Orientation study: Rotation Tower

So Thanks again for the impressive tool!

I’m doing my thesis on generative process in tower design; my approach is to generate morphology from the environment data, starting with the radiation analysis. My aim is to find the best orientation for minimize the total radiation and manage the one that i got on the corners with vegetation.

As you can see it is working great! But I've two questions

0_ how can I increase the resolution on the analysis's mesh? It depends of the brep?

1_ I see that you did an orientation capsule, but I didn't manage to adapt the rotation of a tower. So I did it with Galapagos but it was slow, Do you have any thought of how it could be with lady bug tools?

pd: the tower next to it is the context

Gracias

Cheers

José Miguel

Attachments:

0_1_base giro_radiacion.gh, 60 KB

Replies are closed for this discussion.

Replies to This Discussion

Permalink Reply by Abraham Yezioro on May 7, 2013 at 11:28pm

The resolution is defined in the LB_Radiation component in the item "gridSize".

As for the rotation see the previous Mosthapa's post (Orientation Study). It is defined on 2 angles: The total rotation you want to simulate (let's say 180) and the step angle of this rotation (i.e. 45, so you'll get 5 cases).

-A.

Permalink Reply by Berk Ekici on May 8, 2013 at 4:26am

Hi Jose,

I prefer using Octopus for multi-objective optimization with SPEA-2 evolutionary algorithm. You can find the optimum solution for solar radiation, dayligh calculation or heating and cooling periods. Here, you have one objective which is just solar radiation. In my opinion, there should more objectives for tower generations which makes it better! Please check the video for using Octopus with Ladybug...

Best,

http://www.youtube.com/watch?v=6c32kZN19FU

Permalink Reply by Mostapha Sadeghipour Roudsari on May 10, 2013 at 10:34pm

Hi Jose!

Sorry for the slow reply! Fortunately Abraham and Berk already answered the questions. Here is my 2 cents:

1. In your definition you don't need to connect the test geometry to the context. The component does consider the test geometry as the context as a default.

2. As you can read in radiation analysis component description the result for indoor studies won't be accurate. I assume in your case you can run the studies on the building envelope instead of the floors and optimize the geometry?

3. Really interesting topic for a thesis. I wonder if you have already read Nathaniel Jones' thesis. It has a really good introduction to optimization and Genetic Algorithms: http://www.graphics.cornell.edu/~nlj5/thesis.html

I wish I had the Honeybee developed enough to share it with you for your studies. A project like this is the right fit for using Honeybee.

Hope it helps!

Best,

Mostapha

RSS