Faculty of Architecture, Technical University of Berlin
The project tackles the problem of cross-domain optimization to enhance visual comfort of users by optimizing the geometry of the facade panels. The western facade is highly exposed to the afternoon direct sunlight which causes significant glare effect. Architects proposed facade made up of two lids-upper-south and bottom-north to achieve the most comfortable working conditions.

Project Year
2018
Location
Berlin
Size
17.000 sqm
Role
Optimization Expert
Client
Built by Data Consortium
Summary
The goal of the project is to find the best position of the facade covers to enhance visual comfort. The study looks for a well-lit space which is defined by the availability of daylight, exposure to the direct sunlight, position of sun patch and view to the outside.
Daylight plays a key factor in designing workspaces. Various researches proved that in the locations under 50° latitude there is no need or very limited need of artificial light (fig.01). Berlin lays on 52° N which means that well design workspaces and accurate work schedules assure that during the 80% of occupancy time there is no need for artificial light. The study looks to find a solution that would enhance the daylight performance of the interior space.

The new facade is a second skin offset from the original by 100 cm. Each windowed unit receives a new cover. The cover is divided into two parts south-top lid (STL) and north-bottom lid (NBL). The test has proven that the north-bottom lid influences the solution to a very low percentage. For that reason, the optimization process considers only STL. The STL is built out of 4 connected triangles. The control points of triangles move in predefined domains to enhance visual comfort. The preliminary optimization supplied the optimization process with fenestration ratio for each unit and is based for further optimization and design.

Optimization Goals
Minimize the Annual Sunlight Exposure
Maximize Daylight Autonomy
Variables
The position of the south-top cover points, that moved in 3 dimensions space. The control points are located on the upper lid of the new cover.
Constants
Angle and maximal point movement depending on the lid type
The optimization process optimizes the geometry to maximize Daylight Availability and minimizes Annual Sunlight Exposure. A posteriori test on Daylight Glare Probability is conducted to verify the results. To measure
Daylight Availability the average DA for all pre-selected random and representative points is calculated. To minimize Annual Sunlight to the exposure the algorithms minimizes the ASE value for the worst performing test point.

In the result below the floor, points were evaluated for ASE. 4528 points present ASE(1000lux, 250h) which makes 52% while only 10% is acceptable. A DGP test run on the most overlit space proves that adequate interior layout can minimize the negative effect of glare. Also, interior partitions play important role in decreasing DGP. Use of full height non-transmissive internal partitions revealed that the daylight conditions significantly enhances.
