Possible Software Solution - Structural Static and Dynamic Analysis - Frame3DD - FOSS
Eric Marceau
Hello everyone,
I hope that I am bringing good news for those who were looking for this kind of solution.
No, I did not develop it! I've discovered that I am out of my depth from lack of experience. Hence this email!
I came across an application, Frame3DD
https://frame3dd.sourceforge.net/
https://sourceforge.net/projects/frame3dd/
that was developed at Duke's Pratt School of Engineering by Dr. Henri P. Gavin, with contributions from Dr. John Pye of the Australian National University.
I created a proof-of-concept definition file (Icosahedron__1v_iter5.3dd,
6.6 kB) as input for Frame3dd ...
- to verify that I was able to correctly construct the input format file, and
- to confirm that it can handle a problem of the nature of geodesic domes (it does).
While my previously submitted simplified spreadsheets obviously over-simplified (having all structural joints pinned, not rigid), Frame3DD makes the assumption that all joints are rigid, that movements in one strut imposes movement in all other joint-related struts, which is the more realistic model.
This is a sample of what it will display:
To allow people to reproduce these basic results, I've included my input model file for this problem definition.
A basic run creates the report (Icosahedron__1v_iter5_Results.out, 70.9 kB) and an instruction file (Icosahedron__1v_iter5_Results.plt, 376.4 kB) for gnuplot (http://www.gnuplot.info/) to display both the static results and dynamic simulations.
IMPORTANT: The *.plt files are NOT self-contained. They give gnuplot instructions to make use of the many temporary files (/tmp) created by Frame3DD, which are time-slice snapshots of structural positioning, which are played in sequence for the simulations. I am sure there would be ways of reclaiming those for repeated later use by making small tweaks in the *.plt file.
Obviously, my proof-of-concept model ...
- spherical radius of 20 ft,
- geodesic 1v Icosahedron (truncated at bottom pentagon), and
- struts specified to be 1" OD, 0.875" ID steel pipe
is not sufficiently realistic, but it gives a baseline for the approach to using the Frame3dd software, offering a sufficiently small template file to use for reference while trying to construct much more complex/sizable models.
Scanning thru the code for Frame3DD made me realize that I do not have the Civil Engineering knowledge or experience necessary to tackle the task of creating such a tool from scratch.
I do plan to look deeper into the code to identify where I might be able to add some desirable enhancements on "interraction", "display", or "formating", but I categorically make no commitments regarding those at this time.
That fact that an Open-Source tool, with this scope of functionality, which is cross-platform for each of Windows, MacOS and Linux already exists, is something I believe the Geodesic Help Group's Community may wish to consider banding together to cooperatively manage either its own full "fork" of the tool (I realize that may be to big an ask), or putting together some small geodesic-specific "wrapper" tools (scripts or compiled) to perform the pre-processing (creating the formatted input file) and post-processing (converting gnuplot display ouputs into MPEG). Some postings on StackOverflow suggest methods that are promising for the latter. I have submitted a request to Gnuplot to enhance with option to generate an mpeg/mp4 file:
#566 Option to generate animation as video file (mpeg or mp4)
https://sourceforge.net/p/gnuplot/feature-requests/566/
In amongst the membership of the Group, there must be those with the necessary Civil Engineering experience, that would be better positioned to make suggestions regarding the input file parameters, that would be best-suited for the cases (steel pipe/tubing/bars, aluminum pipe/tubing/bars, bamboo beams, wood beams, polyethylene pipe, etc.) that are most-commonly used for the scenarios where the general public are involved in smaller-scale construction (or for more complex cases for those who want to push further afield).
For those who are interested, there is already an interface that works with Sketchup (http://sketchucation.com/forums/viewtopic.php?t=46405), but I am not in a position to comment on whether that works or how practical/useful it might be, but the video
suggest that it might be.
All in all, I believe this tool warrants a closer inspection by someone more knowledgeable or capable than myself.
I hope this information finds a welcoming audience!
I am now stepping away from this self-imposed commitment to the Group, until asked to be of service in any other way.
Eric
Eric Marceau
For those who can manipulate the GnuPLOT script file,
Icosahedron__1v_iter5_Results.plt
directly, to modify the reference to the location of the data files (replace all references of "/tmp" with any other containing the data files), here is a tar file containing that file along with the transient data files used for the structure mode display simulation, to give you a taste of the modal simulations involved (6).
"runSim.sh" is the script I created to "manage" my output files
in relation to my supplied input file.
Eric
Eric Marceau
... in the continuing saga of a structural analysis package
for the masses ...
For some reason, my version of Gnuplot does not offer me the webp option, but it does provide me with the gif option.
Added 5 lines at the top of the *.plt file:
#set term webp size 400,400 animate
#set output 'icosahedron.webp'
set term gif size 400,400 animate
set output 'icosahedron.gif'
{empty line}
I managed to create a GIF animation (3.6 MB), by modifying the start of the gnuplot script file. That took about 10 minutes of running time before exiting. That option will not output the display to the screen, so you would have to run it at least twice, once for the interactive and once to create the "video".
The attached file, again for the same simplistic scenario, will give you a better idea of what to expect.
Eric
Mark Remark
Ashok Mathur
Ashok
--
--
You received this message because you are subscribed to the "Geodesic Help" Google Group
--
To unsubscribe from this group, send email to GeodesicHelp...@googlegroups.com
--
To post to this group, send email to geodes...@googlegroups.com
--
For more options, visit http://groups.google.com/group/geodesichelp?hl=en
---
You received this message because you are subscribed to the Google Groups "Geodesic Help Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to geodesichelp...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/geodesichelp/618472ae-761d-4b8b-be41-a1763413ccd6n%40googlegroups.com.
Ashok Mathur
Or the question dies not make any sense?
Ashok
Dx G
Very interesting. Just in case it escaped your notice, here is another resource that offers some excellent capabilities. Sometimes we hear from Adrian on this group chat and there is also an Antiprism Group. If you are not already familiar with this, you might find it useful, or at least interesting.
http://www.antiprism.com
Adrian Rossiter
adr...@antiprism.com
http://antiprism.com/adrian
anti...@googlegroups.com
Mark Remark
Ashok Mathur
Ashok
To view this discussion visit https://groups.google.com/d/msgid/geodesichelp/66f47ed7-83c2-4373-be60-b38102298ab3n%40googlegroups.com.