A tool for generation of stochastic occupant-based internal loads using a functional data analysis approach to re-define 'activity'

摘要

In building energy simulation (BES), internal loads are typically defined as hourly schedules based on occupant-related 'activities' assigned to each building zone. In this paper, a data-centric bottom-up functional data analysis model is used to examine how activities in a building correlate with energy demand for plug loads and lighting. Functional principal component analysis and hierarchical clustering of the principal component scores have been used to explore the links between the data and zone activity. The results show that plug loads show limited links to activity to the extent that the activity determines the variability of the data. The lighting loads show little correlation with zone activity but instead are determined primarily by the building control system. A novel methodology is proposed for the generation of stochastic load data for input into BES. This methodology has been developed into a tool for stochastic load generation which is available online.