How to use the pipeline postprocessing

About the development of the FEM module/workbench.

Moderator: bernd

User avatar
HarryvL
Posts: 275
Joined: Sat Jan 06, 2018 7:38 pm

Re: How to use the pipeline postprocessing

Postby HarryvL » Wed May 16, 2018 10:11 pm

I am looking into it Thomas. For a full 3D analysis we may need plots for shear utilisation (based on Mohr Coulomb, as described earlier, to check crushing and shear failure) and principal tensile stress fields (magnitude and direction). All of that can be displayed in Paraview through manipulation of available FC/CCX data. However, the issue I see is that this information still does not uniquely define required reinforcement percentages. Even if you would limit reinforcement to x, y and z directions only (which is not a necessary restriction), then you would still have freedom to choose more of one or the other to meet the tensile utilisation requirements. A simple plot of required reinforcement percentages in the 3 directions may therefore be out of reach, unless we make further constraining assumptions. In due course I hope to show this with the formulas. I am also looking through the literature on this.

So in summary, I think the Paraview challenge is perhaps not the biggest one :shock:
User avatar
HarryvL
Posts: 275
Joined: Sat Jan 06, 2018 7:38 pm

Re: How to use the pipeline postprocessing

Postby HarryvL » Thu May 17, 2018 6:23 pm

Concrete Design.jpg
Concrete Design.jpg (239.83 KiB) Viewed 147 times
User avatar
HarryvL
Posts: 275
Joined: Sat Jan 06, 2018 7:38 pm

Re: How to use the pipeline postprocessing

Postby HarryvL » Thu May 17, 2018 8:03 pm

Note: σy in the above is the yield stress of steel.
User avatar
HarryvL
Posts: 275
Joined: Sat Jan 06, 2018 7:38 pm

Re: How to use the pipeline postprocessing

Postby HarryvL » Fri May 18, 2018 12:16 pm

I did some further research and found that the above problem has been solved elegantly by my ex colleagues in the Netherlands ...

http://heronjournal.nl/53-4/3.pdf

The paper provides 11 closed-form expressions for reinforcement ratios ωxx, ωyy and ωzz that need to be evaluated for every stress point in the result set. The one to chose is the one that ensures that σc1<0 (no tension in any direction), is valid (ωxx, ωyy, ωzz >0) and minimizes total reinforcement ωtotalxxyyzz.

I will give it a try and report back.
thschrader
Posts: 902
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: How to use the pipeline postprocessing

Postby thschrader » Fri May 18, 2018 2:41 pm

HarryvL wrote:
Fri May 18, 2018 12:16 pm
I did some further research and found that the above problem has been solved elegantly by my ex colleagues in the Netherlands ...

http://heronjournal.nl/53-4/3.pdf

The paper provides 11 closed-form expressions for reinforcement ratios ωxx, ωyy and ωzz that need to be evaluated for every stress point in the result set. The one to chose is the one that ensures that σc1<0 (no tension in any direction), is valid (ωxx, ωyy, ωzz >0) and minimizes total reinforcement ωtotalxxyyzz.

I will give it a try and report back.
Harry, thanks for your effort on this topic. But I think this is a very complicated (and time consuming)
approach. What I need is an "optical" support for adjusting the rebars. I mean this:
With the program "ForcePAD2"
http://forcepad.sourceforge.net/screenshots.htm
you can visualize the stresses "flowing" through the body. Thats what I need. I tried this with paraview
on the vtk-files from FC-FEM, but I dont get it. I want to see the 3D stress-vectorfield.
ForcePAD2.JPG
ForcePAD2.JPG (39.78 KiB) Viewed 100 times
ForcePAD2_2.JPG
ForcePAD2_2.JPG (52.68 KiB) Viewed 100 times
thschrader
Posts: 902
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: How to use the pipeline postprocessing

Postby thschrader » Fri May 18, 2018 3:40 pm

OK, here is an other example.
I calculated a 12m concrete beam with FC-FEM. I will see the syy-stresses as arrows and used the glyph filter.
But there is a problem with the orientation. In my opinion all syy-vectors should be exactly parallel to the y-axis.
Under "active-attributes/vectors" you can select between disp or none. When using none, all arrows
are parallel to the x-axis. When using disp, you see the arrows in the picture.
Maybe I am using paraview completely wrong... :oops:
T_Beam_paraview.JPG
T_Beam_paraview.JPG (178.89 KiB) Viewed 93 times
User avatar
HarryvL
Posts: 275
Joined: Sat Jan 06, 2018 7:38 pm

Re: How to use the pipeline postprocessing

Postby HarryvL » Fri May 18, 2018 4:01 pm

Thomas, looking at the entries you made I think the vectors represent the displacements and the colors Sigyy
User avatar
bernd
Posts: 5582
Joined: Sun Sep 08, 2013 8:07 pm
Location: Zürich, Switzerland

Re: How to use the pipeline postprocessing

Postby bernd » Fri May 18, 2018 4:14 pm

thschrader wrote:
Fri May 18, 2018 2:41 pm
Harry, thanks for your effort on this topic. But I think this is a very complicated (and time consuming)
approach.
May be not if someone would be able to implement this somehow in FreeCAD or paraview result viewing. I assume this is what you have in mind harry?



@Thomas:
this one ... https://forum.freecadweb.org/viewtopic.php?f=18&t=24637
User avatar
HarryvL
Posts: 275
Joined: Sat Jan 06, 2018 7:38 pm

Re: How to use the pipeline postprocessing

Postby HarryvL » Fri May 18, 2018 4:52 pm

Yes that's what I will try Bernd. The aim is to have 4 plots. 3 reinforcement ratios x, y, z and one for shear/crushing. This would allow judging optimum reinforcement requirements for 3D stress states.
User avatar
HarryvL
Posts: 275
Joined: Sat Jan 06, 2018 7:38 pm

Re: How to use the pipeline postprocessing

Postby HarryvL » Sat May 19, 2018 12:07 pm

I opened a new topic dedicated to the topic of plotting the concrete reinforcement ratio: https://forum.freecadweb.org/viewtopic. ... 21#p234582