Hello,
when running the minimal example attached below the following problem occurs. When I consider
L = inner(f, v)*dx
I can solve a linear elasticity equation with adaptive mesh refinement.
Whenever I include terms with ds in the linear form, like
L = inner(f, v)*dx + inner(g, v)*ds(2)
I receive the following error message:
In instant.recompile: The module did not compile with command 'make VERBOSE=1 ', see '/home/kimmerle/.instant/error/instant_module_f85117a588b102c37e97eb08e8ba64a1195dcf6a/compile.log'
Traceback (most recent call last):
File "Test_MinimalExample.py", line 57, in <module>
solve(a == L, u, bc, tol=5.0E-7, M=MF) #Solve adaptively
File "/usr/lib/python2.7/dist-packages/dolfin/fem/solving.py", line 264, in solve
_solve_varproblem_adaptive(*args, **kwargs)
File "/usr/lib/python2.7/dist-packages/dolfin/fem/solving.py", line 339, in _solve_varproblem_adaptive
solver = AdaptiveLinearVariationalSolver(problem, M)
File "/usr/lib/python2.7/dist-packages/dolfin/fem/adaptivesolving.py", line 58, in __init__
ec = generate_error_control(self.problem, goal)
File "/usr/lib/python2.7/dist-packages/dolfin/fem/adaptivesolving.py", line 150, in generate_error_control
forms = [Form(form, form_compiler_parameters=p) for form in ufl_forms]
File "/usr/lib/python2.7/dist-packages/dolfin/fem/form.py", line 56, in __init__
common_cell)
File "/usr/lib/python2.7/dist-packages/dolfin/compilemodules/jit.py", line 66, in mpi_jit
return local_jit(*args, **kwargs)
File "/usr/lib/python2.7/dist-packages/dolfin/compilemodules/jit.py", line 154, in jit
return jit_compile(form, parameters=p, common_cell=common_cell)
File "/usr/lib/python2.7/dist-packages/ffc/jitcompiler.py", line 77, in jit
return jit_form(ufl_object, parameters, common_cell)
File "/usr/lib/python2.7/dist-packages/ffc/jitcompiler.py", line 212, in jit_form
cache_dir = cache_dir)
File "/usr/lib/python2.7/dist-packages/ufc_utils/build.py", line 64, in build_ufc_module
**kwargs)
File "/usr/lib/python2.7/dist-packages/instant/build.py", line 541, in build_module
recompile(modulename, module_path, new_compilation_checksum, build_system)
File "/usr/lib/python2.7/dist-packages/instant/build.py", line 150, in recompile
instant_error(msg % (cmd, compile_log_filename_dest))
File "/usr/lib/python2.7/dist-packages/instant/output.py", line 49, in instant_error
raise RuntimeError(text)
RuntimeError: In instant.recompile: The module did not compile with command 'make VERBOSE=1 ', see '/home/kimmerle/.instant/error/instant_module_f85117a588b102c37e97eb08e8ba64a1195dcf6a/compile.log'
By choosing a fine mesh manually, I can solve the problem by FEniCS accurately. The PDE is well-posed and numerical locking effects have been taken care of.
It seams to me that there is a bug or what might be the issue?
I am working with FEniCS v1.2 in python.
Thanks in advance,
SJ
PS: Here is a minimal example:
from dolfin import *
parameters["linear_algebra_backend"] ="PETSc"
parameters["form_compiler"]["quadrature_degree"] = 4
parameters["allow_extrapolation"] = True
b2 = 0.8
mesh = BoxMesh(0, -0.5*b2, -0.4, 45.0, 0.5*b2, 0.4, 10, 6, 4)
print mesh
V = VectorFunctionSpace(mesh, 'Lagrange', 2)
# Initialize mesh function for boundary domains
boundary_parts = MeshFunction("size_t", mesh, mesh.topology().dim()-1)
# Define boundary conditions
class DirichletBoundaryCondLeft (SubDomain):
def inside(self, x, on_boundary):
return on_boundary and (x[0] <= DOLFIN_EPS)
class NeumannBoundaryCondBottom (SubDomain):
def inside(self, x, on_boundary):
return on_boundary and (x[1] <= -0.4 + DOLFIN_EPS)
u0_boundary = DirichletBoundaryCondLeft()
u0_boundary.mark(boundary_parts, 1)
g_boundary = NeumannBoundaryCondBottom()
g_boundary.mark(boundary_parts, 2)
u0 = Constant((0.0, 0.0, 0.0))
bc = DirichletBC(V, u0, boundary_parts, 1)
# Define parameters for Cauchy tensor
E, nu = pow(10,7)/6.0, 0.285
lambdaS=Constant(E*nu/((1.0 + nu)*(1.0 - 2.0*nu)))
muS=Constant(E/(2.0*(1.0 + nu)))
# Define variational problem
u = TrialFunction(V)
v = TestFunction(V)
# Define Cauchy tensor
def sigma(v):
return lambdaS*tr(sym(grad(v)))*Identity(v.cell().d) + 2.0*muS*sym(grad(v))
# Define volume/boundary forces
f = Constant((0.0, -0.2, 0.0))
g = Constant((0.0, -10.0, 0.0))
# Define (bi)linear forms
a = (inner(sigma(u), grad(v)))*dx
L = inner(f, v)*dx + inner(g, v)*ds(2)
#L = inner(f, v)*dx
u = Function(V)
# Define goal functional for adaptive mesh refinement
n = FacetNormal(mesh)
MF = -inner(u, n)*ds #(inner(f, u)-inner(sigma(u), grad(u)))*dx #inner(sigma(u)*n, n)*ds(2)+inner(g, n))*ds(2)
solve(a == L, u, bc, tol=5.0E-7, M=MF) #Solve adaptively
plot(u, mode="displacement")
interactive()
