ARCH 655
In project two, it has been tried to first, create the major parts of the Chameleon Office using Python script, and then, optimize the opening and protrusion of the modules using GA optimization of Galapagos in Grasshopper.
Grasshopper is used to create the hexagon mesh and basic structural components and the interior parts of the building are modeled in Rhino.
 |
| Grasshopper model |
Script
The following screenshots of this section show the Python script used to generate the main components of the facade:
 |
| Defining the parameters |
 |
| Creating one triangular module |
 |
| Creating the whole hexagon component |
 |
| Creating the whole facade by copying hexagon module |
 |
| Adding the components at edges |
 |
| Applying the script on one surface |
 |
| Adding extra components from Grasshopper |
Optimization
Optimization Parameters:
The objective of utilizing optimization technique in this project is to find the near-optimum solution based on different criteria for daylighting. Consequently, two main parameters are recognized as the dominant parameters that can impact daylighting performance of the building. Opening scale of the modules (WWR) is one of the effective parameters. Furthermore, the protrusion of the components may act as shading device in the exterior side of the upper parts of each component and as light shelves (reflectors) in the interior side of the lower part of each component. Therefore, the value of the protrusion can be seen as an important variable in the optimization.
In order to simplify the process, only the offices adjacent to one face of the building are taken into account for the optimization.
Fitness Function:
In order to evaluate the daylighting performance of the building two metrics are considered to ensure that the optimization engine accounts for both daylight availability as well as glare. Spatial Daylight Autonomy (sDA) and Annual Sun Exposure (AES), which are represented as a percentage, are chosen as the metrics to evaluate each individual case. As the main objective is to maximize the sDA and minimize the ASE, and Galapagos accepts one Fitness Function, the two metrics are combined to feed Galapagos with a single value. Also, as the ASE was small based on the configurations of the building and the orientation, it is multiplied by 5 to be taken into account as a more strict objective in this project. The following formulae is used as the fitness function in this study and the objective is to maximize the Fitness Function:
Fitness Function = sDA + (100 - 5 * ASE)
 |
Best case of the first population |
 |
| Best case of the last population |
 |
| Verification of the results of Galapagos |
 |
| Sample screenshot of the sDA analysis of the whole floor |
