Next Article in Journal
Quality and Reliability-Exploitation Modeling of Power Supply Systems
Previous Article in Journal
A Novel Surface Parameterization Method for Optimizing Radial Impeller Design in Fuel Cell System
Previous Article in Special Issue
A Powerful Tool for Optimal Control of Energy Systems in Sustainable Buildings: Distortion Power Bivector
Article

Comparative Evaluation of Non-Intrusive Load Monitoring Methods Using Relevant Features and Transfer Learning

1
Laboratoire IBISC (Informatique, BioInformatique, Systèmes Complexes), EA 4526, University Evry/Paris-Saclay, 91020 Evry-Courcouronnes, France
2
Institut de Recherche en Energie Electrique de Nantes Atlantique (IREENA), EA 4642, University of Nantes, 44602 Saint-Nazaire, France
3
Laboratoire des Systèmes Electriques, Université de Tunis El Manar, Tunis 1002, Tunisia
*
Author to whom correspondence should be addressed.
Academic Editors: Junemo Koo and Jinkyun Cho
Energies 2021, 14(9), 2726; https://doi.org/10.3390/en14092726
Received: 16 March 2021 / Revised: 16 April 2021 / Accepted: 23 April 2021 / Published: 10 May 2021
(This article belongs to the Special Issue Data-Driven Energy-Cost Analysis of HVAC System for Buildings)
Non-Intrusive Load Monitoring (NILM) refers to the analysis of the aggregated current and voltage measurements of Home Electrical Appliances (HEAs) recorded by the house electrical panel. Such methods aim to identify each HEA for a better control of the energy consumption and for future smart grid applications. Here, we are interested in an event-based NILM pipeline, and particularly in the HEAs’ recognition step. This paper focuses on the selection of relevant and understandable features for efficiently discriminating distinct HEAs. Our contributions are manifold. First, we introduce a new publicly available annotated dataset of individual HEAs described by a large set of electrical features computed from current and voltage measurements in steady-state conditions. Second, we investigate through a comparative evaluation a large number of new methods resulting from the combination of different feature selection techniques with several classification algorithms. To this end, we also investigate an original feature selection method based on a deep neural network architecture. Then, through a machine learning framework, we study the benefits of these methods for improving Home Electrical Appliance (HEA) identification in a supervised classification scenario. Finally, we introduce new transfer learning results, which confirm the relevance and the robustness of the selected features learned from our proposed dataset when they are transferred to a larger dataset. As a result, the best investigated methods outperform the previous state-of-the-art results and reach a maximum recognition accuracy above 99% on the PLAID evaluation dataset. View Full-Text
Keywords: Non-Intrusive Load Monitoring (NILM); Home Electrical Appliances (HEAs); identification; feature selection; transfer learning; deep learning Non-Intrusive Load Monitoring (NILM); Home Electrical Appliances (HEAs); identification; feature selection; transfer learning; deep learning
Show Figures

Figure 1

MDPI and ACS Style

Houidi, S.; Fourer, D.; Auger, F.; Sethom, H.B.A.; Miègeville, L. Comparative Evaluation of Non-Intrusive Load Monitoring Methods Using Relevant Features and Transfer Learning. Energies 2021, 14, 2726. https://doi.org/10.3390/en14092726

AMA Style

Houidi S, Fourer D, Auger F, Sethom HBA, Miègeville L. Comparative Evaluation of Non-Intrusive Load Monitoring Methods Using Relevant Features and Transfer Learning. Energies. 2021; 14(9):2726. https://doi.org/10.3390/en14092726

Chicago/Turabian Style

Houidi, Sarra, Dominique Fourer, François Auger, Houda B.A. Sethom, and Laurence Miègeville. 2021. "Comparative Evaluation of Non-Intrusive Load Monitoring Methods Using Relevant Features and Transfer Learning" Energies 14, no. 9: 2726. https://doi.org/10.3390/en14092726

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop