Virtual metering of electrical appliances through analysis of data from contactless sensing

Metering of electricity consumption, both at the building-level and appliance-level, provides stakeholders like residents, facility managers, building owners, etc. with information requisite to engage in energy efficient practices. Currently available solutions for appliance-level energy metering require the installation of plug-through power meters; this is often difficult and costly for appliances with inaccessible wires/outlets, or for appliances that draw large amounts of current. In this paper, we utilize a framework that performs the step of energy estimation for load disaggregation, in order to substitute the need of plug-level sensors with cheap and easily deployable contactless sensors (e.g., light, sound, magnetic field sensors). The framework combines data from these contactless sensors and aggregate metering, in order to virtually meter the electricity consumption of specific appliances. The solution requires minimal calibration, and is easily performed using commercially available sensors. We test it in a commercial building with 6 appliances that are monitored using magnetic field sensors and find that the inferred energy values have an average error of 10.9%.