Publications

A method called Simulation-based Optimization (SBO) is a rising approach in the scientific communities of building technology. It consists in the use of optimization algorithms coupled to simulation software. SBO techniques have shown great potential to achieve energy efficient design goals, however, they are seldom used by engineers and architects, and are mostly studied in academic research. The present research consists on a study of an SBO method using computational tools with friendly interfaces from the point of view of architects. Aspects related to the geometric aspects of buildings were studied, regarding their complex relationship with the thermal-energy performance. The study was divided into three stages - the initial analysis using the Brazilian Technical Quality Requirements for the Energy Efficiency of Residential Buildings (RTQ-R) indicated the prominent geometry characteristics; the optimization of a thermal zone confirmed the potential for interoperability between software and plug-ins related to the process and, finally, the optimization of the housing unit highlighted geometric parameters of higher impact for the city of São Paulo. The cooling and heating degree-hours  (GHr and GHa) were used, with base temperatures of 26°C and 18°C, respectively, for the optimization of the housing unit. as the following parameters were identified as those with greater impact on the energy performance: area, roof inclination, glazed and ventilation area. A minimum of 200°Ch of cooling degree-hours and 800°Ch of heating degree-hours were achieved in the Pareto Front and a surrogate model was proposed to minimize the computational expenditure. The results indicate that methods of SBO applied to models with a small number of variables are already accessible for preliminary and / or general design studies from the point of view of the architects. More complex models can confirm the results through the proposed surrogate model developed in this study.

The thesis will be published soon in this site