2014年11月30日星期日

JIAHE BIAN Arch655 Final Project


Basing on the model I built in project 1, more sliders are added to control the size and shape of the structure.

After that, I reorganized the grasshopper nodes and data structure to make the model works more efficiently.

1. Question 1

If the structure would be built in real projects, the sizes of the structure needed be changed depending on the space project can provide. In another word, the problem is how to make the structure fit into the space, if certain area of space is given?
For example, when the structure need to be built inside a circle area whose diameter is 150, how to adjust those sliders to make that happen.
Adjusting slider F can dramatically influence the overall shape of this structure. In order to solve this problem, I use genetic algorithm to calculate the best solution. 


 Top view is the best view to see the process of fitness.

2. Question 2

After deciding the overall shape, some panels will be installed onto the structure to protect people from sunshine in summer. But if all of the structure will be covered by panels, it is not only a waste of money, but also makes the pavilion look dull. 

Using panels to cast shadow at where people always stay (the S shape) is efficient for keeping people away from direct sunshine.


But the positions of sun are variable, and it depends on the date and time. So a python node is used to calculate the position of sun and the light path which would light up the S shape in college station (30.6N).



According to the calculation, the algorithm can decide which panels should be closed.
 The solution will be changed according to the input time and date.





Reference:

http://www.esrl.noaa.gov/gmd/grad/solcalc/solareqns.PDF
http://www.jgiesen.de/astro/suncalc/calculations.htm
http://agwestinternational.wa.gov.au/objtwr/imported_assets/content/_archive/tr137.pdf
http://orbitingfrog.com/2009/12/21/the-winter-solstice
/





2014年11月3日星期一

Jiahe Bian ARCH655 project1

The model I build is swoosh pavilion.
The basic concept of my model is I divide the structure into two categories. The first one is column and another one is beam.
I start from two curves.

I divided the first curve into A parts, and the second curve into B parts.
I am thinking that the columns are parts of semi-circles. So I build several semi-circles at first. The radiuses of those semi-circles are from 20 to 45.
And then I build a solid to trim those curves, and get the first lines of columns.
With Similar process, I get the second lines of columns.
With those two lists of lines, I get the columns solids.
After the columns, I start to build the beams. At first, I divided each semi-circle into C parts. And get first line of beams (before cull).

I offset the first lines and get the second lines of beams; the width of beams can be adjusted by slider D.
Using extrude node and cap holes node, I get the beams (before cull). And the thickness of beams can be adjusted by E slider.

Using the same solid I used before to evaluate those beams, and get the beams which are totally inside the solid.
 
Next, I rotate it and get another part.

The last step is to build some beams which can connect those two parts.
When A=14, B=17, C=60, D=0.0018, E=4, this is the top view of the model.

When A=11, B=10, C=60, D=0.0018, E=4, this is the top view of the model.
When A=11, B=10, C=60, D=0.0018, E=1, this is the perspective view of the model.
When A=11, B=10, C=60, D=0.0018, E=5, this is the perspective view of the model.

I build the soil which has same thickness. The volume of it is 64976.11.
But using this structure, when A=14, B=17, C=60, D=0.0018, E=4 the volume of all the beams and columns is just 14005.03.
So I can save 78.4% of material.

The final analysis is the force analysis. The result shows that the columns are too long, those columns cannot afford those own dead weight. So I need to cut some parts of the columns.