This version is available at https://strathprints.strath.ac.uk/56413/ Strathprints is designed to allow users to access the research output of the University of Strathclyde. Unless otherwise explicitly stated on the manuscript, Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Please check the manuscript for details of any other licences that may have been applied. You may not engage in further distribution of the material for any profitmaking activities or any commercial gain. You may freely distribute both the url (https://strathprints.strath.ac.uk/) and the content of this paper for research or private study, educational, or not-for-profit purposes without prior permission or charge.Any correspondence concerning this service should be sent to the Abstract-With ongoing massive smart energy metering deployments, disaggregation of household's total energy consumption down to individual appliances using purely software tools, aka. non-intrusive appliance load monitoring (NALM), has generated increased interest. However, despite the fact that NALM was proposed over 30 years ago, there are still many open challenges. Indeed, the majority of approaches require training and are sensitive to appliance changes requiring regular re-training. In this paper, we tackle this challenge by proposing a "blind" NALM approach that does not require any training. The main idea is to build upon an emerging field of graph-based signal processing to perform adaptive thresholding, signal clustering and feature matching. Using two datasets of active power measurements with 1min and 8sec resolution, we demonstrate the effectiveness of the proposed method using a state-of-the-art NALM approaches as benchmarks.