Energy Disaggregation Using Fractional Calculus

Abstract

Non-Intrusive Load Monitoring aims to extract the energy consumption of individual electrical appliances through disaggregation of the total power load measured by one smart-meter. In this article we introduce the use of fractional calculus in the Non-Intrusive Load Monitoring task. Specifically the aggregated active power signal is transformed to its fractional derivatives incorporating temporal information properties of the input signal to the Non-Intrusive Load Monitoring architecture. The performance of the proposed methodology was evaluated in two publicly available datasets namely REDD and AMPds2 using Convolutional Neural Networks and Recurrent Neural Networks as regression models. The proposed approach improves the estimation accuracy by 3.4% when compared to the baseline energy disaggregation setup achieving a maximum disaggregation accuracy of 90.8%.