A method to quantify uncertainties in airtightness measurements: Zero-flow and envelope pressure

Reliability of quantitative and qualitative assessment of air leakage paths through reductive sealing

“…The normal distribution of the zero-flow pressure points is confirmed by the results of the Kolmogorov-Smirnov (D = 0.17, p = 0.08) and Shapiro-Wilk tests (D = 0.91, p = 0.25), in line with previous findings in the literature [20]. Ratios between the considered regression methods were calculated and presented in Table 3.…”

“…When quantifying airtightness by fan pressurization tests there are three common types of errors: precision and bias errors on measurements and modelization errors on the used model [17,18].Wind speed and direction fluctuations are one of the main sources of precision and modelization errors [19]. The zero-flow measurements intend to tackle part of it, but still, for uncertainty purposes, the assumptions made show significant importance [20,21]. In addition to these errors there is the influence of the tester behaviour, which is the main reason behind certification programmes for testers and quality management schemes [22,23].…”

“…The accuracy and resolution uncertainties of the used equipment were considered as uniform distributions while the standard deviation on zero-flow pressure was approximated as a normal distribution, as pointed in recently developed work [20,31]. These sources of error and their joint uncertainty are used in the quantification of weights for the application of WLOC.…”

“…A simple linear model ε = (0.11 + 0.98σ) was found to be a proper predictor of this component [31], being σ the standard deviation of the measured zero-flow pressure points. As the points distribution tested significant for normality [20], considering an interval containing 50% of its values results in ε/1.35.…”

Improving air leakage prediction of buildings using the fan pressurization method with the Weighted Line of Organic Correlation

“…Sherman et al [20] analyzed in 1995 the errors of extrapolation to low pressures using the Ordinary Least Square (OLS) regression method. Recent studies also confirmed that the uncertainty of fan pressurization measurements could not be neglected [21]. Furthermore, Carrié et al [22] recently highlighted, in particular, the influence of wind fluctuations and frequency on the uncertainty of building airtightness pressurization tests.…”