Electric Load Disaggregation in Smart Metering Using a Novel Feature Extraction Method and Supervised Classification

Abstract

Improving energy efficiency by monitoring household electrical consumption is of significant importance with the present-day climate change concerns. A solution for the electrical consumption management problem is the use of a nonintrusive appliance load monitoring system. This system captures the signals from the aggregate consumption, extracts the features from these signals and classifies the extracted features in order to identify the switched on appliances. This paper complements a novel feature extraction scheme presented in a previous work for load disaggregation with a comparative study of supervised classification methods. The objective of the current work is hence to make use of the feature extraction scheme to construct a database of signatures and then to compare different supervised learning methods for load classification. Preliminary results indicate high classification accuracy of all tested methods.