A influência da porcentagem da área de janela no consumo de energia elétrica em residências na cidade de São Carlos - SP

• Ano de defesa: 2012
• Autor(a) principal: Bisinotto, Stela Leticia
• Orientador(a): Souza, Lea Cristina Lucas de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia Civil - PPGECiv
• País: BR
• Palavras-chave em Português: Construção civil; Conforto ambiental; Iluminação natural; Energia elétrica - consumo; WWR; Redes neurais artificiais
• Palavras-chave em Inglês: Electric energy; WWR; Daylighting; Artificial neural networks
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA CIVIL::CONSTRUCAO CIVIL
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/4667

This study seeks to verify the influence of windows areas in the electrical energy consumption of dwellings in the city of São Carlos. In order to do this, the index WWR (window to wall ratio) was applied, representing the ratio of window area and the area of external surface (wall). The applied methodology consisted on field campaigns, modeling and computational simulation. First, taking into account a preexistent data basis, the characteristics of the electrical energy consumption for the whole area of the city of São Carlos was verified. Then, specific buildings were selected to be measured in loco. The field data allowed the determination of building parameters (orientation, volume and constructed area, area of façades and window areas), which were directly or indirectly used as data entry for modeling the electrical energy consumption of the dwellings set. The modeling process applied Artificial Neural Networks (ANN), which is a computational technique inspired on the structure of human neurons. Several models were developed and their performance was evaluated by the determination coefficient (R2) and relative errors, shown by the relationship between simulated and real data. The best model was selected to be applied in the simulation of the influence of WWR and electrical energy consumption. The results showed that the WWR index has influence on the residential consumption of the dwellings studied and that NE orientation causes the lowest impact on electrical energy consumption. Furthermore, an ideal range of WWR/volume between 0,06 and 0,07 is related to the lowest energy consumption per constructed area in the city of São Carlos.