Paper Conference

Proceedings of Building Simulation 2021: 17th Conference of IBPSA


A bootstrap method to investigate the variability of overheating risk against the future climate uncertainty in dwellings

Cheng Cui, Rokia Raslan, Ivan Korolija
Institute for Environmental Design and Engineering, University College London, London, UK

Abstract: Future overheating risk in dwellings can be potentially mitigated by minimising the variability of overheating hours against uncertainties in future climate via robust optimisation. However, the estimation of this variability value through the utilisation of percentile-based probabilistic weather data has yet to be sufficiently investigated. In this simulation-based study, the bootstrap method is used to quantify the accuracy of the variability estimation via percentilebased weather data. The results indicate significant overheating risk in regulation-compliant houses. An increased degree of difficulty is also suggested in obtaining accurate estimations when considering time periods further in the future and when assuming higher carbon emissions. In addition, the skew normal distribution can be used for a simpler and faster estimation, but the underlying uncertainties must be strengthened throughout its implementation.
Keywords: uncertainty, climate change, retrofit, resilience, overheating
Pages: 933 - 940