Mailman 3 December 2010 - SfePy

Re: Torque
by Robert Cimrman 10 Mar '18

10 Mar '18

In examples/large_deformation/hyperelastic.py a rotation by displacements is applied. By using a similar function the vectors defining the force couples could be defined for dw_surface_ltr (IMHO). Does it make sense? r. ----- Reply message ----- From: "Andre Smit" <freev...(a)gmail.com> To: <sfepy...(a)googlegroups.com> Subject: Torque Date: Sat, Dec 18, 2010 05:10 What is the best way to apply a torque load to a model? -- Andre -- You received this message because you are subscribed to the Google Groups "sfepy-devel" group. To post to this group, send email to sfepy...(a)googlegroups.com. To unsubscribe from this group, send email to sfepy-devel...(a)googlegroups.com. For more options, visit this group at http://groups.google.com/group/sfepy-devel?hl=en.

4 7

components of fields and variables
by Robert Cimrman 05 Jan '11

05 Jan '11

Hi, I am thinking about decoupling the shape of a Field from the number of components of a Variable. The current state of affairs is that the field shape defines the components of related variables, but in reality the FE base is always scalar, applied to each component of a vector variable separately. This is in contrast with truely vector FE spaces like the edge elements used for the Maxwell equations. To fix that I propose: 1. For scalar FE approximations the field shape will ve 1 - this is for all examples we have. 2. Variable syntax will be changed to allow for the number of components. Apart from the clean-up reasons, it will be IMHO easier to implement the edge elements then. Opinions/ideas? r.

1 1

Re: material_nonlinearity example
by Robert Cimrman 21 Dec '10

21 Dec '10

Yes, the function should return a value of mu in each quadrature point. So the "fill" line initializes those values to a constant 1, which is then modified according to the strain, also evaluated in the quadrature points. And all is for all quad. points, for all elements at once. So essentialy, mu = tile(mu, n_element * n_qp) I see that many more things need to be documented :) r. ----- Reply message ----- From: "Andre Smit" <freev...(a)gmail.com> To: <sfepy...(a)googlegroups.com> Subject: material_nonlinearity example Date: Mon, Dec 20, 2010 20:44 Robert In the get_pars function you prepare the mu or val matrix: if mode != 'qp': return val = nm.empty((coors.shape[0], 1, 1), dtype=nm.float64) val.fill(1e0) I'm trying to make sense of this. Are you repeating the material parameter for each gauss point? So for example, for a quad, there are 4 gauss points and essentially mu = tile(mu,4) and reshaped? a On Mon, Dec 20, 2010 at 11:08 AM, Robert Cimrman <cimr...(a)ntc.zcu.cz> wrote: On 12/20/10 17:45, Andre Smit wrote: Robert - in the new example I notice you are not using presolve in the solver definition or definition the history term for the displacement variable: 'ls' : ('ls.scipy_direct', { 'presolve' : True }), presolve set to True means that, for linear problems ('problem' : 'linear' in newton configuration), the matrix is assembled and factorized before the time-stepping starts - the factors are then reused in each time step to quickly solve the linear systems. It also assumes that the matrix is the same in each time step - this is not the case here, as 'mu' changes... So presolve does not have sense in this example. variables = { 'u' : ('unknown field', 'displacement', 0., 'previous'), 'v' : ('test field', 'displacement', 'u'), } but consistently used these for other time step problems. Any reason? This really emphasizes my lack of understanding of the internals. For linear problems with constant matrix, the most efficient would be: solvers = { 'ls' : ('ls.scipy_direct', {'presolve' : True }), 'newton' : ('nls.newton', { 'i_max' : 1, 'eps_a' : 1e-10, 'eps_r' : 1.0, 'problem' : 'linear'}), 'ts' : ('ts.simple', {'t0' : 0.0, 't1' : 1.0, 'dt' : None, 'n_step' : 5, 'quasistatic' : True, # or False }), } I have just fixed a small bug with 'quasistatic' : True r. -- You received this message because you are subscribed to the Google Groups "sfepy-devel" group. To post to this group, send email to sfepy...(a)googlegroups.com. To unsubscribe from this group, send email to sfepy-devel...(a)googlegroups.com. For more options, visit this group at http://groups.google.com/group/sfepy-devel?hl=en. -- Andre -- You received this message because you are subscribed to the Google Groups "sfepy-devel" group. To post to this group, send email to sfepy...(a)googlegroups.com. To unsubscribe from this group, send email to sfepy-devel...(a)googlegroups.com. For more options, visit this group at http://groups.google.com/group/sfepy-devel?hl=en.

