Towards improved thermal comfort predictions for building controls

Abstract

This paper updates the findings of a prior work that found evidence to suggest that predictions of thermal comfort can be improved by adding measurements of indoor CO2 concentrations. This work first updates these findings by adding 150 new samples of IEQ measurements collected from occupants of office spaces at the University of British Columbia in 2019. This paper then formulates and proposes a novel Hierarchical Bayesian model, trained on the expanded field dataset, that predicts thermal satisfaction based on thermal IEQ metrics and measurements of CO2 levels. Posterior predictive results revealed a robust and statistically significant correlation between perceived thermal comfort and indoor CO2 levels. Cross-validation and posterior checks revealed stronger evidence that including indoor CO2 concentrations as an independent variable when predicting thermal satisfaction improves its prediction accuracy. The proposed model can be integrated into building control systems to predict thermal comfort in office spaces based on thermal conditions and ventilation rates, which improves the prediction accuracy of thermal comfort, mitigate the performance gap between predictions and observations of thermal comfort, and may result in energy savings while not sacrificing indoor air quality and well-being, an important challenge to building controls.