Abstract
The application of the dynamic integrated method, implemented in the form of a linear regression based on averages over a certain period for the evaluation of the thermal performance of the building envelope, is presented and discussed. Previous researches and results on this method are corroborated and extended. A data set obtained from two well insulated twin houses, located in Germany, measured during the winter period, was analysed. Both dwellings were unoccupied but the heating and ventilation systems were operated to simulate the behaviour of a family with two children. The analysis was focused on obtaining accurate heat transfer coefficient (HTC) and solar aperture (gA) estimates characterising the behaviour of the building envelope regarding heat loses and solar gains respectively. To do so, different models for the main heat exchange phenomena were evaluated, finding precise models for the solar gains and the ventilation system. Different options were considered for the dependent variable in the regression equation and the most suitable was determined. The optimum average period was also identified. Finally, the selected model was tested to find the minimum number of days required to obtain accurate results. The consistency of the results obtained from different data series is considered as key validation criteria. With a test duration of 18 days, HTC and gA estimates were within +/- 5% and +/- 8% from the mean, respectively, in both houses.
NSTL
Elsevier