Mailman 3 December 2010 - SfePy

Re: Primer
by Robert Cimrman Dec. 14, 2010

Dec. 14, 2010

Great job! I have looked at it only briefly, will try to run the examples as soon as I get to my computer. Thank you, it's a very nice example showing many recent features of the code. r. ----- Reply message ----- From: "Andre Smit" <freev...(a)gmail.com> To: <sfepy...(a)googlegroups.com> Subject: Primer Date: Sun, Dec 12, 2010 17:16 I have completed the update on the Primer at the new wiki location [1]. Please review/revise/edit as you see fit. Thanks [1] https://github.com/sfepy/sfepy/wiki/Primer -- 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 2

Re: Point load
by Robert Cimrman Dec. 13, 2010

Dec. 13, 2010

I am not sure about the line load, but the point load both in 2D and 3D should not be difficult, either by starting with the surface traction term or the spring term. Could you create a new issue? r. ----- Reply message ----- From: "Andre Smit" <freev...(a)gmail.com> To: <sfepy...(a)googlegroups.com> Subject: Point load Date: Mon, Dec 13, 2010 18:42 Yes, I meant the concentrated point load in 2D or a line load in 3D - a term to address these would be very useful. On Mon, Dec 13, 2010 at 9:23 AM, Robert Cimrman <cimr...(a)ntc.zcu.cz> wrote: You can use the same term also in 2D, the traction is then applied to a line (surface in 2D). But if you really mean concentrated point load, then there is no term for this yet. But you could try making directly a zero rhs vector and setting the its components directly. Solution outside simple.py machinery... Does it make sense? r. ----- Reply message ----- From: "Andre Smit" <freev...(a)gmail.com> To: <sfepy...(a)googlegroups.com> Subject: Point load Date: Mon, Dec 13, 2010 15:18 In the linear_elastic_tractions.py example I note that tractions can be applied to surfaces in 3D models. What about point loads for 2D models (without hacking the dw_point_lspring term). -- 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. -- 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.

1 0

Re: Point load
by Robert Cimrman Dec. 13, 2010

Dec. 13, 2010

You can use the same term also in 2D, the traction is then applied to a line (surface in 2D). But if you really mean concentrated point load, then there is no term for this yet. But you could try making directly a zero rhs vector and setting the its components directly. Solution outside simple.py machinery... Does it make sense? r. ----- Reply message ----- From: "Andre Smit" <freev...(a)gmail.com> To: <sfepy...(a)googlegroups.com> Subject: Point load Date: Mon, Dec 13, 2010 15:18 In the linear_elastic_tractions.py example I note that tractions can be applied to surfaces in 3D models. What about point loads for 2D models (without hacking the dw_point_lspring term). -- 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 1

Point load
by Andre Smit Dec. 13, 2010

Dec. 13, 2010

In the linear_elastic_tractions.py example I note that tractions can be applied to surfaces in 3D models. What about point loads for 2D models (without hacking the dw_point_lspring term). -- Andre

1 0

Primer
by Andre Smit Dec. 12, 2010

Dec. 12, 2010

I have completed the update on the Primer at the new wiki location [1]. Please review/revise/edit as you see fit. Thanks [1] https://github.com/sfepy/sfepy/wiki/Primer -- Andre

1 0

Patch - convenience function to remove boundary conditions
by Andre Smit Dec. 10, 2010

Dec. 10, 2010

Attached -- Andre

2 3

Update Primer on wiki
by Andre Smit Dec. 10, 2010

Dec. 10, 2010

I'm updating the Primer tut: https://github.com/sfepy/sfepy/wiki/Primer This is the first of probably many questions: I'm stuck on the last line in the tut as shown. Previously I was able to save the force vector f, as in pb.save_state(file,f) now it fails. f is no longer a State variable? -- Andre

