Parametric modeling and building performance: lighting and energy simulation based optimization through genetic algorithms

Download

Performance-based design is being increasingly used as a design methodology in architecture. 21st Century computational advances allow the development of software and tools capable of integrating parametric modeling and energy efficiency processes in buildings. The objective of this work is to evaluate the potential of DIVA on energy analysis. DIVA is a software capable of performing dynamic light simulation, which is integrated with the energy simulations of Archsim/EnergyPlus, through the availability of daylight in the interior rooms. The use of genetic algorithms during the optimization process was also evaluated, considering the time and the pattern of convergence of the parameters and objectives. Therefore, a multiobjective optimization problem was elaborated in order to maximize the occurrence of daylight and to minimize energy consumption with artificial lighting and air conditioning systems. Two cases were parameterized, the Base Case corresponding to the real form of the building and the Initial Case, corresponding to the building with modifications in the main façade openings and addition to solar protection elements. The simulation-based optimization process was initiated after the simulation of the Initial Case, but was finished when the 54th Generation was completed, totaling 7293 individuals. The improvement in lighting and energy performance of the building was already perceived in the simulation phase of the Initial Case, as a better distribution of daylight in the interior rooms was achieved with a reduction of 14% in the energy consumption of the building. In the Optimized Case, there was an improvement, compared to the Initial Case, of 20% in the average useful daylight illuminance value, as well as a reduction of 5% in the energy use intensity.