Non-intrusive monitoring of overlapping home appliances using smart meter measurements

Energy savings and energy management are some of the important considerations in the deployment of smart grids. Non-intrusive load monitoring (NILM) is a set of methodologies and techniques that analyze the total aggregated smart meter measurements and break them down into the individual consumptions by devices present in a household. Through NILM, the household owners will be informed about detailed consumptions by individual devices and this information was found to be influencing their energy consumption behavior, resulting in energy savings. In this work, we focus on a specific challenge in the NILM realm, which is disaggregating appliances with close or overlapping power consumptions. An approach for devices clusters splitting is proposed based on testing clusters cohesion. The proposed approach was tested on a real house from the REDD public data set. The performance of the proposed approach was found to be dependent on clusters tightness, size and degree of overlapping.