Occupancy-aided energy disaggregation

Abstract

Energy disaggregation helps to identify major energy guzzlers in the house without introducing extra metering cost. It motivates users to take proper actions for energy saving and facilitates demand response programs. To reduce the computational complexity of pure energy disaggregation, we propose an occupancy-aided energy disaggregation (OAED) approach in this paper. Specifically, we make use of the occupancy information (whether or not the house/room is occupied by users) and classify the whole time interval into occupied and unoccupied periods. In unoccupied periods, we perform lightweight energy approximation; in occupied periods, we apply energy disaggregation with existing methods. Real-world experiments are conducted in an apartment hosting typical household appliances. Comparing with energy disaggregation without considering occupancy information, our occupancy-aided approach can significantly reduce the computational overhead while ensuring the accuracy of energy disaggregation.