March 2015 - firedrake - imperial.ac.uk

Perturbed extruded meshes
by Eike Mueller 07 Mar '15

07 Mar '15

Dear firdrakers, how can I perturb an extruded mesh? I want to add 'mountains' (i.e. bumps) to my spherical icosahedral mesh, so I want to change the radius according to something like r = r(theta) = R_earth+H*exp(-0.5*theta^2/D^2) Can I do this by moving around the underlying 2d host grid and then extruding that? How can I perturb the 2d grid? Thanks a lot, Eike

5 9

Mapping nodal values from parent to extruded mesh
by Tuomas Karna 04 Mar '15

04 Mar '15

Hi all, I'd need to copy nodal values between fields on parent and extruded meshes. For example, copy 2d->3d (constant over vertical) or 3d->2d (extract surface/bottom level). The horizontal function space is the same. Is there an easy way to do this? Is there a map of extruded nodes somewhere? Thanks, Tuomas

1 0

Re: [firedrake] Problem with nonlinear solver on extruded mesh
by Andrew McRae 04 Mar '15

04 Mar '15

You may or may not need to do the whole chain: PETSc, petsc4py, FIAT, UFL, FFC, COFFEE, PyOP2, and Firedrake Good luck! On 4 March 2015 at 18:16, Anna Kalogirou <a.kalogirou(a)leeds.ac.uk> wrote: > I haven't updated after the new year, so that might be it. Thanks! > > Anna. > > > > On 04/03/15 18:14, Andrew McRae wrote: > > Is your Firedrake completely up-to-date? > > It sounds not-unrelated to > https://github.com/firedrakeproject/firedrake/issues/472, which was fixed > recently. > > On 4 March 2015 at 18:09, Anna Kalogirou <a.kalogirou(a)leeds.ac.uk> wrote: > >> Dear all, >> >> I am experiencing a problem with extruded meshes. When I try to solve >> the nonlinear VP: >> >> m = IntervalMesh(N,L) >> mesh = ExtrudedMesh(m, layers=10) >> >> F = ( v*(p1-p0)/dt + v*(h0-1) + sigma*inner(grad(v),grad(h0)) + >> 0.5*inner(grad(p0_5),grad(p0_5))*v )*dx + E_t*v*ds_v(1) >> >> I get the error >> >> File >> >> "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/variational_solver.py", >> line 97, in __init__ >> form_compiler_parameters=self._problem.form_compiler_parameters) >> File >> "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/assemble.py", >> line 58, in assemble >> form_compiler_parameters=form_compiler_parameters) >> File >> "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/assemble.py", >> line 79, in _assemble >> kernels = ffc_interface.compile_form(f, "form", >> parameters=form_compiler_parameters) >> File >> >> "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/ffc_interface.py", >> line 273, in compile_form >> ((kernel, needs_orientations), ) = ffc_kernel.kernels >> ValueError: need more than 0 values to unpack >> >> However, this problem does not arise when I switch to a RectangleMesh ( >> and change the ds_v(1) to ds(1) ), OR when I solve the corresponding >> linear VP. >> Anyone has any ideas why this is happening? >> >> Thank you. >> >> Best, Anna. >> >> _______________________________________________ >> firedrake mailing list >> firedrake(a)imperial.ac.uk >> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >> > > > -- > > Dr Anna Kalogirou > Research Fellow > School of Mathematics > University of Leeds > >

1 0

Re: [firedrake] Problem with nonlinear solver on extruded mesh
by Andrew McRae 04 Mar '15

