I've don'e a lot of programming between Rhino and Structural analysis programs, but not Ansys yet (if there's enough demand, I'll look at it in the new year), but mainly for skeletal models. This includes intercepting the basis of a structural model generated in Grasshopper.
If you want to learn more, or discuss it further, feel free to get in touch. Discussions with a few in Academia have so far led to them trying the free academic version of GSA instead of using Ansys. Horses for courses, depends on what you are trying to model.
Thanks for your response. That and your blog are intriguing me now.
Im following a course (architecture-londonmet) for which I have set out to model a small pavilion/shelter design with the help of grasshopper which is site and location responsive (mainly program, topography, view and access). I am currently working on the design as its not there yet (just floor elements) As a second portion to this my intention is to bring the top half of the shelter's canopy into ansys (I am open minded so it could be or other) and do some structural tests on it which could inform the modelling of the construction parts by adressing issues of minimizing materiality though maximising friction(of slotted parts) Think along the lines of alvaro Siza and de Moura's serpentine pavilion.
Though this is all very new to me im keen to give this a run, my idea was something along the lines of where the derformations/stresses are greatest this can be fed back into grasshopper (via excel?) and scale up components this way. So to conclude; I am wondering what (in a nutshell) the way is to do this.
I am aware that obviously if this is too ambitious given time constraints the research question will involve not feeding the information back (drawing conclusions) although I really would like to..
Chris
Permalink Reply by taz on December 18, 2009 at 8:45am
Chris,
If you're interested it may be easiest if you try to utilize all the great work that Jon has done.
If you're trying to integrate structural analysis into design be aware that Ansys isn't really a program used to model structural systems. If Ansys is ever used for structural engineering it's for very, very special details or in acadamia for very, very detailed study.
We make extensive use of ANSYS for structural design. We take advantage of an extension called CIVILFEM (www.civilfem.com) It makes life so easy on ANSYS world for structural engineers.
As you might know ANSYS is parametric with its own script language APDL. So we have developed a workflow were inputs are taken from EXCEL and outputs are sent back to EXCEL. When the results are huge MATLAB is used to filter in order to avoid the row limits from EXCEL.
The idea of connect with grasshopper sounds interesting. I will give it a try. If something useful come out I will post it....
Good luck, it sounds like a great project and it's a popular field at the moment. I've seen on Twitter and online a few firms progressing and researching this area. I'm happy to help/offer advice as I can.
A couple of quick points, you'll soon find out you can spend lots of effort programming this type of thing. Like in a design office, you'll have time (budget) constraints that will effect how thorough a result you can achieve. You can do something quite powerful using the GSA export of data/results to a spreadsheet with some manual intervention/supervison. You can sort elements by resulting stress/force and manually allocate propertys in groups.
At Expedition, this was the process we used 4 years ago for Vauxhall Towers http://www.expedition.uk.com/index.php?pid=83 and you can do something really powerful without expending lots of effort on programming.
Over the 4 years we evolved a number of tools and routines, and used a lot of them for something with Genetic algorithm decision making automating the design of Santa Maria Del Pianto Station http://www.expedition.uk.com/index.php?pid=141 There was a lot of progamming expertise and effort that went into this, and computing power.