Clustering appliance energy consumption data for occupant energy-behavior modeling

ABSTRACTEnergy consumption of buildings varies significantly across buildings with similar functions and locations. Occupant behavior is one of the most significant sources of uncertainty related to energy consumption in buildings. A deeper understanding of occupant energy behavior can help in designing personalized behavior intervention strategies to save energy and predict energy consumption. This paper uses the Pecan Street dataset to cluster building occupants based on the energy they consume for each appliance in the household, and then developed load profiles for each of the clusters.