Suporte à decisão projetual quanto ao desempenho energético da envoltória em projetos colaborativos de edificações em fases iniciais

Abstract

The quantitative methods to support decision-making in energy-efficient building design have been evolving due to the energy crises since the 1970s and the current need to reduce CO2 emissions to mitigate climate change. Computer simulations are the primary tool for quantification, and there are several obstacles to their integration into the design process: most of the required building characteristics are defined only at the end of the process; simulation times exceed the designers' tolerance; design space simulation requires high computational cost; there is no discrimination of characteristics by their impact on results; there is no control over result uncertainties, and other obstacles. The objective of this master thesis is to propose an approach to support design decisions regarding the impact on the energy performance of commercial building envelopes, compatible with the early stages of the design process, which range from strategic pre-design definitions to the conceptual design. The scope is limited to hot and humid climates, and the performance is restricted to cooling loads. The proposal's conception is based on the literature review of the main obstacles and available solutions. The development begins by creating an information matrix corresponding to design spaces using Latin hypercube sampling, performance simulated by the INI-C metamodel, and sensitivity analysis using the Sobol method. Graphic interfaces with filters and graphical visualizations are developed for users to explore the design space. The approach's applicability is demonstrated by mapping its use in theoretical scenarios of collaborative design processes since the initial phases. The results confirmed the possibilities of use and improvement of energy efficiency