there is no point in trying fancy preconditioners if you are not sure of what you are doing.

8

it works better that way. Thanks you