Thresholding methods in non-intrusive load monitoring

Non-intrusive load monitoring (NILM) is the problem of predicting the status or consumption of individual domestic appliances only from the knowledge of the aggregated power load. NILM is often formulated as a classification (ON/OFF) problem for each device. However, the training datasets gathered b...

Descripción completa

Detalles Bibliográficos
Autores: Gómez-Ullate, David, Precioso, Daniel
Tipo de recurso: artículo
Fecha de publicación:2023
País:España
Institución:IE
Repositorio:Repositorio IE
OAI Identifier:oai:repositorio.ie.edu:20.500.14417/3565
Acceso en línea:https://doi.org/10.1007/s11227-023-05149-8
https://hdl.handle.net/20.500.14417/3565
Access Level:acceso abierto
Palabra clave:Non-intrusive load monitoring (NILM)
Recurrent neural networks
Convolutional neural networks
Binary cross-entropy loss
Mean squared error loss
33 Ciencias Tecnológicas
ODS 7 - Energía asequible y no contaminante
ODS 11 - Ciudades y comunidades sostenibles
id ES_12e4ae28337ac53e68abca0129ca3dba
oai_identifier_str oai:repositorio.ie.edu:20.500.14417/3565
network_acronym_str ES
network_name_str España
repository_id_str
spelling Thresholding methods in non-intrusive load monitoringGómez-Ullate, DavidPrecioso, DanielNon-intrusive load monitoring (NILM)Recurrent neural networksConvolutional neural networksBinary cross-entropy lossMean squared error loss33 Ciencias TecnológicasODS 7 - Energía asequible y no contaminanteODS 11 - Ciudades y comunidades sosteniblesNon-intrusive load monitoring (NILM) is the problem of predicting the status or consumption of individual domestic appliances only from the knowledge of the aggregated power load. NILM is often formulated as a classification (ON/OFF) problem for each device. However, the training datasets gathered by smart meters do not contain these labels, but only the electric consumption at every time interval. This paper addresses a fundamental methodological problem in how a NILM problem is posed, namely how the different possible thresholding methods lead to different classification problems. Standard datasets and NILM deep learning models are used to illustrate how the choice of thresholding method affects the output results. Some criteria that should be considered for the choice of such methods are also proposed. Finally, we propose a slight modification to current deep learning models for multi-tasking, i.e. tackling the classification and regression problems simultaneously. Transfer learning between both problems might improve performance on each of them.yesPublishedSpringer Nature Linkhttps://ror.org/02jjdwm75202520252023info:eu-repo/semantics/articleapplication/pdfapplication/pdfhttps://doi.org/10.1007/s11227-023-05149-8https://hdl.handle.net/20.500.14417/3565reponame:Repositorio IEinstname:IEInglésIE School of Science & TechnologyIE UniversityApplied MathematicsAttribution 4.0 Internationalhttps://creativecommons.org/licenses/by/4.0/deedinfo:eu-repo/semantics/openAccessoai:repositorio.ie.edu:20.500.14417/35652026-06-15T12:40:57Z
dc.title.none.fl_str_mv Thresholding methods in non-intrusive load monitoring
title Thresholding methods in non-intrusive load monitoring
spellingShingle Thresholding methods in non-intrusive load monitoring
Gómez-Ullate, David
Non-intrusive load monitoring (NILM)
Recurrent neural networks
Convolutional neural networks
Binary cross-entropy loss
Mean squared error loss
33 Ciencias Tecnológicas
ODS 7 - Energía asequible y no contaminante
ODS 11 - Ciudades y comunidades sostenibles
title_short Thresholding methods in non-intrusive load monitoring
title_full Thresholding methods in non-intrusive load monitoring
title_fullStr Thresholding methods in non-intrusive load monitoring
title_full_unstemmed Thresholding methods in non-intrusive load monitoring
title_sort Thresholding methods in non-intrusive load monitoring
dc.creator.none.fl_str_mv Gómez-Ullate, David
Precioso, Daniel
author Gómez-Ullate, David
author_facet Gómez-Ullate, David
Precioso, Daniel
author_role author
author2 Precioso, Daniel
author2_role author
dc.contributor.none.fl_str_mv https://ror.org/02jjdwm75
dc.subject.none.fl_str_mv Non-intrusive load monitoring (NILM)
Recurrent neural networks
Convolutional neural networks
Binary cross-entropy loss
Mean squared error loss
33 Ciencias Tecnológicas
ODS 7 - Energía asequible y no contaminante
ODS 11 - Ciudades y comunidades sostenibles
topic Non-intrusive load monitoring (NILM)
Recurrent neural networks
Convolutional neural networks
Binary cross-entropy loss
Mean squared error loss
33 Ciencias Tecnológicas
ODS 7 - Energía asequible y no contaminante
ODS 11 - Ciudades y comunidades sostenibles
description Non-intrusive load monitoring (NILM) is the problem of predicting the status or consumption of individual domestic appliances only from the knowledge of the aggregated power load. NILM is often formulated as a classification (ON/OFF) problem for each device. However, the training datasets gathered by smart meters do not contain these labels, but only the electric consumption at every time interval. This paper addresses a fundamental methodological problem in how a NILM problem is posed, namely how the different possible thresholding methods lead to different classification problems. Standard datasets and NILM deep learning models are used to illustrate how the choice of thresholding method affects the output results. Some criteria that should be considered for the choice of such methods are also proposed. Finally, we propose a slight modification to current deep learning models for multi-tasking, i.e. tackling the classification and regression problems simultaneously. Transfer learning between both problems might improve performance on each of them.
publishDate 2023
dc.date.none.fl_str_mv 2023
2025
2025
dc.type.none.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://doi.org/10.1007/s11227-023-05149-8
https://hdl.handle.net/20.500.14417/3565
url https://doi.org/10.1007/s11227-023-05149-8
https://hdl.handle.net/20.500.14417/3565
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv IE School of Science & Technology
IE University
Applied Mathematics
dc.rights.none.fl_str_mv Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/deed
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/deed
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Springer Nature Link
publisher.none.fl_str_mv Springer Nature Link
dc.source.none.fl_str_mv reponame:Repositorio IE
instname:IE
instname_str IE
reponame_str Repositorio IE
collection Repositorio IE
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869403643213512704
score 15,81155

Similares en LA Referencia