Prediction of thermal transmittance for lightweight wood-frame houses using hierarchical regression analysis

Abstract

This study aims to establish a prediction equation for the thermal transmittance of the thermal bridges in a lightweight wood-frame house comprising composite wall materials using a statistical method. A heat transfer analysis was performed on 288 cases with different independent variables that affect the linear thermal transmittance and equivalent thermal transmittance of a wooden structure, and independent variables with a high degree of influence were selected through hierarchical regression analysis in order to establish a regression equation. The predicted values were compared with the heat transfer analysis values for verification. The results showed an average error of 5.9% and maximum error of 15.1% for the linear thermal transmittance and an average error of 0.9% and maximum error of 3.0% for the equivalent thermal transmittance, demonstrating an engineering level of reliability ( < 10%). The results of this study contribute to improving the rapidity and accuracy of assessing the energy performan...