Impact of people flows on energy consumption in an office building in Tokyo using a state–space model

Understanding the dynamics of energy consumption in buildings, which are closely related to occupant behavior, is important for mitigating climate change in urban areas. Obtaining information on the flow of people using GPS data has become possible in recent years. In this study, we used the state–space model to quantitatively clarify the relationship between (a) heat load, the electricity consumption used for the (b) heating, ventilation, air conditioning (HVAC) fans, (c) outlets, and (d) lighting as a dependent variables, and occupants in the building as an explanatory variable. In addition, we added outdoor temperature (relevant only to (a) and (b) in relation to HVAC) and a weekday daytime dummy as explanatory variables, performed model selection, and estimated the variables after controlling for their influence. Consequently, in summer, a 1 % increase in the number of people within a building resulted in energy increases of 0.114 $\pm$ 0.051 % (mean $\pm$ S.E.) for the (a) heat load, 0.218 $\pm$ 0.032 % for the (b) HVAC fan, 0.024 $\pm$ 0.005 % for (c) outlets, and 0.038 $\pm$ 0.010 % for (d) lighting. In winter, a 1 % increase in the number of people resulted in energy increases of 0.009 $\pm$ 0.033 % for the (a) heat load, 0.212 $\pm$ 0.034 % for the (b) HVAC fan, 0.031 $\pm$ 0.006 % for (c) outlets, and 0.039 $\pm$ 0.010 % for (d) lighting. This study clarified the association between energy consumption and the influence of people in a building. These results are useful for precisely reflecting the influence of occupant behavior in urban building energy modeling.