Novel Semi-Supervised Seasonal-Trend VTN for Multimode Process IoT Soft Sensing

The rapid development of soft sensors has significantly enhanced industrial operations by promoting sustainability, safety, and efficiency. However, modern process industries involve highly dynamic systems with multimode and nonlinear data, posing challenges for conventional soft sensor models that...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE internet of things journal Jg. 12; H. 22; S. 47658 - 47666
Hauptverfasser:	He, Yan-Lin, Zhou, Yang-Xiao-Yu, Xu, Yuan, Zhu, Qun-Xiong, Li, Xingyuan
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Piscataway IEEE 15.11.2025 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:	Autoencoders Data models Feature extraction Gas turbines Internet of Things Multimode process Predictive models Probabilistic models soft sensing Soft sensors Statistical analysis Time series analysis Training transfer learning transformer Transformers variational Bayesian network
ISSN:	2327-4662, 2327-4662
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Abstract	The rapid development of soft sensors has significantly enhanced industrial operations by promoting sustainability, safety, and efficiency. However, modern process industries involve highly dynamic systems with multimode and nonlinear data, posing challenges for conventional soft sensor models that assume uniform data distributions. Additionally, limited target mode data hinders effective training. To address these issues, we propose the seasonal-trend variational transformer network (ST-VTN), a deep probabilistic model based on seasonal-trend decomposition. ST-VTN employs a Transformer-based encoder to extract seasonal and trend patterns from multimode data, enhancing latent Gaussian feature representation. A mode discriminator further improves mode-specific feature learning by distinguishing trend characteristics across modes. Built on a variational Bayesian framework, ST-VTN models the regression between latent features and quality variables. A two-stage pretraining and fine-tuning strategy enables effective use of source mode data, even with scarce target mode samples. Experiments on gas turbine and sulfur recovery datasets confirm ST-VTN's superior performance for multimode regression tasks.
AbstractList	The rapid development of soft sensors has significantly enhanced industrial operations by promoting sustainability, safety, and efficiency. However, modern process industries involve highly dynamic systems with multimode and nonlinear data, posing challenges for conventional soft sensor models that assume uniform data distributions. Additionally, limited target mode data hinders effective training. To address these issues, we propose the seasonal-trend variational transformer network (ST-VTN), a deep probabilistic model based on seasonal-trend decomposition. ST-VTN employs a Transformer-based encoder to extract seasonal and trend patterns from multimode data, enhancing latent Gaussian feature representation. A mode discriminator further improves mode-specific feature learning by distinguishing trend characteristics across modes. Built on a variational Bayesian framework, ST-VTN models the regression between latent features and quality variables. A two-stage pretraining and fine-tuning strategy enables effective use of source mode data, even with scarce target mode samples. Experiments on gas turbine and sulfur recovery datasets confirm ST-VTN's superior performance for multimode regression tasks.
