Continuous Remaining Useful Life Prediction by Self-Guided Attention Convolutional Neural Network and Memory Consciousness Adjustment
To accurately predict the remaining useful life (RUL) of rotating machinery while continuously providing the task data, a novel continuous RUL prediction methodology was proposed. The methodology comprises a self-guided attention convolutional neural network (SGACNN) and memory consciousness adjustm...
Gespeichert in:
Veröffentlicht in: | IEEE internet of things journal 2024-10, Vol.11 (19), p.31947-31958 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 31958 |
---|---|
container_issue | 19 |
container_start_page | 31947 |
container_title | IEEE internet of things journal |
container_volume | 11 |
creator | Zhou, Jianghong Qi, Junyu Chen, Dingliang Qin, Yi |
description | To accurately predict the remaining useful life (RUL) of rotating machinery while continuously providing the task data, a novel continuous RUL prediction methodology was proposed. The methodology comprises a self-guided attention convolutional neural network (SGACNN) and memory consciousness adjustment (MCA) mechanism. First, a multihead focal channel-wise self-attention (MFCWSA) mechanism was implemented to effectively capture the degradation information across all the channels and achieve the attentional focus. Next, the SGACNN was constructed using the MFCWSA, squeeze-and-excitation mechanism, and convolutional block attention module. A new network gradient direction was synthesized by leveraging the gradients from both the previous task and the current task. Further, a weight constraint loss term based on the gradient magnitude was designed to constrain the learning process of important parameters. With the new network gradient direction and weight constraint loss, a novel MCA mechanism was proposed and integrated into the SGACNN for implementing the continuous RUL prediction tasks. Finally, various RUL prediction experiments on the life-cycle bearing and gear data sets were carried out, and its outcomes were compared to those of the advanced methods of the same kind. The comparative results validated the superiority of the proposed methodology. |
doi_str_mv | 10.1109/JIOT.2024.3421673 |
format | Article |
fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_crossref_primary_10_1109_JIOT_2024_3421673</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>10579798</ieee_id><sourcerecordid>3109489707</sourcerecordid><originalsourceid>FETCH-LOGICAL-c176t-69f70411f854865faec45a14aa1f52d38809df89b8712d3018e5de288a9e815c3</originalsourceid><addsrcrecordid>eNpNkMtOwzAQRSMEElXpByCxsMQ6xc7L9rKqeBQViqBdW24yRi5JXOwY1A_gv3HaLrqaGc29d0Yniq4JHhOC-d3zbLEcJzjJxmmWkIKmZ9EgSRMaZ0WRnJ_0l9HIuQ3GONhywotB9Dc1badbb7xD79BI3er2E60cKF-juVaA3ixUuuy0adF6hz6gVvGj1xVUaNJ10O4XIeTH1L7vZY1ewdt96X6N_UKyrdALNMbuep0rdbjVgnNoUm2865qQcRVdKFk7GB3rMFo93C-nT_F88TibTuZxSWjRxQVXFGeEKJZnrMiVhDLLJcmkJCpPqpQxzCvF-JpREkZMGOQVJIxJDozkZTqMbg-5W2u-PbhObIy34Wcn0gAyY5xiGlTkoCqtcc6CElurG2l3gmDRAxc9cNEDF0fgwXNz8GgAONHnlFPO0n8xbX5V</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3109489707</pqid></control><display><type>article</type><title>Continuous Remaining Useful Life Prediction by Self-Guided Attention Convolutional Neural Network and Memory Consciousness Adjustment</title><source>IEEE Xplore</source><creator>Zhou, Jianghong ; Qi, Junyu ; Chen, Dingliang ; Qin, Yi</creator><creatorcontrib>Zhou, Jianghong ; Qi, Junyu ; Chen, Dingliang ; Qin, Yi</creatorcontrib><description>To accurately predict the remaining useful life (RUL) of rotating machinery while continuously providing the task data, a novel continuous RUL prediction methodology was proposed. The methodology comprises a self-guided attention convolutional neural network (SGACNN) and memory consciousness adjustment (MCA) mechanism. First, a multihead focal channel-wise self-attention (MFCWSA) mechanism was implemented to effectively capture the degradation information across all the channels and achieve the attentional focus. Next, the SGACNN was constructed using the MFCWSA, squeeze-and-excitation mechanism, and convolutional block attention module. A new network gradient direction was synthesized by leveraging the gradients from both the previous task and the current task. Further, a weight constraint loss term based on the gradient magnitude was designed to constrain the learning process of important parameters. With the new network gradient direction and weight constraint loss, a novel MCA mechanism was proposed and integrated into the SGACNN for implementing the continuous RUL prediction tasks. Finally, various RUL prediction experiments on the life-cycle bearing and gear data sets were carried out, and its outcomes were compared to those of the advanced methods of the same kind. The comparative results validated the superiority of the proposed methodology.