ABSTRACT

This paper discusses novel technologies for energy efficiency and predictive maintenance using hardware accelerated energy disaggregation. The disaggregation process involves the use of custom designed smart sensors that collect and treat aggregated information on the current and voltage waveforms. The treated data are further on transmitted to the cloud where they are stored and processed to enable the extraction of advanced information on individual device consumption patterns and health status. This information can be extremely useful for the management of electric devices in residential or commercial sites as well as for predictive maintenance in industrial sites. The paper reviews the underlying methodologies, and presents preliminary work and results from data collection in the offices of a software company. The presented work involves the installation of measurement devices and the development of complementary hardware and software. This is part of the ongoing 4-year project PREDIVIS (PREdictive, Disaggregation Intelligent VIS (meaning “power” in Latin)).