Thanks Robert!

I'm working on the revisions and updates and will get back. The abstract is due at the end of January but I hope to get the bulk of the paper done by then as well, which is due in Feb sometime i.e. if abstract is accepted.

On Mon, Jan 17, 2011 at 7:13 AM, Robert Cimrman <cimr...@ntc.zcu.cz> wrote:
At [1] I have posted the code with "correct" avgqp() function. Yours was ok, as long as all the quadrature point weights were the same.

The new implementation defines an auxiliary material with the quadrature point data, and calls the de_volume_average_mat term. Note that the order of the quadrature has to be the same as the one used for evaluating the strain.

Then it seems to me, that you could avoid all those loops in
functions Ft(), Fc(), Ec(), by using numpy vectorized operations, i.e. numpy.exp() instead of math.exp(). Also the failure loops (around line 193) could be IMHO vectorized using numpy.where() and fancy indexing.
Try it as a NumPy excercise :) - you get big speed gain by this, and also the code would be shorter and more readable. I can give you a hand if you get stuck, of course.

cheers,
r.

[1] http://paste.pocoo.org/show/322183/


On Mon, 17 Jan 2011, Robert Cimrman wrote:

Hi Andre,

(sorry for the delayed response - I'm accommodating in Fribourg again, and have not yet wifi at home...)

On Fri, 14 Jan 2011, Andre Smit wrote:

Robert

I think I have something to share. How would you feel about co-authoring
a paper to this conference - mainly to further expose SfePy but also to
check that I'm not applying it incorrectly?

That would be great, thanks for proposing it!

Using the strain rate dependency example you provided I recoded an
example looking at the failure of a cylindrical specimen under
compression. The cylinder is made up of elements that I treat
individually or discretely, checking the strain rates in each and using a
relationship established in the lab, I calculate the compressive
strengths in the elements based on these strain rates. If the stress in
an element exceeds its strength then the element fails and I assign a low
stiffness to that element. The code I'm using is available here:
http://paste.pocoo.org/show/320646/

I will look at the code and try it ASAP. What is the deadline, anyway?
A few notes straight from my head follow.

What happens with the linear elastic assumption when you reduce stiffness in some elements? The code has some hyperelastic terms, in case they would be needed.

Would you mind looking and commenting on this code. I'm a bit skeptical
about the avgqp function which averages the strains at the quad points.
Note that I use a normal distribution to vary the moduli of the elements
initially.  

To get averaged element values of something defined in quadrature points, one can use either directly de_volume_average_mat term (a fake material definition would then be needed, I can show you), or use directly similar code to the one in the body of that term.
 
I've attached the mesh: cylgeo.vtk.

What I aim to do with the paper is demonstrate the influence of specimen
geometry on strength. For simple compression and tension tests (on
cylinders) this is trivial but with indirect tensile tests, for example,
there is both compression and tension failure, and I hope to show using
FEM that the strengths of materials determined using these tests are
significantly influenced by the geometry of the specimen.     

Interesting!

The force-displacement curves currently generated by the code look
promising. I'm not sure the analysis can be refined by using smaller
elements?

The mesh is not very fine, yes. Another way to refine the analysis might be to use another material relation (hyperelastic?).

r.

On Fri, Jan 14, 2011 at 3:19 AM, Robert Cimrman <cimr...@ntc.zcu.cz>
wrote:
     I see :) Good luck then!

     r.


On Thu, 13 Jan 2011, Andre Smit wrote:

     No plans currently - still working on an analysis.
     Maybe if I can get it
     done before the end of the month :)

     On Thu, Jan 13, 2011 at 2:43 AM, Robert Cimrman
     <cimr...@ntc.zcu.cz>
     wrote:
          Hi Andre,

          are you planning to go there?

          r.


     On Mon, 10 Jan 2011, Andre Smit wrote:

          FYI:

        
      http://congress.cimne.com/CFRAC2011/frontal/default.asp


          --
          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...@googlegroups.com.
To unsubscribe from this group, send email to
sfepy-devel...@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...@googlegroups.com.
To unsubscribe from this group, send email to
sfepy-devel...@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...@googlegroups.com.
To unsubscribe from this group, send email to sfepy-devel...@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...@googlegroups.com.
To unsubscribe from this group, send email to sfepy-devel...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/sfepy-devel?hl=en.




--
Andre