Heat flux pattern matching using numerical model to reduce required measurement duration for U-value

초록

Assessing the thermal performance of building envelopes is essential for improving energy efficiency and reducing carbon emissions. The heat flux method (HFM) is a widely used in situ measurement technique to evaluate thermal transmittance; however, its practical application is limited by prolonged measurement durations and strict environmental requirements. This paper proposes an optimization-based simulation model for reducing the required measurement period while maintaining accuracy. The model utilizes a subset of measurement data (24 h) to estimate the long-term heat flux and thermal transmittance by optimizing thermal conductivities of the material to match the observed data. Field measurements are conducted on residential buildings older than 30 years, and two verification scenarios are designed to evaluate the accuracy, numerical reproducibility, and potential of the model toward reducing measurement duration. The results confirmed that the proposed model accurately predicted the thermal transmittance with a root mean square error (RMSE) below 1 W/m2 and an average coefficient of variation of the RMSE of 3.46 %, which closely matched the long-term HFM measurements. Further, reducing the measurement period from three days to one day yielded comparable results, validating the feasibility of the approach. The proposed method enhances the efficiency of in situ thermal performance evaluations, increasing the viability of large-scale building energy assessments. This study contributes to accelerating energy performance diagnostics and supports building energy policies aimed at reducing carbon emissions.

키워드