Measuring the in-use thermal performance of dwellings: improving the precision by accounting for variability

Governments around the world are regulating the thermal performance of new and retrofitted dwellings to reduce the significant energy demand for space heating. These efforts are undermined by case-study evidence of a significant performance gap between the thermal performance predicted for compliance and the thermal performance that is achieved in-use. Overall building thermal performance is most simply described by the Heat Transfer Coefficient (HTC) metric. New low-cost methods are being developed to measure the in-use HTC of dwellings, that is, measured during occupancy. If deployed at scale, these methods could provide much-needed evidence of the performance gap and help to close it through feedback and quality assurance. This paper quantifies, for the first time, the variability in repeated measurements of the in-use HTC of occupied dwellings over a winter heating season and demonstrates that this variability can be explained by changes in the average boundary conditions (indoor air temperatures and weather). This finding could provide measurements of the in-use HTC with a higher precision than when the variability is simply assumed to be part of the uncertainty in the measurement. The in-use HTC of 19 occupied dwellings was repeatedly calculated over rolling 20-day periods for up to 6 months. The in-use HTC of the dwellings had a coefficient of variation (standard deviation divided by mean) of between 1.0% and 11.8% (mean 7.1%). Valid linear models that described the variability in the in-use HTC (dependent variable) by considering variations in the boundary conditions at the time of each repeated measurement (independent variables) were created for 17 of the 19 dwellings. The in-use HTC for each dwelling had a unique relationship with the boundary conditions and required a unique linear model. The linear model explained at least 80% of the variability in 13 of 17 dwellings.