Shading with daylight: development and testing of a parametric model to facilitate design choices of shading devices

Abstract

This study aims to develop and test a parametric model to be used in the early design stages of shading devices. With the adoption of multi-criteria analysis, the tool seeks to optimize solutions in order to meet criteria that are apparently conflicting: blocking solar radiation and providing the incidence of daylight. The geometric model consists of a room with one window and also the possibility of an external obstacle. Four types of shading devices can be selected and adjusted by users. Three performance objectives were defined for multicriteria evaluation with genetic algorithms: to maximise the number of points between 300 and 750 lux, to reduce the incidence of direct solar radiation on the window surface, and to increase average illuminance levels. In the first stage, 35 generations of cases, each with 100 individuals, were calculated. The simulation results are plotted in a three-dimensional graph that has an axis for each performance goal. Therefore, different areas of this graph have different performance features and, consequently, varied solutions. Based on their relative position on the graph, six cases were selected to be further investigated with annual daylight simulation. The results indicate that, in the simulated scenario with a generous window facing west, there is convergence between the lighting and shading objectives: the lower the incidence of solar radiation, the higher the average percentage of hours between 100 and 2000lux and between 300 and 750 lux.

Keywords:
Generative design; Sun shading; Daylight; Multi-criteria analysis