</description><identifier>ISSN: 2327-4662</identifier><identifier>EISSN: 2327-4662</identifier><identifier>DOI: 10.1109/JIOT.2024.3421673</identifier><identifier>CODEN: IITJAU</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Accuracy ; Artificial neural networks ; Attention ; Attention mechanism ; Computational modeling ; Consciousness ; Constraints ; continuous learning (CL) ; deep learning. remaining useful life (RUL) ; Degradation ; Feature extraction ; Life prediction ; Methodology ; Neural networks ; Predictive models ; Rotating machinery ; Task analysis ; Useful life ; Vectors</subject><ispartof>IEEE internet of things journal, 2024-10, Vol.11 (19), p.31947-31958</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024</rights><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c176t-69f70411f854865faec45a14aa1f52d38809df89b8712d3018e5de288a9e815c3</cites><orcidid>0000-0002-3180-7654 ; 0000-0003-1669-8987 ; 0000-0001-7338-2407 ; 0000-0002-2160-4300</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/10579798$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,780,784,796,27924,27925,54758</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/10579798$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Zhou, Jianghong</creatorcontrib><creatorcontrib>Qi, Junyu</creatorcontrib><creatorcontrib>Chen, Dingliang</creatorcontrib><creatorcontrib>Qin, Yi</creatorcontrib><title>Continuous Remaining Useful Life Prediction by Self-Guided Attention Convolutional Neural Network and Memory Consciousness Adjustment</title><title>IEEE internet of things journal</title><addtitle>JIoT</addtitle><description>To accurately predict the remaining useful life (RUL) of rotating machinery while continuously providing the task data, a novel continuous RUL prediction methodology was proposed. The methodology comprises a self-guided attention convolutional neural network (SGACNN) and memory consciousness adjustment (MCA) mechanism. First, a multihead focal channel-wise self-attention (MFCWSA) mechanism was implemented to effectively capture the degradation information across all the channels and achieve the attentional focus. Next, the SGACNN was constructed using the MFCWSA, squeeze-and-excitation mechanism, and convolutional block attention module. A new network gradient direction was synthesized by leveraging the gradients from both the previous task and the current task. Further, a weight constraint loss term based on the gradient magnitude was designed to constrain the learning process of important parameters. With the new network gradient direction and weight constraint loss, a novel MCA mechanism was proposed and integrated into the SGACNN for implementing the continuous RUL prediction tasks. Finally, various RUL prediction experiments on the life-cycle bearing and gear data sets were carried out, and its outcomes were compared to those of the advanced methods of the same kind. The comparative results validated the superiority of the proposed methodology.</description><subject>Accuracy</subject><subject>Artificial neural networks</subject><subject>Attention</subject><subject>Attention mechanism</subject><subject>Computational modeling</subject><subject>Consciousness</subject><subject>Constraints</subject><subject>continuous learning (CL)</subject><subject>deep learning. remaining useful life (RUL)</subject><subject>Degradation</subject><subject>Feature extraction</subject><subject>Life prediction</subject><subject>Methodology</subject><subject>Neural networks</subject><subject>Predictive models</subject><subject>Rotating machinery</subject><subject>Task analysis</subject><subject>Useful life</subject><subject>Vectors</subject><issn>2327-4662</issn><issn>2327-4662</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNpNkMtOwzAQRSMEElXpByCxsMQ6xc7L9rKqeBQViqBdW24yRi5JXOwY1A_gv3HaLrqaGc29d0Yniq4JHhOC-d3zbLEcJzjJxmmWkIKmZ9EgSRMaZ0WRnJ_0l9HIuQ3GONhywotB9Dc1badbb7xD79BI3er2E60cKF-juVaA3ixUuuy0adF6hz6gVvGj1xVUaNJ10O4XIeTH1L7vZY1ewdt96X6N_UKyrdALNMbuep0rdbjVgnNoUm2865qQcRVdKFk7GB3rMFo93C-nT_F88TibTuZxSWjRxQVXFGeEKJZnrMiVhDLLJcmkJCpPqpQxzCvF-JpREkZMGOQVJIxJDozkZTqMbg-5W2u-PbhObIy34Wcn0gAyY5xiGlTkoCqtcc6CElurG2l3gmDRAxc9cNEDF0fgwXNz8GgAONHnlFPO0n8xbX5V</recordid><startdate>20241001</startdate><enddate>20241001</enddate><creator>Zhou, Jianghong</creator><creator>Qi, Junyu</creator><creator>Chen, Dingliang</creator><creator>Qin, Yi</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><orcidid>https://orcid.org/0000-0002-3180-7654</orcidid><orcidid>https://orcid.org/0000-0003-1669-8987</orcidid><orcidid>https://orcid.org/0000-0001-7338-2407</orcidid><orcidid>https://orcid.org/0000-0002-2160-4300</orcidid></search><sort><creationdate>20241001</creationdate><title>Continuous Remaining Useful Life Prediction by Self-Guided Attention Convolutional Neural Network and Memory Consciousness Adjustment</title><author>Zhou, Jianghong ; Qi, Junyu ; Chen, Dingliang ; Qin, Yi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c176t-69f70411f854865faec45a14aa1f52d38809df89b8712d3018e5de288a9e815c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Accuracy</topic><topic>Artificial neural networks</topic><topic>Attention</topic><topic>Attention mechanism</topic><topic>Computational modeling</topic><topic>Consciousness</topic><topic>Constraints</topic><topic>continuous learning (CL)</topic><topic>deep learning. remaining useful life (RUL)</topic><topic>Degradation</topic><topic>Feature extraction</topic><topic>Life prediction</topic><topic>Methodology</topic><topic>Neural networks</topic><topic>Predictive models</topic><topic>Rotating machinery</topic><topic>Task analysis</topic><topic>Useful life</topic><topic>Vectors</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhou, Jianghong</creatorcontrib><creatorcontrib>Qi, Junyu</creatorcontrib><creatorcontrib>Chen, Dingliang</creatorcontrib><creatorcontrib>Qin, Yi</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005–Present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998–Present</collection><collection>IEEE Xplore</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>IEEE internet of things journal</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhou, Jianghong</au><au>Qi, Junyu</au><au>Chen, Dingliang</au><au>Qin, Yi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Continuous Remaining Useful Life Prediction by Self-Guided Attention Convolutional Neural Network and Memory Consciousness Adjustment</atitle><jtitle>IEEE internet of things journal</jtitle><stitle>JIoT</stitle><date>2024-10-01</date><risdate>2024</risdate><volume>11</volume><issue>19</issue><spage>31947</spage><epage>31958</epage><pages>31947-31958</pages><issn>2327-4662</issn><eissn>2327-4662</eissn><coden>IITJAU</coden><abstract>To accurately predict the remaining useful life (RUL) of rotating machinery while continuously providing the task data, a novel continuous RUL prediction methodology was proposed. The methodology comprises a self-guided attention convolutional neural network (SGACNN) and memory consciousness adjustment (MCA) mechanism. First, a multihead focal channel-wise self-attention (MFCWSA) mechanism was implemented to effectively capture the degradation information across all the channels and achieve the attentional focus. Next, the SGACNN was constructed using the MFCWSA, squeeze-and-excitation mechanism, and convolutional block attention module. A new network gradient direction was synthesized by leveraging the gradients from both the previous task and the current task. Further, a weight constraint loss term based on the gradient magnitude was designed to constrain the learning process of important parameters. With the new network gradient direction and weight constraint loss, a novel MCA mechanism was proposed and integrated into the SGACNN for implementing the continuous RUL prediction tasks. Finally, various RUL prediction experiments on the life-cycle bearing and gear data sets were carried out, and its outcomes were compared to those of the advanced methods of the same kind. The comparative results validated the superiority of the proposed methodology.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/JIOT.2024.3421673</doi><tpages>12</tpages><orcidid>https://orcid.org/0000-0002-3180-7654</orcidid><orcidid>https://orcid.org/0000-0003-1669-8987</orcidid><orcidid>https://orcid.org/0000-0001-7338-2407</orcidid><orcidid>https://orcid.org/0000-0002-2160-4300</orcidid></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | ISSN: 2327-4662 |
ispartof | IEEE internet of things journal, 2024-10, Vol.11 (19), p.31947-31958 |
issn | 2327-4662 2327-4662 |
language | eng |
recordid | cdi_crossref_primary_10_1109_JIOT_2024_3421673 |
source | IEEE Xplore |
subjects | Accuracy Artificial neural networks Attention Attention mechanism Computational modeling Consciousness Constraints continuous learning (CL) deep learning. remaining useful life (RUL) Degradation Feature extraction Life prediction Methodology Neural networks Predictive models Rotating machinery Task analysis Useful life Vectors |
title | Continuous Remaining Useful Life Prediction by Self-Guided Attention Convolutional Neural Network and Memory Consciousness Adjustment |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T10%3A58%3A54IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Continuous%20Remaining%20Useful%20Life%20Prediction%20by%20Self-Guided%20Attention%20Convolutional%20Neural%20Network%20and%20Memory%20Consciousness%20Adjustment&rft.jtitle=IEEE%20internet%20of%20things%20journal&rft.au=Zhou,%20Jianghong&rft.date=2024-10-01&rft.volume=11&rft.issue=19&rft.spage=31947&rft.epage=31958&rft.pages=31947-31958&rft.issn=2327-4662&rft.eissn=2327-4662&rft.coden=IITJAU&rft_id=info:doi/10.1109/JIOT.2024.3421673&rft_dat=%3Cproquest_RIE%3E3109489707%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3109489707&rft_id=info:pmid/&rft_ieee_id=10579798&rfr_iscdi=true |