How to compute curvature of a boundary

1

Dear Fenics community
I am new to Fenics and hence the following question.
How to compute the curvature of a curved boundary of the domain? I have tried curvature, H, as
H =-\text{\div( n)}, where \text n is the unit normal vector of the boundary. When I compute assemble( H* \mathbf dS) for a circle, it results in zero, instead of 2\pi. I have gone through this forum and found a thread on the explanation for the same. However, I could not compute curvature. Could anyone kindly help me with computing it in any possible way by posting a code snippet?
Thanks

2

Using functionality implemented in dolfin-adjoint, which is an extension/generalization of an implementation in FEmorph, you can do the following (only works for CG1-approximations of the curvature and normal vector.

The following code first projects the facet normal to the boundary of a volume CG-1 function space, then transfering it to the boundary-mesh, consisting only of the exterior facets.
Then, we compute the curvature on the boundary-mesh, and transfer the values back to the volume mesh.

import numpy
import dolfin
from dolfin import *


def mesh_to_boundary(v, b_mesh):
    """
    Returns a the boundary representation of the CG-1 function v
    """
    # Extract the underlying volume and boundary meshes
    mesh = v.function_space().mesh()

    # We use a Dof->Vertex mapping to create a global
    # array with all DOF values ordered by mesh vertices
    DofToVert = dolfin.dof_to_vertex_map(v.function_space())
    VGlobal = numpy.zeros(v.vector().size())

    vec = v.vector().get_local()
    for i in range(len(vec)):
        Vert = dolfin.MeshEntity(mesh, 0, DofToVert[i])
        globalIndex = Vert.global_index()
        VGlobal[globalIndex] = vec[i]
    VGlobal = SyncSum(VGlobal)

    # Use the inverse mapping to se the DOF values of a boundary
    # function
    surface_space = dolfin.FunctionSpace(b_mesh, "CG", 1)
    surface_function = dolfin.Function(surface_space)
    mapa = b_mesh.entity_map(0)
    DofToVert = dolfin.dof_to_vertex_map(dolfin.FunctionSpace(b_mesh, "CG", 1))

    LocValues = surface_function.vector().get_local()
    for i in range(len(LocValues)):
        VolVert = dolfin.MeshEntity(mesh, 0, mapa[int(DofToVert[i])])
        GlobalIndex = VolVert.global_index()
        LocValues[i] = VGlobal[GlobalIndex]

    surface_function.vector().set_local(LocValues)
    surface_function.vector().apply('')
    return surface_function

def vector_mesh_to_boundary(func, b_mesh):
    v_split = func.split(deepcopy=True)
    v_b = []
    for v in v_split:
        v_b.append(mesh_to_boundary(v, b_mesh))
    Vb = dolfin.VectorFunctionSpace(b_mesh, "CG", 1)
    vb_out = dolfin.Function(Vb)
    scalar_to_vec = dolfin.FunctionAssigner(Vb, [v.function_space() for
                                                  v in v_b])
    scalar_to_vec.assign(vb_out, v_b)
    return vb_out


def SyncSum(vec):
    """ Returns sum of vec over all mpi processes.
    Each vec vector must have the same dimension for each MPI process """

    comm = dolfin.MPI.comm_world
    NormalsAllProcs = numpy.zeros(comm.Get_size() * len(vec), dtype=vec.dtype)
    comm.Allgather(vec, NormalsAllProcs)

    out = numpy.zeros(len(vec))
    for j in range(comm.Get_size()):
        out += NormalsAllProcs[len(vec) * j:len(vec) * (j + 1)]
    return out

def boundary_to_mesh(f, mesh):
    b_mesh = f.function_space().mesh()
    SpaceV = dolfin.FunctionSpace(mesh, "CG", 1)
    SpaceB = dolfin.FunctionSpace(b_mesh, "CG", 1)

    F = dolfin.Function(SpaceV)
    GValues = numpy.zeros(F.vector().size())

    map = b_mesh.entity_map(0)  # Vertex map from boundary mesh to parent mesh
    d2v = dolfin.dof_to_vertex_map(SpaceB)
    v2d = dolfin.vertex_to_dof_map(SpaceV)

    dof = SpaceV.dofmap()
    imin, imax = dof.ownership_range()

    for i in range(f.vector().local_size()):
        GVertID = dolfin.Vertex(b_mesh, d2v[i]).index()  # Local Vertex ID for given dof on boundary mesh
        PVertID = map[GVertID]  # Local Vertex ID of parent mesh
        PDof = v2d[PVertID]  # Dof on parent mesh
        value = f.vector()[i]  # Value on local processor
        GValues[dof.local_to_global_index(PDof)] = value
    GValues = SyncSum(GValues)

    F.vector().set_local(GValues[imin:imax])
    F.vector().apply("")
    return F


mesh = Mesh(UnitDiscMesh.create(MPI.comm_world, 100, 1, 2))
n = FacetNormal(mesh)
V = VectorFunctionSpace(mesh, "CG", 1)
u = TrialFunction(V)
v = TestFunction(V)
a = inner(u,v)*ds
l = inner(n, v)*ds
A = assemble(a, keep_diagonal=True)
L = assemble(l)

A.ident_zeros()
nh = Function(V)

solve(A, nh.vector(), L)
File("nh.pvd") << nh
bmesh = BoundaryMesh(mesh, "exterior")
nb = vector_mesh_to_boundary(nh, bmesh)
Q = FunctionSpace(bmesh, "CG", 1)

p, q = TrialFunction(Q), TestFunction(Q)
a = inner(p,q)*dx
l = inner(div(nb), q)*dx
A = assemble(a, keep_diagonal=True)
L = assemble(l)
A.ident_zeros()
kappab = Function(Q)
solve(A, kappab.vector(), L)
kappa = boundary_to_mesh(kappab, mesh)

print(assemble(kappa*ds))
File("kappa.pvd") << kappa
1 Like
3

Thanks for the kind help, Dokken!
Is there an alternate way where I can extract the coordinates of the boundary (say internal boundary, a curved interface between two fluids) and then compute curvature using the standard definition from the derivates of the position vector?

Thanks

4

Dear Dokken
With the above method, I could compute curvature for the external boundaries. However, I am unable to compute the same for an internal boundary (a circular interface between two different fluids in a rectangular box ) as it leads to errors. Could you kindly help me with this internal boundary curvature? Sorry for the trouble.
Thanks

5

This method only works for exterior boundaries, as that is what is used to transfer quantities. To do this with interior boundaries, you need to define a SubMesh for a volume with the internal interface as an exterior boundary, and then create a boundary mesh for the submesh.

6

Thanks for the help Dokken.