Multi-label LSTM autoencoder for non-intrusive appliance load monitoring

•Proposed a new approach to NILM.•LSTM autoencoder fuses dynamic and static modeling paradigms.•Improves over the state-of-the-art algorithms. This work follows the multi-label classification based paradigm for non-intrusive load monitoring (NILM). Power consumption signals used for NILM are inheren...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Electric power systems research Jg. 199; S. 107414
Hauptverfasser: Verma, Sagar, Singh, Shikha, Majumdar, Angshul
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Amsterdam Elsevier B.V 01.10.2021
Elsevier Science Ltd
Elsevier
Schlagworte:
Algorithms
Artificial Intelligence
Autoencoder
Computer Science
Datasets
Deep learning
Electric power
Energy consumption
Energy disaggregation
Engineering Sciences
LSTM
Monitoring systems
Non-intrusive load monitoring
Power consumption
Signal classification
Signal monitoring
Deep learning
Non-intrusive load monitoring
LSTM
Autoencoder
Energy disaggregation
ISSN:0378-7796, 1873-2046
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Proposed a new approach to NILM.•LSTM autoencoder fuses dynamic and static modeling paradigms.•Improves over the state-of-the-art algorithms. This work follows the multi-label classification based paradigm for non-intrusive load monitoring (NILM). Power consumption signals used for NILM are inherently time varying. However prior multi-label classification techniques could not model this dynamical behaviour. They used off-the-shelf algorithms for classifying static signals on NILM problems. This is the first work that shows how to account for the temporal variability of input signals in a multi-label classification framework. Results on benchmark datasets like REDD and Pecan Street show considerable improvement over several state-of-the-art shallow and deep techniques.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0378-7796
1873-2046
DOI:10.1016/j.epsr.2021.107414