Optimization of Light Shelf Performance by Genetic Algorithm

Abstract

The use of daylight in buildings with energy efficient consciousness is possible by studying the effect on the heating and cooling loads as well as the effect on the artificial lighting load. It is difficult to make comprehensive assessments because the conditions are variable according to climate, season and time of day. The aim of this study is; comprehensively evaluating the design parameters of light shelf systems and energy consumption and optimizing them. In this study, the literature about evaluating the performance of the light shelf systems were researched. Then, the simulation and optimization method used for the research is given and the workflow created by using Grasshopper which is the visual coding plugin of the Rhino software is declared. The simulations are carried out by the Ladybug/Honeybee plugins which is an interface of Radiance and EnergyPlu simulation engines and modelling steps are explained. In the first stage of this is; the case of only interior or only exterior light shelves is examined by using the simulation method. 810 options were tested for 9 different volumes and 90 different light shelf situations. Light shelves’s contributions to natural lighting and its effects on artificial lighting, heating and cooling loads were evaluated together. In this stage, a systematic review was made that could not be observed during the optimization phase. In the second stage, in order to optimize the performance of the light shelves by genetic algorithm, a solution set with 6.900.000 elements was created for 6 different volume scenarios and 1.250 solutions were evaluated to create the best solution set. In the set of optimal solutions, the solutions with the best performance in terms of daylight performance and cooling, heating and artificial lighting loads were examined, and the set of elements that were better than the base case where there was no light shelf was filtered.

In the study, in addition to natural lighting efficiency; the effects of tilt angle, shelf height, shelf depth, volume depth and the transparency ratio of the façade is evaluated, in terms of energy consumption by using the simulation method. This study shall be a new perspective for evaluating efficient use of light shelves in consideration of the energy. The optimization made with the evolutionary algorithm, practical solution suggestions for the designers are presented.