Defining subdomains

mesh
1

I feel like I am going a bit crazy here. Pretty simple set-up, I have a unit square mesh, and I would like to define a circle subdomain. The code I’ve used is below.

from dolfin import *
import numpy as np

msh = UnitSquareMesh(500, 500)

def circle(x):
    return (x[0]-0.5)**2 + (x[1]-0.5)**2

class Circle(SubDomain):
    def inside(self, x, on_boundary):
        return near(circle(x) , 0.2**2)

domains = MeshFunction("size_t", msh, msh.topology().dim()-1, 0)

circ = Circle()
circ.mark(domains,1)

However, the circle is still marked 0 when I try and plot domains. Refining the mesh does not work and neither does increasing or decreasing the tolerance in the near function. I cannot think of any reason this does not work. If I try and set Circle to return circle(x) > 0.2**2 or circle(x) < 0.2**2, it works perfects fine, so I really do not understand why this doesn’t work.

Any help would be fantastic.

2

Hi,
below code works perfectly fine for me.

from dolfin import *
import numpy as np

mesh = UnitSquareMesh(500, 500)
domains = MeshFunction("size_t", msh, msh.topology().dim()-1)
CompiledSubDomain("(x[0]-0.5)*(x[0]-0.5) + (x[1]-0.5)*(x[1]-0.5) < 0.2*0.2").mark(domains,1)

while using subdomain, you need to specify which area you want to mark, whether it is lesser than the radius or greater than the radius.

1 Like
3

I understand that, but I am trying to mark just the boundary of the circle, not the inside or outside of the circle, as I specified in my question.

4

You can do this:

CompiledSubDomain("0.18*0.18 < (x[0]-0.5)*(x[0]-0.5) + (x[1]-0.5)*(x[1]-0.5) &&  (x[0]-0.5)*(x[0]-0.5) + (x[1]-0.5)*(x[1]-0.5) <0.2*0.2").mark(domains,1)

image
image563×566 4.14 KB

and change the thickness of the boundary accordingly.