Back to Right Light Proceedings startpage
Guy NewshamLightswitch: A Stochastic Model for Predicting Office Lighting Energy Consumption M-24, National Research Council Canada, Montreal Rd, Ottawa, Ontario Canada K1A 0R6 Ardeshir Mahdavi CBPD, Dept of Architecture, Carnegie Mellon University Pittsburgh, PA 15213-3890, USA Ian Beausoleil-Morrison CANMET, Natural Resources Canada 580 Booth St., Ottawa, Ontario Canada K1A 0E4 Abstract Lighting controls are promoted on the basis that they significantly reduce lighting energy consumption. We are currently studying the impact of lighting controls on office building energy consumption using the DOE2.1E building energy model. In DOE2.1E the lighting load is defined by an hourly profile. Modelling lighting controls (other than daylighting, which can be modelled dynamically) is achieved by inputting different profiles. However, representative profiles, particularly those showing the impact of lighting controls, are not readily available. Consequently, we decided to develop a model (LIGHTSWITCH) to predict lighting profiles for a typical office. For example, the impact of occupancy sensors clearly depends on the building occupancy: arrival times, departure times, and times of temporary absence. There is randomnesss to these occupancy parameters, therefore LIGHTSWITCH is a stochastic model which incorporates randomness based on observed occupancy behaviour. LIGHTSWITCH can model the impact of occupancy sensors of variable switching delays, daylighting controls, and zoning.
|