Author	Xu, Yuan He, Yan-Lin Li, Xingyuan Zhu, Qun-Xiong Zhou, Yang-Xiao-Yu
Author_xml	– sequence: 1 givenname: Yan-Lin orcidid: 0000-0001-9999-3679 surname: He fullname: He, Yan-Lin email: heyanlin@ieee.org organization: College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, China – sequence: 2 givenname: Yang-Xiao-Yu surname: Zhou fullname: Zhou, Yang-Xiao-Yu email: 2024200765@buct.edu.cn organization: College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, China – sequence: 3 givenname: Yuan orcidid: 0000-0001-5490-2892 surname: Xu fullname: Xu, Yuan email: xuyuanbuct@163.com organization: College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, China – sequence: 4 givenname: Qun-Xiong orcidid: 0000-0002-2719-0516 surname: Zhu fullname: Zhu, Qun-Xiong email: buctielab@hotmail.com organization: College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, China – sequence: 5 givenname: Xingyuan orcidid: 0000-0002-3047-7219 surname: Li fullname: Li, Xingyuan email: yuan_l1106@163.com organization: College of Information Science and Technology, Beijing University of Chemical Technology, Beijing, China
BookMark	eNpNkE1Lw0AQhhepYK39AYKHgOfU2a9kc5TiR6W2QqPXJe7OSkqarbtNwX9vSgt6mhl4n5fhuSSD1rdIyDWFCaVQ3L3MluWEAZMTngFToM7IkHGWpyLL2ODffkHGMa4BoMckLbIhWS78HptkhZs6XXVbDPs6ou3vKvq2atIyYGuTj3KROB-S167Z1RtvMXkL3mCMycyXycq7XU-0sW6_rsi5q5qI49MckffHh3L6nM6XT7Pp_Tw1TKhdKqpPLnPpKoWuoCiBZkWeSyPBcCWUM4I5Y5VFEJI5FILnaMEaapWThaV8RG6PvdvgvzuMO732Xeg_jpqzvovlDLI-RY8pE3yMAZ3ehnpThR9NQR_U6YM6fVCnT-p65ubI1Ij4l6dUUAWM_wLZG2sK
CODEN	IITJAU
Cites_doi	10.1109/TNNLS.2022.3144162 10.1016/j.engappai.2023.106124 10.1109/TII.2023.3248059 10.1109/TCYB.2021.3090996 10.1109/TIE.2018.2803727 10.48550/ARXIV.1706.03762 10.1109/TNNLS.2021.3085869 10.1109/TII.2022.3183211 10.1016/j.jprocont.2022.09.008 10.1021/acs.iecr.5c00283 10.1109/JSEN.2023.3336789 10.1016/j.chemolab.2020.103981 10.1016/j.engappai.2022.105180 10.1109/TCYB.2020.3010331 10.48550/arXiv.1312.6114 10.3906/elk-1807-87 10.1109/TCYB.2022.3143613 10.1109/TIM.2024.3353844 10.1109/TIE.2020.2984443 10.1016/j.engappai.2022.105737 10.1002/cjce.23665 10.1109/TASE.2023.3309339 10.1109/TIM.2025.3572162 10.1109/TKDE.2023.3268125
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
DBID	97E RIA RIE AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D
DOI	10.1109/JIOT.2025.3602808
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE/IET Electronic Library (IEL) (UW System Shared) CrossRef Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional
DatabaseTitle	CrossRef Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional
DatabaseTitleList	Computer and Information Systems Abstracts
Database_xml	– sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Computer Science
EISSN	2327-4662
EndPage	47666
ExternalDocumentID	10_1109_JIOT_2025_3602808 11141802
Genre	orig-research
GrantInformation_xml	– fundername: National Natural Science Foundation of China grantid: 62073022 funderid: 10.13039/501100001809
GroupedDBID	0R~ 6IK 97E AAJGR AASAJ AAWTH ABAZT ABJNI ABQJQ ABVLG AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS IFIPE IPLJI JAVBF M43 OCL PQQKQ RIA RIE AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D
ID	FETCH-LOGICAL-c248t-4ab3575fa8ef91e50169775c50c3848fc42fcd8de0452fe4437ed0dc1d8f59d13
IEDL.DBID	RIE