2 4

material_nonlinearity example
by Andre Smit 20 Dec '10

20 Dec '10

Robert - in the new example I notice you are not using presolve in the solver definition or definition the history term for the displacement variable: 'ls' : ('ls.scipy_direct', { 'presolve' : True }), variables = { 'u' : ('unknown field', 'displacement', 0., 'previous'), 'v' : ('test field', 'displacement', 'u'), } but consistently used these for other time step problems. Any reason? This really emphasizes my lack of understanding of the internals. -- Andre

2 2

sfepy: linear system not solved! (err = 3.217853e-12)
by Andre Smit 20 Dec '10

20 Dec '10

Although SfePy is indicating a system is NOT solved, the solution appears valid. Is the err term perhaps too rigid? -- Andre

2 1

Torque
by Andre Smit 18 Dec '10

18 Dec '10

What is the best way to apply a torque load to a model? -- Andre

1 0

Dump array to vtk
by Andre Smit 17 Dec '10

17 Dec '10

Hi guys I've generated an array comprising material D matrix or stiffness tensor values. It looks like this - where each block represents the material value for a cell or element (yes, they're all the same but will change in subsequent time steps). [[[ 1604.9382716 864.19753086 0. ] [ 864.19753086 1604.9382716 0. ] [ 0. 0. 370.37037037]] [[ 1604.9382716 864.19753086 0. ] [ 864.19753086 1604.9382716 0. ] [ 0. 0. 370.37037037]] [[ 1604.9382716 864.19753086 0. ] [ 864.19753086 1604.9382716 0. ] [ 0. 0. 370.37037037]] ..., [[ 1604.9382716 864.19753086 0. ] [ 864.19753086 1604.9382716 0. ] [ 0. 0. 370.37037037]] [[ 1604.9382716 864.19753086 0. ] [ 864.19753086 1604.9382716 0. ] [ 0. 0. 370.37037037]] [[ 1604.9382716 864.19753086 0. ] [ 864.19753086 1604.9382716 0. ] [ 0. 0. 370.37037037]]] I'd like to append this array to the vtk file generated by SfePy which contains displacements, stress and strains, etc. I was using the following to generate the array: Dlist=[] for i in range(NELS): Dlist.append(pb.conf.materials.values()[i].values['D']) Dmatrix=array(Dlist) print "~~~~~~~~~~~~~~~~~~~~~~~~~~~" print Dmatrix Dstate=pb.create_state() Dstate.set_full(Dmatrix) outD=Dstate.create_output_dict() out['D_matrix']=Struct(name='output_data',mode='cell',data=outD,dofs=None) but get the following *numpy* error: Traceback (most recent call last): File "/home/grassy/sfepy/simple.py", line 120, in <module> main() File "/home/grassy/sfepy/simple.py", line 117, in main app() File "/home/grassy/sfepy/sfepy/applications/application.py", line 29, in call_basic return self.call( **kwargs ) File "/home/grassy/sfepy/sfepy/applications/simple_app.py", line 112, in call post_process_hook_final=self.post_process_hook_final) File "/home/grassy/sfepy/sfepy/solvers/generic.py", line 216, in solve_direct nls_status=nls_status) File "/home/grassy/sfepy/sfepy/solvers/generic.py", line 148, in solve_evolutionary_op ts=ts) File "/home/grassy/sfepy/sfepy/fem/problemDef.py", line 535, in save_state out = post_process_hook( out, self, state, extend = extend ) File "its.py", line 102, in strain_rate outD=Dstate.create_output_dict() File "/home/grassy/sfepy/sfepy/fem/state.py", line 188, in create_output_dict var_info, extend) File "/home/grassy/sfepy/sfepy/fem/variables.py", line 587, in state_to_output fill_value=fill_value)) File "/home/grassy/sfepy/sfepy/fem/variables.py", line 1468, in create_output (self.n_dof / self.n_components, self.n_components)) File "/usr/lib/python2.7/site-packages/numpy/core/fromnumeric.py", line 170, in reshape return reshape(newshape, order=order) ValueError: total size of new array must be unchanged Can I use existing SfePy methods to write this array to the vtk or should I rather write my own? -- Andre

