Abstract

Climate change continues to impact weather conditions globally, requiring cities to adapt to new environments. In Canada, indoor summer overheating in buildings is problematic due to their primary design for cold winters, making them susceptible to extreme heatwave events. This research focuses on calibrating building models and employing model calibration methodologies to evaluate and address overheating risks in various building types across Canada using current and future weather data. The study's objectives are to accurately assess summertime overheating risks in selected buildings in Montreal, Quebec, and evaluate effective mitigation strategies. Bayesian calibration and multi-objective genetic algorithms are used as calibration methods, with the latter showing superiority in producing highly accurate calibrated models based on five performance criteria. By incorporating field measurements and novel methodologies, the research ensures precise assessment and mitigation of summertime overheating risks. After calibrating several buildings, including schools, a hospital, and a residential building, which demonstrates the reliability and repeatability of the calibration process, the assessment of overheating is conducted on calibrated models to determine the number of overheating hours during the summer. In conclusion, this thesis demonstrates the repeatability of the calibration methods on a variety of existing Canadian buildings and the effective use of passive cooling techniques, such as external shading, night cooling, and high albedo surfaces, that are implemented in the building models, to mitigate overheating based on current and future weather data.