Paper Conference

Proceedings of eSim 2020: 11th Conference of IBPSA-Canada

     

Applying data driven thermal modeling techniques to provide office occupants with time to setpoint estimates

Connor Brackley 1, William O'Brien 2, Chantal Trudel 3
1 Department of Mechanical and Aerospace Engineering, Carleton University, Ottawa, ON, Canada
2 Department of Civil and Environmental Engineering, Carleton University, Ottawa, ON, Canada
3 School of Industrial Design, Carleton University, Ottawa, ON, Canada


Abstract: While satisfaction with the ability to adjust personal thermostats offers significant benefits to peoples’ productivity, health, and satisfaction in the workplace, surveys have shown that current office workers have low levels of perceived control over their thermal environment. Meanwhile, using building sensors to develop data-driven thermal models for advanced building control has become increasingly popular. Currently, there are no known examples of presenting insights from these models to office building occupants to further engage them with their heating and cooling systems. In this study, three months of heating season operational data from 25 offices of an academic building was used to develop a grey box model of the offices. This model was used to provide occupants with estimates for time to reach the setpoint temperature. When tested on the same months of the following year, the timeto-setpoint estimations had a mean absolute error of 20, 42, and 65 minutes for 30, 60, 120 minute prediction horizons, respectively. This model was implemented into the building controller, with the predictions displayed to occupants through wall-mounted thermostats.
Paper:
esim2020_1186