2 7

History error on dw_lin_elastic_th term
by Andre Smit 17 Dec '10

17 Dec '10

Hi guys Playing around with the dw_lin_elastic_th term. Unfortunately, no examples for this one. I've pasted my code at [1]. I'm assuming the material parameter H is similar to that for an isotropic material D. I'm getting a 'history error' as shown below. Any ideas? Thx [1] http://paste.pocoo.org/show/306024/ [grassy@myhost sfepy]$ ~/sfepy/simple.py nits.py sfepy: left over: ['disp', 'verbose', '__builtins__', '_filename', '__file__', '__name__', 'stress_strain', 'youngpoisson_to_lame', 'young', '__package__', 'mesh_file', 'Mesh', 'stiffness_tensor_youngpoisson', 'poisson', '__doc__'] sfepy: reading mesh (/home/grassy/sfepy.wiki/msh/its2D.mesh)... sfepy: ...done in 0.00 s sfepy: setting up domain edges... sfepy: ...done in 0.00 s sfepy: creating regions... sfepy: Top sfepy: Bottom sfepy: Omega sfepy: Left sfepy: ...done in 0.01 s sfepy: equation "equilibrium": sfepy: dw_lin_elastic_iso.2.Omega(asphalt.lam,asphalt.mu,v,u)+ dw_lin_elastic_th.2.Omega(ts,asphalt.D,v,du/dt)=0 sfepy: setting up dof connectivities... sfepy: ...done in 0.00 s sfepy: using solvers: ts: ts nls: newton ls: ls sfepy: ====== time 0.000000e+00 (step 1 of 10) ===== sfepy: updating variables... sfepy: ...done ====================================================================== Displacement: -0.111111111111 sfepy: updating materials... sfepy: asphalt sfepy: ...done in 0.00 s sfepy: matrix shape: (93, 93) sfepy: assembling matrix graph... sfepy: ...done in 0.00 s sfepy: matrix structural nonzeros: 1063 (1.23e-01% fill) sfepy: nls: iter: 0, residual: 1.015969e+02 (rel: 1.000000e+00) sfepy: rezidual: 0.00 [s] sfepy: solve: 0.00 [s] sfepy: matrix: 0.01 [s] sfepy: nls: iter: 1, residual: 7.768306e-14 (rel: 7.646202e-16) sfepy: equation "tmp": sfepy: de_cauchy_strain.2.Omega(u) sfepy: equation "tmp": sfepy: de_cauchy_stress.2.Omega(asphalt.D,u) sfepy: ====== time 1.111111e-01 (step 2 of 10) ===== sfepy: updating variables... sfepy: ...done ====================================================================== Displacement: -0.222222222222 sfepy: updating materials... sfepy: asphalt sfepy: ...done in 0.00 s history update! Traceback (most recent call last): File "/home/grassy/sfepy/simple.py", line 120, in <module> main() File "/home/grassy/sfepy/simple.py", line 117, in main app() File "/home/grassy/sfepy/sfepy/applications/application.py", line 29, in call_basic return self.call( **kwargs ) File "/home/grassy/sfepy/sfepy/applications/simple_app.py", line 112, in call post_process_hook_final=self.post_process_hook_final) File "/home/grassy/sfepy/sfepy/solvers/generic.py", line 216, in solve_direct nls_status=nls_status) File "/home/grassy/sfepy/sfepy/solvers/generic.py", line 138, in solve_evolutionary_op for ts, state in time_solver( state0 ): File "/home/grassy/sfepy/sfepy/solvers/ts.py", line 118, in __call__ state = step_fun( self.ts, state0, *step_args ) File "/home/grassy/sfepy/sfepy/solvers/generic.py", line 116, in time_step_function state = problem.solve(state0=state0) File "/home/grassy/sfepy/sfepy/fem/problemDef.py", line 799, in solve vec = solvers.nls(vec0) File "/home/grassy/sfepy/sfepy/solvers/nls.py", line 183, in __call__ vec_r = fun( vec_x ) File "/home/grassy/sfepy/sfepy/fem/evaluate.py", line 25, in eval_residual vec_r = pb.equations.eval_residuals(vec) File "/home/grassy/sfepy/sfepy/fem/equations.py", line 510, in eval_residuals self.evaluate(mode='weak', dw_mode='vector', asm_obj=residual) File "/home/grassy/sfepy/sfepy/fem/equations.py", line 495, in evaluate asm_obj=asm_obj) File "/home/grassy/sfepy/sfepy/fem/equations.py", line 674, in evaluate ret_status=True) File "/home/grassy/sfepy/sfepy/terms/terms.py", line 1258, in evaluate **kwargs): File "/home/grassy/sfepy/sfepy/terms/terms_base.py", line 81, in _call for ii, fargs in iter_kernel(): File "/home/grassy/sfepy/sfepy/terms/termsLinElasticity.py", line 179, in iter_kernel state=state, get_vector=self.get_vector) File "/home/grassy/sfepy/sfepy/terms/cache.py", line 233, in __call__ out = self._call(key, term, ih, **kwargs) File "/home/grassy/sfepy/sfepy/terms/cache.py", line 215, in _call self.update(key, term, ih, **kwargs) File "/home/grassy/sfepy/sfepy/terms/cachesBasic.py", line 92, in update print kwargs['history'] KeyError: 'history' -- Andre