04 Mar '15

Is your Firedrake completely up-to-date? It sounds not-unrelated to https://github.com/firedrakeproject/firedrake/issues/472, which was fixed recently. On 4 March 2015 at 18:09, Anna Kalogirou <a.kalogirou(a)leeds.ac.uk> wrote: > Dear all, > > I am experiencing a problem with extruded meshes. When I try to solve > the nonlinear VP: > > m = IntervalMesh(N,L) > mesh = ExtrudedMesh(m, layers=10) > > F = ( v*(p1-p0)/dt + v*(h0-1) + sigma*inner(grad(v),grad(h0)) + > 0.5*inner(grad(p0_5),grad(p0_5))*v )*dx + E_t*v*ds_v(1) > > I get the error > > File > > "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/variational_solver.py", > line 97, in __init__ > form_compiler_parameters=self._problem.form_compiler_parameters) > File > "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/assemble.py", > line 58, in assemble > form_compiler_parameters=form_compiler_parameters) > File > "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/assemble.py", > line 79, in _assemble > kernels = ffc_interface.compile_form(f, "form", > parameters=form_compiler_parameters) > File > > "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/ffc_interface.py", > line 273, in compile_form > ((kernel, needs_orientations), ) = ffc_kernel.kernels > ValueError: need more than 0 values to unpack > > However, this problem does not arise when I switch to a RectangleMesh ( > and change the ds_v(1) to ds(1) ), OR when I solve the corresponding > linear VP. > Anyone has any ideas why this is happening? > > Thank you. > > Best, Anna. > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

2 1

Problem with nonlinear solver on extruded mesh
by Anna Kalogirou 04 Mar '15

04 Mar '15

Dear all, I am experiencing a problem with extruded meshes. When I try to solve the nonlinear VP: m = IntervalMesh(N,L) mesh = ExtrudedMesh(m, layers=10) F = ( v*(p1-p0)/dt + v*(h0-1) + sigma*inner(grad(v),grad(h0)) + 0.5*inner(grad(p0_5),grad(p0_5))*v )*dx + E_t*v*ds_v(1) I get the error File "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/variational_solver.py", line 97, in __init__ form_compiler_parameters=self._problem.form_compiler_parameters) File "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/assemble.py", line 58, in assemble form_compiler_parameters=form_compiler_parameters) File "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/assemble.py", line 79, in _assemble kernels = ffc_interface.compile_form(f, "form", parameters=form_compiler_parameters) File "/opt/firedrake/2.0/lib/python2.7/site-packages/firedrake/ffc_interface.py", line 273, in compile_form ((kernel, needs_orientations), ) = ffc_kernel.kernels ValueError: need more than 0 values to unpack However, this problem does not arise when I switch to a RectangleMesh ( and change the ds_v(1) to ds(1) ), OR when I solve the corresponding linear VP. Anyone has any ideas why this is happening? Thank you. Best, Anna.

1 0

Re: [firedrake] Perturbing mesh extrusion at runtime
by David Ham 04 Mar '15

04 Mar '15

Hi Tuomas, I suspect you may be hitting a bug in the assembly cache. Does the issue go away if you specify: parameters["assembly_cache"]["enabled"] = False ? David On 3 March 2015 at 23:17, Tuomas Karna <tuomas.karna(a)gmail.com> wrote: > Hi all, > > Custom mesh extrusions can be generated by manipulating mesh.coordinates > directly. This works fine as long as the coordinates are changed prior > to the first solve. After that changing the coordinates has no effect as > precomputed mesh properties do not get updated. > > What do I need to do to force a mesh update? I need to update the > extrusion only, not the parent mesh. Can this be done efficiently? > > > Cheers, > > Tuomas > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > -- Dr David Ham Departments of Mathematics and Computing Imperial College London http://www.imperial.ac.uk/people/david.ham

2 1

Perturbing mesh extrusion at runtime
by Tuomas Karna 03 Mar '15

03 Mar '15

Hi all, Custom mesh extrusions can be generated by manipulating mesh.coordinates directly. This works fine as long as the coordinates are changed prior to the first solve. After that changing the coordinates has no effect as precomputed mesh properties do not get updated. What do I need to do to force a mesh update? I need to update the extrusion only, not the parent mesh. Can this be done efficiently? Cheers, Tuomas

1 0