Occupancy profiles modelling based on Indoor Measurements and Clustering Analysis: Application in an Office Building
Abstract
Sensors were applied in an office building to obtain information regarding user presence and absence intervals. Occupancy was also recorded by manual observation, and indoor parameters such as air temperature, relative humidity, carbon dioxide (CO2), volatile organic compounds (VOC) were monitored. Occupants’ behaviors regarding door/window (open/closed) and electric power were considered. Clustering analysis by manual observation was employed to identify similarities in daily or monthly occupancy and to describe possible occupancy profiles. Similar approach was carried out with each monitored parameter and the results of clustering elaboration were compared with the real occupancy profiles to identify which sensor is more effective to measure office occupancy. Furthermore, data were analyzed to explore relationships between occupancy and the magnitude of indoor environmental changes with the objective to identify daily, weekly, or monthly patterns. Single-linkage, complete-linkage, and average-linkage clustering were applied to each dataset. The cophenetic correlation coefficient was used to verify the quality of the results obtained for each variable, and the complete linkage was selected to define the groups. Comparison between occupancy real data clustering and VOC and open/closed door groups demonstrated not similarities. The electricity consumption and CO2 data showed some similarities.
Keywords: Occupancy detection, environmental sensor, clustering analysis, Office buildings
References
[1] Aldenderfer MS, Blashfield RK (1984) Cluster Analysis. In: Cluster Analysis (Quantitative applications in the social sciences. SAGE Publications, Inc., pp 11–28
[2] Andersen RV (2009) Occupant behaviour with regard to control of the indoor environment. Technical University of Denmark
[3] Dong B, Andrews B, Lam KP, Höynck M, Zhang R, Chiou YS, Benitez D (2010) An information technology-enabled sustainability test-bed (ITEST) for occupancy detection through an environmental sensing network. Energy and Buildings 42:1038–1046. doi: 10.1016/j.enbuild.2010.01.016
[4] Dong B, Lam KP (2011) Building energy and comfort management through occupant behaviour pattern detection based on a large-scale environmental sensor network. Journal of Building Performance Simulation 4:359–369. doi: 10.1080/19401493.2011.577810
[5] Ebadat A, Bottegal G, Varagnolo D, Wahlberg B, Johansson KH (2013) Estimation of building occupancy levels through environmental signals deconvolution. In: Proceedings of the 5th ACM Workshop on Embedded Systems For Energy-Efficient Buildings - BuildSys’13. pp 1–8
[6] Jain AK, Murty MN, P.J. Flynn (1999) Data clustering: a review. ACM Computing Surveys 31:264–323. doi: 10.1145/331499.331504
[7] Page J, Robinson D, Morel N, Scartezzini J-L (2008) A generalised stochastic model for the simulation of occupant presence. Energy and Buildings 40:83–98. doi: 10.1016/j.enbuild.2007.01.018
[8] R Development Core Team (2014) R: a language and environment for statistical computing. R Foundation for Statistical Computing.
[9] Wiesemann & Theis (2015) WuTility - management and inventorying tool.
[10] Zhang R, Lam KP, Chiou Y-S, Dong B (2012) Information-theoretic environment features selection for occupancy detection in open office spaces. Building Simulation 5:179–188. doi: 10.1007/s12273-012-0075-6