Non-Intrusive Load Monitoring for Domestic Installations with Renewable Energy Sources
by Awadelrahman M. A. Ahmed, Sudan (EUREC 2015-16)
Marie-Curie research fellow at CIRCE, Spain
PhD student at University of Zaragoza, Spain
*The project is a part of MEAN4SG-ITN – Metrology Excellence Academic Network for Smart Grids- Innovative Training Network, www.mean4sg-itn.eu
Monitoring the electricity consumption in the residential buildings has become an important task that can facilitate enhancing the energy efficiency and implementing the demand side management programs as key topics of smart grids. In this context, Non-Intrusive Load Monitoring (NILM) is an approach based on the analysis of the single aggregate electricity consumption signal of the building in order to give detailed breakdown and disaggregated signals of the energy consumption. The deployment of the smart meters provides a platform to collect aggregate household electricity consumption data which can make it appropriate for this purpose.
NILM existing approaches require a manual training phase to build models of the household appliances, which are subsequently used to disaggregate the household's electricity data which makes it difficult to generalize and upscale to the national scales of smart meter data currently being collected, also the current approaches do not incorporate the local energy generation (e.g. rooftop solar systems).
This project is aiming to develop unsupervised training methods which does not require a manual training phase and using just the aggregate smart meter data to perform the disaggregation. The methods will be implemented, validated and evaluated in a real platform.
Keywords: Energy Efficiency, Load Disaggregation, Machine Learning, NILM, Smart Grid, Smart Meter