From Sensor Network To Social NetworkÐ A Study On The Energy Impact In Buildings

Graphical models built on sensor networks have been used extensively in smart home projects to improve occupancy comfort and building energy use. Simulation tools use profiles of occupants to predict future building energy use. Previous research focused on past or present occupancy and mobility. But the social interactions are often ignored. In this study, we model occupancy activities by constructing a social network from information provided by physical sensor networks in an open-plan office building. We propsed a Time Series Maximum Margin Markov Network model (TMN ) to incorporate information from evolving networks, e.g., number of occupants, occupant activities and indoor and outdoorCO2 changes. We then constructed an energy simulation model of the building from inference results. Simulation results show that energy savings reached 20% in the demonstration building while maintaining indoor occupancy thermal comfort.

[1]  Christoph F. Reinhart,et al.  Adding advanced behavioural models in whole building energy simulation: A study on the total energy impact of manual and automated lighting control , 2006 .

[2]  Edoardo M. Airoldi,et al.  Bayesian mixed-membership models of complex and evolving networks , 2006 .

[3]  Michael I. Jordan,et al.  Loopy Belief Propagation for Approximate Inference: An Empirical Study , 1999, UAI.

[4]  Parag A. Pathak,et al.  Massachusetts Institute of Technology , 1964, Nature.

[5]  Standard Ashrae.  Thermal Environmental Conditions for Human Occupancy , 1992 .

[6]  David E. Claridge.  Accounting for the Occupancy Variable in Inverse Building Energy Baselining Models , 2001 .

[7]  Ben Taskar,et al.  Learning structured prediction models: a large margin approach , 2005, ICML.

[8]  J. E. Janssen,et al.  Ventilation for acceptable indoor air quality , 1989 .

[9]  Diane J. Cook,et al.  Data Mining for Hierarchical Model Creation , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[10]  Rui Zhang,et al.  Occupancy detection through an extensive environmental sensor network in an open-plan office building , 2009 .

[11]  Darren Robinson,et al.  A generalised stochastic model for the simulation of occupant presence , 2008 .

[12]  Xuanjing Huang,et al.  Sparse higher order conditional random fields for improved sequence labeling , 2009, ICML '09.

[13]  L. Degelman.  A MODEL FOR SIMULATION OF DAYLIGHTING AND OCCUPANCY SENSORS AS AN ENERGY CONTROL STRATEGY FOR OFFICE BUILDINGS , .

[14]  Drury B. Crawley,et al.  EnergyPlus: A New-Generation Building Energy Simulation Program , 1999 .

[15]  Svetha Venkatesh,et al.  Human Behavior Recognition with Generic Exponential Family Duration Modeling in the Hidden Semi-Markov Model , 2006, 18th International Conference on Pattern Recognition (ICPR'06).