An intuitive explanation of graph signal processing-based electrical load disaggregation

Graph Signal Processing (GSP)-based techniques for disaggregation are obtained as a solution of appropriately formulated optimization problem. There is a possibility to work out the intuitive steps behind this solution to gain insights on improvising the existing algorithm and building new ones. In this direction, the paper presents an attempt made to bring out the intuitive explanation of the GSP-based disaggregation solution in a clear manner. Additionally, the promising results obtained by extending the existing GSP-based solution for 15 minutes sampled power data are presented. Also captured are the results towards disaggregating the loads of interest when unknown high power loads are present in the aggregate power data.