Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters

<p>Non-intrusive detection of household occupancy status using smart meter data presents challenges due to the intricate relationship between user behavior, energy usage, and occupancy. Effective occupancy classification relies on real-world factors beyond load consumption. In this study, we i...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Sakib Mahmud (15302404) (author)
مؤلفون آخرون: Faycal Bensaali (12427401) (author), Muhammad E․ H․ Chowdhury (22330219) (author), Mahdi Houchati (16891560) (author)
منشور في: 2025
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513538147483648
author Sakib Mahmud (15302404)
author2 Faycal Bensaali (12427401)
Muhammad E․ H․ Chowdhury (22330219)
Mahdi Houchati (16891560)
author2_role author
author
author
author_facet Sakib Mahmud (15302404)
Faycal Bensaali (12427401)
Muhammad E․ H․ Chowdhury (22330219)
Mahdi Houchati (16891560)
author_role author
dc.creator.none.fl_str_mv Sakib Mahmud (15302404)
Faycal Bensaali (12427401)
Muhammad E․ H․ Chowdhury (22330219)
Mahdi Houchati (16891560)
dc.date.none.fl_str_mv 2025-02-03T06:00:00Z
dc.identifier.none.fl_str_mv 10.1016/j.buildenv.2025.112635
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Multimodal_feature_fusion_and_ensemble_learning_for_non-intrusive_occupancy_monitoring_using_smart_meters/30233791
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Engineering
Electrical engineering
Information and computing sciences
Machine learning
Non-intrusive occupancy monitoring
Multimodal feature fusion
Sequence-to-sequence-to-point learning
Load disaggregation
Smart meters
Sustainable energy systems
Deep machine learning
dc.title.none.fl_str_mv Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p>Non-intrusive detection of household occupancy status using smart meter data presents challenges due to the intricate relationship between user behavior, energy usage, and occupancy. Effective occupancy classification relies on real-world factors beyond load consumption. In this study, we introduce the multimodal feature fusion for non-intrusive occupancy monitoring (MMF-NIOM) framework, which leverages both classical and deep machine learning algorithms to achieve state-of-the-art occupancy detection performance using smart meter data. Three modes of feature extraction are employed within MMF-NIOM. Using sequence-to-sequence-to-point (s2s2p) learning, we capitalize on cascaded deep learning systems to associate load-switching events with occupancy changes. Additionally, statistical features from aggregated smart meter data help correlate load consumption magnitude and patterns with occupancy labels. To gain deeper insights into user occupancy patterns over time, we integrate datetime information into the multimodal feature set. We also demonstrate that eliminating redundant devices during s2s2p learning further enhances occupancy detection. We combine features from all three modes of MMF-NIOM to achieve a state-of-the-art non-intrusive occupancy classification performance of 91.5 % accuracy and 91.5 % f1-score, approximately, by an ensemble of fine-tuned classifiers on the electricity consumption & occupancy (ECO) dataset. The proposed method is sustainable, robust, adaptable to various households, and can be mass-implemented within smart meters at a much lower cost and effort compared to the traditional internet of things (IoT)-based intrusive systems.</p><h2>Other Information</h2> <p> Published in: Building and Environment<br> License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.buildenv.2025.112635" target="_blank">https://dx.doi.org/10.1016/j.buildenv.2025.112635</a></p>
eu_rights_str_mv openAccess
id Manara2_34095ce758fcf9967afb9b4d20ddf944
identifier_str_mv 10.1016/j.buildenv.2025.112635
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/30233791
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart metersSakib Mahmud (15302404)Faycal Bensaali (12427401)Muhammad E․ H․ Chowdhury (22330219)Mahdi Houchati (16891560)EngineeringElectrical engineeringInformation and computing sciencesMachine learningNon-intrusive occupancy monitoringMultimodal feature fusionSequence-to-sequence-to-point learningLoad disaggregationSmart metersSustainable energy systemsDeep machine learning<p>Non-intrusive detection of household occupancy status using smart meter data presents challenges due to the intricate relationship between user behavior, energy usage, and occupancy. Effective occupancy classification relies on real-world factors beyond load consumption. In this study, we introduce the multimodal feature fusion for non-intrusive occupancy monitoring (MMF-NIOM) framework, which leverages both classical and deep machine learning algorithms to achieve state-of-the-art occupancy detection performance using smart meter data. Three modes of feature extraction are employed within MMF-NIOM. Using sequence-to-sequence-to-point (s2s2p) learning, we capitalize on cascaded deep learning systems to associate load-switching events with occupancy changes. Additionally, statistical features from aggregated smart meter data help correlate load consumption magnitude and patterns with occupancy labels. To gain deeper insights into user occupancy patterns over time, we integrate datetime information into the multimodal feature set. We also demonstrate that eliminating redundant devices during s2s2p learning further enhances occupancy detection. We combine features from all three modes of MMF-NIOM to achieve a state-of-the-art non-intrusive occupancy classification performance of 91.5 % accuracy and 91.5 % f1-score, approximately, by an ensemble of fine-tuned classifiers on the electricity consumption & occupancy (ECO) dataset. The proposed method is sustainable, robust, adaptable to various households, and can be mass-implemented within smart meters at a much lower cost and effort compared to the traditional internet of things (IoT)-based intrusive systems.</p><h2>Other Information</h2> <p> Published in: Building and Environment<br> License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.buildenv.2025.112635" target="_blank">https://dx.doi.org/10.1016/j.buildenv.2025.112635</a></p>2025-02-03T06:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.buildenv.2025.112635https://figshare.com/articles/journal_contribution/Multimodal_feature_fusion_and_ensemble_learning_for_non-intrusive_occupancy_monitoring_using_smart_meters/30233791CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/302337912025-02-03T06:00:00Z
spellingShingle Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
Sakib Mahmud (15302404)
Engineering
Electrical engineering
Information and computing sciences
Machine learning
Non-intrusive occupancy monitoring
Multimodal feature fusion
Sequence-to-sequence-to-point learning
Load disaggregation
Smart meters
Sustainable energy systems
Deep machine learning
status_str publishedVersion
title Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
title_full Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
title_fullStr Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
title_full_unstemmed Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
title_short Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
title_sort Multimodal feature fusion and ensemble learning for non-intrusive occupancy monitoring using smart meters
topic Engineering
Electrical engineering
Information and computing sciences
Machine learning
Non-intrusive occupancy monitoring
Multimodal feature fusion
Sequence-to-sequence-to-point learning
Load disaggregation
Smart meters
Sustainable energy systems
Deep machine learning