A FASTER MULTI-CRITERIA OPTIMIZATION MODEL TO PARAMETRIC FACADES FOR DAYLIGHTING AND THERMAL PERFORMANCE

Abstract

In the Performance-based design method the 3D model is connected to computational simulation processes and an optimization process is associated with a specific design problem. This work addresses the design of facade solution, which have problems associated with the architectural programing formulation, interoperability with the architectural 3d model, and slow computer simulation processes. The objective of the paper is to present a simplified strategy for faster optimization processes of shading devices and window solutions based on thermal performance and daylighting requirements for an entire year in Sao Paulo, Brazil. The multi-criteria optimization model integrates three performance indexes: solar radiation in the window plane in hot periods, UDI (useful daylight illuminance), and area of the shading devices. The computational model is based on the Radiance two-phase method that was up to 10 times faster than the Daysim based methods. The process uses parametric modeling through Grasshopper, and presents a solution sensitive to complex geometries and choice of glass to assist the performance-based design process. It was performed 10 generations of simulations. The optimized solutions were evaluated and presented a gain that reach 62% in the reduction of material proposed for the shading devices.