Uncertainties in predicting the impact of climate change on thermal performance of domestic buildings in the UK

Abstract

Buildings typically have a long life span, which can easily reach 50 or 100 years. In the light of expected climate change it is therefore important to look towards the future, and to analyse how buildings will cope with the changes in climate that are predicted by the climatologists. In such long-term predictions of building performance, uncertainties play a large role. This paper describes a preliminary study that aims to better map out the consequences of dealing with a whole range of uncertainties in the specific case of predicting the effect of climate change on the energy use and thermal comfort (overheating) in the large stock of terraced houses in the UK. Uncertainties in climate change prediction are compared with other variable factors like building occupancy patterns, actual thickness of construction materials and HVAC control settings. Uncertainties have been propagated in a transient model of these terraced houses using the transient simulation program EnergyPlus. In order to explore the large range of input variants use has been made of a genetic algorithm. The current search and solution spaces are discussed, putting the impact of climate change in perspective with regards to other changes and developments that might increase or decrease overheating. Overall, the findings indicate that uncertainties in long term thermal performance predictions run high, with standard deviations of over 100%. However, the robustness of the contemporary dwelling design is fortunately high.