Lawrence Berkeley National Laboratory

The fan pressurization method is a common practice in many countries for measuring the air leakage of houses. The test results are sensitive to uncertainties in the measured pressures and airflows. In particular, changing wind conditions during a test result in some pressure stations having more or less uncertainty than others. Usually, it is necessary to fit the measured data to the power-law equation. Using the ordinary least square (OLS) fitting method, the pressure exponent and flow coefficient can be determined, and the reported data at high pressures can be extrapolated to small pressures where natural infiltration occurs. However, this fitting method neglects the existing of the uncertainty of these measurements, which may lead to errors in the prediction of flows at low pressures and therefore to unreliable input data for energy simulations. The weighted line of organic correlation (WLOC) takes the uncertainty at each pressure station into account and minimizes the fitting residuals for both pressure and flow. This paper shows the results of a statistical analysis of an extensive data set of over 7.400 fan pressurization test of six houses in 109 different leakage configurations. It was found that in over 90 % of the analyzed cases, WLOC enables a more reliable prediction of pressure exponent and flow coefficient at low pressure compared to OLS and appears to be a better fitting technique