Thermal comfort principles are essential to ensure good quality housing, but in most social housing projects in Brazil, such criteria have not been considered satisfactorily. Detailed computer simulation, that may preditc the performance of constructions in the design stage, are mainly used in the design's final stages to validate decisions already made. Since the high complexity of such tools present difficulties on its application during the early design stages, when they could provide quantitative information to guide de designers' decision making. Simplified simulations, although easy, generate imprecise results, due to its simplification of the complex mechanisms that rules the iterations involved. This research is part of a larger project (to be submitted by the adviser during the year 2013) which aims to develop a tool that is in accordance to the reality of Brazilian bioclimatic zones, being able to provide accurate results, and may be used during the early design stages to evaluate the thermal comfort in social houses. Such tool, consisting of one regression model, will be obtained by regression methods, based on the iteration of a highly detailed simulation software. The Monte Carlo method will be applied to generate the random representative samples, according to the building's type considered, to generate de data that will be used on the regression analysis. This tool will be optimized, ensuring the compatibility of its responses to those obtained by the software EnergyPlus This research will focus on the envelope parameters ensuring that the proposed tool will consider important aspects, such as "U" (transmittance) "a" (absorbance) and "Ct" (thermal capacity), on an integrated manner and with appropriate attention.
News published in Agência FAPESP Newsletter about the scholarship: