Valid boundary conditions for partial model

Project Support
1

Good afternoon,
I want to apply pressure to the exterior normal of my model in order to assess the global buckling behavior. I have already reduced/simplified my model significantly, but it is still a challenge to get it meshed and analyzed. I was able to mesh a more complex version of it, but the analysis ran out of resources. So I reduced it to the current form (fewer than 1/2 the previous faces), but so far have not been able to mesh it. Even after that, I suspect a nonlinear static analysis is going to run out of resources still.
There is a lot of symmetry, but I don’t see that the boundary conditions for a reduced model can be made valid, since it is a closed structure, and the boundary would be supplied by the truncated portion of the model. Fixed, or elastic conditions would not accurately reproduce the affect of the missing portion. Maybe if there was a means to translate one boundary motion/forces to the opposite side, effectively connecting the edges with appropriate angular adjustment.
A secondary concern for a reduced model, is that the ratio of columns to cables for a complete model is 1:1, but the ratio for a partial model is never balanced, since it always contains one more column than cable sections.
Any suggestions on how I can perform a global buckling analysis on this? The point is to maximize specif strength of the structure by optimizing the cable bracing. I,e., to achieve simultaneous column yield and bucking with the most efficient bracing possible.

Reduced model

2

Hi there,
Thank you for reaching out to us on the Forum :slight_smile:
I will have a look at the project and get back to you soon.

3

Thanks for taking a look at it.

To state my prior ramblings a bit more concisely, I believe that one possible solution would be to permit a bonded contact between remote faces. Then I could slice the boundary columns in half longitudinally, and stick the separated ends of the model together with said imaginary bonded contact. This would both transfer the stress, and enforce the 1:1 column to cable section ratio.

Paul

4

Any word on how to make this simulation succeed? The model is likely to get larger and more complex as it progresses. It would help a lot with time and resources if it can be sliced into segments.

6

Hi Paul, I have asked my team to consult on this and someone should get back to you soon! I appreciate your patience Thank you!

7

Hi Paul,
I appreciate your patience and understanding. Unfortunately I do not come bearing good news. I’m afraid this is analysis is just not possible with SimScale at the moment. We would need:

  • buckling analysis
  • Beam and Rod elements

At SimScale we take our user’s opinion and needs quite seriously. This is why we’ve put together a Roadmap page . You can go here and vote on the feature you want.

Your interest in using SimScale for this analysis is highly valued, and it’s regrettable that we cannot accommodate your request at this time. We hope that you’ll continue to use our software for other projects and analyses where it can provide the solutions you need.

Thank you and have a nice day :slight_smile:

8

Ok, I thought nonlinear analysis would cover buckling.
Thanks to you and your team for investigating this.
Regards,
Paul

1 Like