2 1

sfepy.org domain
by Robert Cimrman 16 Dec '10

16 Dec '10

Hi, you may have noticed that http://sfepy.org, that should point to [1], either does not work, or points to the main google site. That's because something changed, and my old DNS settings (ip 69.56.173.218) do not work anymore. I have tried changing the ip to 209.85.227.100 which is what I found to be the new ip of code.google.com (one of many), but that just redirects to google.com. Is there anyone experienced more with those issues around? r. [1] http://code.google.com/p/sfepy/

2 6

mesh_hook
by Andre Smit 15 Dec '10

15 Dec '10

With the code at [1] I was hoping to retrieve the number of elements from the mesh to use in the problem definition. I'm running into problems - seems mesh_hook isn't processed early enough to define NELS and although mesh_hook reads the mesh from file it fails with this error: grassy@x sfepy]$ ~/sfepy/simple.py its.py sfepy: left over: ['disp', 'young', '__builtins__', 'youngpoisson_to_lame', 'mesh_file', 'poisson', '_filename', 'Eq', 'NELS', 'verbose', '__doc__', 'i', 'mesh_hook', '__package__', 'stiffness_tensor_youngpoisson', '__file__', '__name__'] sfepy: reading mesh (function:mesh_hook)... sfepy: reading mesh (/home/grassy/sfepy.wiki/msh/its2D.mesh)... sfepy: ...done in 0.01 s sfepy: ...done in 0.01 s Traceback (most recent call last): File "/home/grassy/sfepy/simple.py", line 120, in <module> main() File "/home/grassy/sfepy/simple.py", line 113, in main app = SimpleApp( conf, options, output_prefix ) File "/home/grassy/sfepy/sfepy/applications/simple_app.py", line 71, in __init__ **kwargs) File "/home/grassy/sfepy/sfepy/fem/problemDef.py", line 70, in from_conf mesh = Mesh.from_file(conf.filename_mesh, prefix_dir=conf_dir) File "/home/grassy/sfepy/sfepy/fem/mesh.py", line 323, in from_file mesh._set_shape_info() File "/home/grassy/sfepy/sfepy/fem/mesh.py", line 454, in _set_shape_info self.n_nod, self.dim = self.coors.shape AttributeError: 'Mesh' object has no attribute 'coors' Any suggestions on a workaround? thx [1] http://paste.pocoo.org/show/305010/ -- Andre

3 3