How can high-resolution temperature traces be used to assess building thermal performance?
David Veitch, UCL
David Veitch talks about this project:
In the context of UK dwelling thermal energy efficiency, performance monitoring and estimation is strongly focused around results from theoretical building physics models; however, these have been shown to have systematic deviations from reality. Conventional empirical validation methods are equipment intensive and require the dwelling to be vacated, effectively restricting their use to case studies. An alternate approach was proposed based on space temperature trace data, collected non-intrusively from occupied dwellings, with the aims of opening up the potential for large scale feedback. The focus of the project was to empirically investigate the feasibility of this approach.
Four types of temperature observation, related to dwelling fabric performance, were determined and subsequently explored through collection of temperature data from three case study dwellings. Two of the phenomena were found to have a quantitative relationship to main descriptors of fabric performance: the natural balance of the dwelling internal air temperature above the external ambient, which represents the ratio of incidental heat gains to the heat loss coefficient; and the rate of overnight internal temperature decay, or time constant, which represents the ratio of thermal mass to the heat loss coefficient.
As the measures represent a ratio, their individual application is limited to either longitudinal investigation of the percentage change in the variables e.g. before and after an efficiency retrofit, or direct feedback on their associated representations within the SAP calculation. This ratio form was subsequently established as fundamental limit of temperature-only based analysis. The approaches used to determine the measures showed significant variation in the results, both in part from their current basic method, but also due to the fundamental nature of temperature as a continuous variable in both space and time.
In conclusion, intrinsic challenges to the use of space temperature to produce successful and repeatable metrics were established around this theme. Translation of these ideas to conventional applications suggest that internal space temperature monitoring, as an entire field, has an unfounded basis, focussed around an arbitrary ‘average’ which is ill defined. It is proposed that greater clarity of thought is required regarding the different purposes measures of temperature are used for, and that a more nuanced approach is required for progression within the field.
I would like to thank my supervisor for his continued support and patience throughout the project. I would also like to thank Mike, Fraser, Magnus and Edwin at Passiv Systems Ltd. for their valuable time and input, without which the project would not have been possible.