A Heating Agent Using a Personalised Thermal Comfort Model to Save Energy

We present a novel, personalised thermal comfort model and a heating agent using this model to reduce energy consumption with minimal comfort loss. At present, heating agents typically use simple models of user comfort when deciding on a set point temperature for the heating or cooling system. These models however generally fail to adapt to an individual user's preferences, resulting in poor performance. To address this issue, we propose a personalised thermal comfort model using a Bayesian network to learn and adapt to a user's individual preferences. Through an empirical evaluation based on the ASHRAE RP-884 data set, we show that our model is 17.5-23.5% more accurate than current models, regardless of environmental conditions and type of heating system. Further, our model has several additional outputs such as expected user feedback, optimal comfort temperature and thermal sensitivity that allow it to save between 18-20% of energy while still maintaining comfort.