Question about doing optimisation in Grasshopper with Femap

Hello all the experts,

I am doing my graduation work and want to use grasshopper to do optimisation. At the same time, I want to use Femap to do the FEM analysis. So far I am at the beginning of my graduation work and I know very few about scripting.

Can anyone give me some clue what should I do to connect Grasshopper and Femap.

Plus, for optimisation(to get a light-weight structure) , can I use Galapagos?

Sincerely,

Xue

Replies to This Discussion

Permalink Reply by peter fotiadis on November 21, 2015 at 1:09am

Also:

Attachments:

Flatten_V1.gh, 38 KB
Drape_on_something_V1.gh, 43 KB

Permalink Reply by peter fotiadis on November 21, 2015 at 1:50am

Since depth in answers is rather limited, I start from node 0:

1. You remember that I've mentioned tensile membranes? (and you provided ... er ... an explanation about what tensegrity is, NOT actually required, he he).

2. So the paranoid thing that you have in mind is:

2.a. Create a minimal/relaxed surface (or a minimal/relaxed line graph) via Kangaroo by additionally using gravity (for "shaping along" a desired vector).

2.b. Using division points (or line graph node/grid points) create a tensegrity Truss.

2.c. Using any method ... do some FEA on that.

Hope that you understand why this is BAD engineering approach to the max: a minimal surface BEGS for some tensile membrane NOT a tensegrity truss. T trusses are just a very expensive way to skin a cat mostly used (real-life, real-projects, real-people) in planar glazing systems or in some artistic/decorative stuff/objects. Of course you can "mix" the 2 worlds ... but is bad engineering AND a pointless overkill AND kitsch AND very expensive.

Permalink Reply by Zhengyu Xue on November 21, 2015 at 9:51am

Holly Jesus!

You Are AMAZING!! I never expected someone online could offer me such tremendous help!

I will read all your messages carefully and follow your instructions.

I was a bit depressed because today I saw a close friend off back to China(I am in Netherland). Now I feel much better after reading all your helpful reply.. I am a man need either love, or friendship, or a sense of achievement/accomplishment...Anyway, this about this research got a bit further clear. You cheered me up.

I wish I could finish form finding part in the coming weekdays.

Wish you all the best , prettige weekends, and good night.

Sincerely,

Someone who respects omnipotent Lord Sardineland.

Permalink Reply by peter fotiadis on November 21, 2015 at 11:21am

OK, now we are in the path (I mean: we are in the rabbit hole).

BEFORE attempting anything:

1. Have you got the gist of what a real-life T truss is? (cables add an ENORMOUS level of complexity thus such trusses are rarely used in real-life AEC projects). I mean the parts that make the damn thing not Academic lines and abstract tubes.

2. If so (and having in mind my reply in the other thread of yours) forget for the moment relaxing this and relaxing that > focus to the end of the road: the Truss itself. Form finding is 0.01% of the whole task, the nuts and bolts are 99.99%.

3. If the T thingy it's made out of NON flat division (grid) points then the required system is rather complex and expensive: something that could achieve rigidity of the X whist's been capable to adjust according to a variety of required levels of freedom (I'll post soon such a system that I've used in a project long time ago, made with CATIA and Microstation).

In an ideal world (in planet Utopia, what else?) and IF you were a bit advanced in C# I would suggest to write the code that does the truss (and forget FEA et all). I mean a real truss with real components.

This should been sufficient enough for ANY thesis in ANY Academic institution.