Paper Conference

Proceedings of ASim Conference 2016: 3rd Asia conference of IBPSA-China, Japan, Korea


Mining and modeling occupant-behavior patterns in residential buildings based on field survey results

H. Zhou, B. Lin, Y. Jiang, H. Sun

Abstract: The variety of occupant behaviors in buildings have led to a significant gap between simulated building energy performance and measured one. It is crucial to explore and model occupant-behavior patterns based on field survey data to achieve more accurate simulation purpose. The team continuously collected environment and behavioral data in 3 residential rooms for over half a year, and then proposed a data mining approach, frequent-pattern mining, to mine the correlation rules between behaviors and related driving forces from collected data. Occupant-behavior patterns were finally extracted and modeled based on the generated correlation rules, and expected to be applied in housing energy simulation.
Keywords: Occupant-behavior, Frequent-pattern mining, Driving force, Correlation rule, Probit analysis