2 6

Re: sfepy wiki
by cimrman3＠ntc.zcu.cz Dec. 9, 2010

Dec. 9, 2010

Lucky guy with a free time :-) ----- Reply message ----- From: "Andre Smit" <freev...(a)gmail.com> To: <sfepy...(a)googlegroups.com> Subject: sfepy wiki Date: Thu, Dec 9, 2010 16:50 np, glad to help, gives me something to do :) On Thu, Dec 9, 2010 at 9:47 AM, Robert Cimrman <cimr...(a)ntc.zcu.cz> wrote: Maybe I messed something up. I was looking for the admin tab at the organization page... and it is at the source repo page. It should be ok now, thanks for starting this! r. On 12/09/10 16:41, Andre Smit wrote: I see it now, didn't previously. On Thu, Dec 9, 2010 at 9:28 AM, Robert Cimrman<cimr...(a)ntc.zcu.cz> wrote: So they are enabled. (Were already, in fact.) r. On 12/09/10 16:14, Andre Smit wrote: If I remember you need to enable the Wikis box under the admins tab and then add a wiki to your repo. Yes, I see I have push rights - thanks. On Thu, Dec 9, 2010 at 9:10 AM, Robert Cimrman<cimr...(a)ntc.zcu.cz> wrote: On 12/09/10 15:44, Andre Smit wrote: Sounds good - I've made a start to show the wiki layout on Github. Probably best if you enable wikis on the sfepy page and give me write access and I'll continue to edit from there or we can just transfer at a later stage: https://github.com/freevryheid/tmpWiki/wiki/Primer Nice! I sort of see how to add people to teams within the sfepy organization, but cannot see how to enable wikis. Write access = I put you among developers with push rights, right? r. On Thu, Dec 9, 2010 at 4:57 AM, Robert Cimrman<cimr...(a)ntc.zcu.cz> wrote: That sounds good! Maybe a good start would be to update your cool primer for current sfepy (as some things changed slightly in the input file syntax, term evaluation was simplified etc.) and put it to the github. The updates to the problem description are as follows: 1. from fields = { 'displacement': ((2,1), 'real', 'Omega', {'Omega' : '2_3_P1'}), } to fields = { 'displacement': ('real', 'vector', 'Omega', 1), } 2. remove 'integrals' and use the order directly in place of integral name in equations/terms: equations = { 'balance_of_forces' : """dw_lin_elastic_iso.2.Omega(Asphalt.lam, Asphalt.mu, v, u ) = 0""", } This is not necessary, but let's make the input as simple as possible. 3. remove 'fe' 4. in postprocessing/probe functions, replace strain = eval_term_op(state, 'de_cauchy_strain.i1.Omega(u)', pb) with strain = pb.evaluate('de_cauchy_strain.i1.Omega(u)') etc. 5. Computing nodal stress can be done now IMHO in a simpler way, something like: i. Create a variable to hold the nodal stresses: sfield = Field('stress', nm.float64, (sym,), omega, 1) svar = FieldVariable('sigma', 'parameter', sfield, sym, primary_var_name='(set-to-None)') where omega is some region, sym = (dim + 1) * dim / 2 is the symmetric tensor storage size (3 in 3D, 6 in 3D) ii. Evaluate the stresses using the custom integral (this remains the same) stress = pb.evaluate('dq_cauchy_stress.ivn.Omega(Asphalt.D, u)') iii. Set the variable - this computes the nodal values given the quadrature point values: svar.data_from_qp(stress, integrals['ivn']) Then calling svar() will return the vector of nodal values. Note that the above is just a "pseudo-code"... I can help you with it after other parts are updated. What do you think about this? r. On 12/08/10 14:31, Andre Smit wrote: Benefits: It may be easier to address, for example, Issue 109<http://code.google.com/p/sfepy/issues/detail?id=109>. Figures can be copied and linked to the git repo. It supports rst (although not exactly the same as Sphinx) but it should be easier to convert to/from documentation. Looks like it provides better latex math support. On Wed, Dec 8, 2010 at 6:13 AM, Robert Cimrman<cimr...(a)ntc.zcu.cz> wrote: On 12/08/10 02:24, freevryheid wrote: Robert What do you think about moving the wiki to github? I have heard a lot of good things about github recently, but have not tried it much myself except the source code management functions. I have nothing against moving more things to github - what exactly are the benefits? So for the moment I am indifferent to this, but any effort towards improving the (wiki) pages is more than welcome. :) -- 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.

1 0

sfepy wiki
by freevryheid Dec. 9, 2010

Dec. 9, 2010

Robert What do you think about moving the wiki to github?

3 10

ANN: SfePy 2010.4
by Robert Cimrman Dec. 6, 2010

Dec. 6, 2010

I am pleased to announce release 2010.4 of SfePy. Description ----------- SfePy (simple finite elements in Python) is a software for solving systems of coupled partial differential equations by the finite element method. The code is based on NumPy and SciPy packages. It is distributed under the new BSD license. Home page: http://sfepy.org Mailing lists, issue tracking: http://code.google.com/p/sfepy/ Git (source) repository: http://github.com/sfepy Documentation: http://docs.sfepy.org/doc Highlights of this release -------------------------- - higher order elements - refactoring of geometries (reference mappings) - transparent DOF vector synchronization with variables - interface variables defined on a surface region For more information on this release, see http://sfepy.googlecode.com/svn/web/releases/2010.4_RELEASE_NOTES.txt (full release notes, rather long and technical). Best regards, Robert Cimrman and Contributors (*) (*) Contributors to this release (alphabetical order): Vladimír Lukeš, Logan Sorenson, Olivier Verdier

1 0