Eficiência energética de edifícios residenciais por análise de correlação canônica

Abstract

In both global and national contexts, buildings are important sources of CO2 emissions and energy consumption. The building sector, with a focus on residential buildings, is one of the main consumers of electricity in Brazil. Faced with this challenge, strategies such as energy efficiency (EE) stand out as solutions. In the context of buildings, EE refers to practices aimed at reducing energy consumption, starting from the construction phase and including aspects of architectural design and thermal comfort. This study aimed to investigate the relationship between physical characteristics related to building form (relative compactness, surface area, wall area, roof area, total height, orientation, glazed area, distribution of glazed area) of residential buildings and their energy efficiency, characterized by thermal loads (heating and cooling loads), through canonical correlation analysis (CCA). CCA is a multivariate statistical technique that aims to quantify the relationship between two sets of variables by creating linear combinations that maximize the correlation between them. With it, it is possible to verify which variables have the greatest influence on the relationship between the groups, among other information. The analysis of the first canonical function revealed a canonical correlation of 0.95, indicating a strong linear relationship between building traits and thermal loads. Moreover, it explained 89.05% of the variance of the thermal loads and 33.12% of the variance of the building characteristics. Among the parameters analyzed, roof area and total height stood out as the most influential on thermal loads. A greater total height was associated with higher thermal loads, while larger roof area showed an inverse relationship to thermal loads. These findings suggest that lower heights and more extensive roofs are associated with better energy performance in residential buildings.