Simulações termoenergéticas em brises: Análise de otimização multiobjetivo e algoritmos genéticos aplicado em projeto baseado em desempenho

Abstract

The construction industry is one of the sectors that most significantly impacts the environment throughout the building's life cycle, one of the factors that contributes to this is energy consumption. There are studies about the importance of strategic decisions in the early stages of building design to reduce energy consumption. So, this study proposes a design routine verification to identify the form of shading devices (brises) that allows for more natural lighting while minimizing energy consumption. These are conflicting solutions, because that multi-objective optimization was employed using genetic algorithms and the NSGA II approach. The objective function of the problem was to minimize the area of the shading device and the cooling load while maximizing natural lighting. For this purpose, the following indicators were selected: Area (m²), CgTT (kWh/ano), and UDI (%) – Useful Daylight Illuminance (%). The case study was a floor model subdivided into five zones—four peripheral and one central, with an office typology in Campo Grande (MS), and the analysis was conducted in one zone. The design parameters were guided by NBR 15220-3, NBR 15575, and INI-C, and the variables were the cartesian coordinates of the brises. The analysis was conducted using Grasshopper® along with Rhinoceros® and the Ladybug Tools® plugins (which are based on EnergyPlus®, OpenStudio®, and Radiance®) and Wallacei® (which uses the NSGA II approach). In addition to optimization, a sensitivity analysis was conducted using the Colibri® plugin and software R® to identify which elements have the most significant impact on cooling load and natural lighting. A population of 1000 individuals was generated, and solutions from the Pareto Front were selected. The optimized models showed advantages such as allowing more natural lighting throughout the whole environment and reducing the shading device area (which would require less material), but with a slight increase in cooling energy consumption. Thus, multi-objective optimization can assist professionals in the early stages of design.