MinNet: toward more intelligent smart home energy management systems with fewer sensors

Abstract

The smart home technology has attracted considerable attention for decades. Nowadays, as the drastic increase of home energy consumption, efficient home energy management system HEMS has become a major building block. Aiming at building an easy-to-deploy intelligent HEMS, we design MinNet to i monitor the states of individual home appliances using the minimum number of power sensors and ii provide occupancy states of a house/room with inference intelligence. In detail, we use convenient plug-in power sensors to construct a wireless sensor network for data collection and establish an analytical framework based on the power model of appliances. By solving a carefully-designed optimisation problem, we find the minimum number of sensors to decode the states of appliances from their aggregated power and infer the occupancy states of the house/room. We implement MinNet in a house and validate its accuracy in appliance states monitoring and occupancy inference.