Urban temperature and electricity demand: probabilistic modeling and adaptive decision-making

Severe heatwaves cause tens of thousands of deaths worldwide, and extreme temperatures increase energy use for space-cooling in built environments. The regional heatwave can be amplified locally around built environments due to the thermal and radiative properties of the urban fabric. Therefore, built environments are exposed to excessive heat stressors, and better modeling is needed to identify the complicated interaction between heat stressors and built environments. 

This thesis aims at three research objectives. The first research develops probabilistic models to enable rapid and efficient forecasts of near-surface temperature with adequate accuracy, motivated by the necessity of highly resolution forecasts for mitigating urban heat risk. The second research develops an approach for quantifying the propagating uncertainty of urban temperature to the energy use of buildings. The study aims to properly consider the heat-induced uncertainty in urban electricity systems at a single-building scale. Finally, extending the previous research, the last research proposes to use an information theory to evaluate the value of accurate temperature modeling to save electricity costs for a building. The overall impact of the study is related to enhancing the resilience of the urban community by leveraging the knowledge of probability and urban informatics.