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001611617900037&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2327-4662
IngestDate	Thu Nov 20 16:51:22 EST 2025 Sat Nov 29 06:54:34 EST 2025 Wed Nov 19 08:27:19 EST 2025
IsPeerReviewed	false
IsScholarly	true
Issue	22
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c248t-4ab3575fa8ef91e50169775c50c3848fc42fcd8de0452fe4437ed0dc1d8f59d13
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-3047-7219 0000-0001-5490-2892 0000-0001-9999-3679 0000-0002-2719-0516
PQID	3269727206
PQPubID	2040421
PageCount	9
ParticipantIDs	ieee_primary_11141802 proquest_journals_3269727206 crossref_primary_10_1109_JIOT_2025_3602808
PublicationCentury	2000
PublicationDate	2025-11-15
PublicationDateYYYYMMDD	2025-11-15
PublicationDate_xml	– month: 11 year: 2025 text: 2025-11-15 day: 15
PublicationDecade	2020
PublicationPlace	Piscataway
PublicationPlace_xml	– name: Piscataway
PublicationTitle	IEEE internet of things journal
PublicationTitleAbbrev	JIoT
PublicationYear	2025
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	ref13 Paige (ref24); 30 ref12 ref15 ref14 ref31 ref30 ref11 ref10 Oreshkin (ref22) 2019 ref2 ref1 ref17 Fortuna (ref26) 2007 Liu (ref32) 2023 ref18 Wen (ref19); 33 ref25 ref21 ref28 ref27 ref29 ref8 Lopez (ref16); 31 ref7 ref9 ref4 ref3 ref6 ref5 Wu (ref23); 34 Wen (ref20)
References_xml	– ident: ref8 doi: 10.1109/TNNLS.2022.3144162 – ident: ref13 doi: 10.1016/j.engappai.2023.106124 – ident: ref7 doi: 10.1109/TII.2023.3248059 – ident: ref14 doi: 10.1109/TCYB.2021.3090996 – volume: 33 start-page: 5409 volume-title: Proc. AAAI Conf. Artif. Intell. ident: ref19 article-title: RobustSTL: A robust seasonal-trend decomposition algorithm for long time series – volume: 31 start-page: 6117 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref16 article-title: Information constraints on auto-encoding variational Bayes – ident: ref4 doi: 10.1109/TIE.2018.2803727 – volume-title: Soft Sensors for Monitoring and Control of Industrial Processes year: 2007 ident: ref26 – ident: ref31 doi: 10.48550/ARXIV.1706.03762 – ident: ref29 doi: 10.1109/TNNLS.2021.3085869 – ident: ref17 doi: 10.1109/TII.2022.3183211 – ident: ref9 doi: 10.1016/j.jprocont.2022.09.008 – ident: ref3 doi: 10.1021/acs.iecr.5c00283 – ident: ref10 doi: 10.1109/JSEN.2023.3336789 – ident: ref5 doi: 10.1016/j.chemolab.2020.103981 – volume: 34 start-page: 22419 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref23 article-title: Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting – ident: ref12 doi: 10.1016/j.engappai.2022.105180 – year: 2019 ident: ref22 article-title: N-BEATS: Neural basis expansion analysis for interpretable time series forecasting publication-title: arXiv:1905.10437 – ident: ref28 doi: 10.1109/TCYB.2020.3010331 – ident: ref11 doi: 10.48550/arXiv.1312.6114 – ident: ref25 doi: 10.3906/elk-1807-87 – ident: ref15 doi: 10.1109/TCYB.2022.3143613 – year: 2023 ident: ref32 article-title: iTransformer: Inverted transformers are effective for time series forecasting publication-title: arXiv:2310.06625 – ident: ref21 doi: 10.1109/TIM.2024.3353844 – ident: ref6 doi: 10.1109/TIE.2020.2984443 – ident: ref27 doi: 10.1016/j.engappai.2022.105737 – ident: ref30 doi: 10.1002/cjce.23665 – start-page: 4653 volume-title: Proc. 13th Int. Joint Conf. Artif. Intell. ident: ref20 article-title: Time series data augmentation for deep learning: A survey – ident: ref1 doi: 10.1109/TASE.2023.3309339 – ident: ref2 doi: 10.1109/TIM.2025.3572162 – ident: ref18 doi: 10.1109/TKDE.2023.3268125 – volume: 30 start-page: 5927 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref24 article-title: Learning disentangled representations with semi-supervised deep generative models
