Predicted and actual indoor environmental quality: Verification ofoccupants’ behaviour models in residential buildings

Occupants’ interactions with the building envelope and building systems can have a large impact onthe indoor environment and energy consumption in a building. As a consequence, any realistic forecastof building performance must include realistic models of the occupants’ interactions with the buildingcontrols (windows, thermostats, solar shading etc.).During the last decade, studies about stochastic models of occupants’ behaviour in relation to controlof the indoor environment have been published. Often the overall aim of these models is to enablemore reliable predictions of building performance using building energy performance simulations (BEPS).However, the validity of these models has only been sparsely tested.In this paper, stochastic models of occupants’ behaviour from literature were tested against mea-surements in five apartments. In a monitoring campaign, measurements of indoor temperature, relativehumidity and CO2concentration was measured in the living room and bedroom at five minute intervals infive apartments with similar layout in a building located in Copenhagen, Denmark. Outdoor temperature,relative humidity, wind speed and solar radiation were obtained from a weather station close by.The stochastic models of window opening and heating set-point adjustments were implemented inthe BEPS tool IDA ICE. Two apartments from the monitoring campaign were simulated using the imple-mented models and the measured weather data. The results were compared to measurements from themonitoring campaign to get an estimate of the forecast’s realism.The simulations resulted in realistic predictions in a sense that the measured values were within orclose to the range of the simulated values. The variation in the simulated and measured variables betweenapartments and over time was similar. However, comparisons of the average stochastic predictions withthe measured temperatures, relative humidity and CO2concentrations revealed that the models did notpredict the actual indoor environmental conditions well.