| Title |
Integrated Design for Rooftop Daylighting of Sports Hall by Multi-purpose Optimization Process with daylighting and energy performance |
| Authors |
이병연(Lee, Byungyun) ; 박서우(Park, Seowoo) |
| DOI |
https://doi.org/10.5659/JAIK.2022.38.10.119 |
| Keywords |
Useful Daylight Index; Daylight Glare Probability; Annual Energy Unit Index; Multi-purpose Optimization; Meta-model |
| Abstract |
This study aimed to analyze multi-purpose optimized design alternatives for sports halls to minimize total energy consumption, maximize
daylight quantity, minimize glare probability and develop meta-models to predict energy and daylight performance during the early design
stage. The automatic optimization tool of Modefrontier integrated with the rhino-grasshopper model was developed and simulated with the
energy plus and radiance engine. Three optimization phases were conducted, and the variable ranges and optimization algorithms were
selected for each phase's aim. In the first and second phases, the optimized cases were selected in the Pareto surfaces and compared to
analyze the influence of glare prevention on the best-performing cases. Lastly, the meta-model was developed and presented to predict energy
and daylight performance with a variation of the three most sensitive variables to predict the performance without energy simulation by
architects and all participants. The rooftop daylighting model with cone-type lightwell was selected for the analysis with four geometric
variables and two material variables for parametric design development. The results revealed that the window-floor-ratio was a dominant
variable for all energy, useful daylight index, and daylight glare probability followed by tilting height and lightwell height. The window-floor
ratio in the Pareto-optimized cases ranged between 11 and 21 percent in the first optimization without the glare-free objective; the range was
reduced to between 8.5 and 14 percent. The range of lightwell height shrunk between 360 and 480 mm to between 240 and 360 in relieving
glare. The developed response surface model with restricted window-floor ratio of 9.5-13 percent is expected to provide relevant information
for future decision-making purposes. |