SSID	ssj0001105196
Score	2.3573287
Snippet	The rapid development of soft sensors has significantly enhanced industrial operations by promoting sustainability, safety, and efficiency. However, modern...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Index Database Publisher
StartPage	47658
SubjectTerms	Autoencoders Data models Feature extraction Gas turbines Internet of Things Multimode process Predictive models Probabilistic models soft sensing Soft sensors Statistical analysis Time series analysis Training transfer learning transformer Transformers variational Bayesian network
Title	Novel Semi-Supervised Seasonal-Trend VTN for Multimode Process IoT Soft Sensing
URI	https://ieeexplore.ieee.org/document/11141802 https://www.proquest.com/docview/3269727206
Volume	12
WOSCitedRecordID	wos001611617900037&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 2327-4662 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001105196 issn: 2327-4662 databaseCode: RIE dateStart: 20140101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS8MwFA9uePDi_Jg4nZKDJyFbs6YfOYo4nEgnrMpupU1eYaDt2Nffb16WMUQ8eGuhCeW9vO_83iPkLjQ2jYMnmDkOmgngBSu8QjFP5UILGapc2aklr1GSxNOpfHNgdYuFAQB7-Qx6-Ghr-bpWa0yV9Y1cCuxY1iCNKAq3YK19QoWjNxK6yiX3ZP9lNE5NBDgIen6IFcT4h-2xw1R-aWBrVoatf_7QCTl2_iN92DL8lBxAdUZau9kM1InqORkn9QY-6QS-ZmyynqNGWII277n1vZm9C0s_0oQar5VaGC4OxaEOOEBHdUonRkWbFRWmE9rkffiUPj4zNzyBqYGIV0zkhW9csTKPoZQcAuy6EkWBCjzlxyIulRiUSscasKd6CUL4EWhPK67jMpCa-xekWdUVXBIKPM9L7BMmwTAwMhGe8vwCUbc6KE0A2SH3O7Jm822PjMzGFp7MkAcZ8iBzPOiQNtJx_6EjYYd0d5zInBgtM-NbSlspDq_-WHZNjnB3RAfyoEuaq8Uabsih2qxmy8WtPSHf4fO5xA
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB58gV58i_WZgychmnSzr6OIxWpdha7S27KbzEJB29KHv99MmiIiHrztwoYsM5l3vhmAi8jaNIlCcXscDFcoK16JSnOhS2VUGulSu6klnTjLkl4vffFgdYeFQUR3-Qyv6NHV8s1QzyhVdm3lUlHHsmVYDZVqijlc6zulIskfiXztUor0-qH9nNsYsBleBRHVEJMf1seNU_mlg51haW3985e2YdN7kOxmzvIdWMLBLmwtpjMwL6x78JwNP_GddfGjz7uzEemECRr7Xjrvm7vbsOwtz5j1W5kD4tJYHOahA6w9zFnXKmm7YkAJhX14bd3lt_fcj0_guqmSKVdlFVhnrC4TrFOJIfVdieNQh0IHiUpqrZq1NolB6qpeo1JBjEYYLU1Sh6mRwQGsDIYDPASGsixr6hSWomVhbGM8LYKKcLcmrG0I2YDLBVmL0bxLRuGiC5EWxIOCeFB4HjRgn-j4_aEnYQNOFpwovCBNCutdpq5WHB39sewc1u_zp07RaWePx7BBOxFWUIYnsDIdz_AU1vTntD8Zn7nT8gW4q70L
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Novel+Semi-Supervised+Seasonal-Trend+VTN+for+Multimode+Process+IoT+Soft+Sensing&rft.jtitle=IEEE+internet+of+things+journal&rft.au=Yan-Lin%2C+He&rft.au=Yang-Xiao-Yu%2C+Zhou&rft.au=Xu%2C+Yuan&rft.au=Zhu%2C+Qun-Xiong&rft.date=2025-11-15&rft.pub=The+Institute+of+Electrical+and+Electronics+Engineers%2C+Inc.+%28IEEE%29&rft.eissn=2327-4662&rft.volume=12&rft.issue=23&rft.spage=47658&rft.epage=47666&rft_id=info:doi/10.1109%2FJIOT.2025.3602808&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2327-4662&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2327-4662&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2327-4662&client=summon