Reduction of HVAC system runtime due to occupancy-controlled smart thermostats in contemporary multi-unit residential building suites

Abstract

Previous studies in single family homes have demonstrated a reduction in space conditioning energy demand through the use of occupancy-controlled smart thermostats. This technology has the potential to reduce space conditioning demand in multi-unit residential buildings (MURBs) as well, however no previous studies have tested the performance of smart thermostats in this building type. Field data were collected from 56 thermostats installed in two condominium buildings located in Toronto, Canada. Thermostats installed in each suite were operated using through three different control scenarios during the monitoring period: 1) a baseline scenario, where the thermostat is operated as a standard programmable thermostat, 2) an occupancy-based control scenario, and 3) a load-shifting control scenario. Baseline runtime data collected while the thermostats were operated as a standard programmable thermostat was used in combination with weather data and a supervised learning regression algorithm (Random Forest) to estimate the baseline runtime for each suite on days that the occupancy-based control strategy was running. When the estimated baseline runtime derived from the regression model was compared with the actual system runtime while the occupancy-based control strategy is running, an average reduction in suite HVAC system runtime of 17% was found.