LSTM-MPC: A Deep Learning Based Predictive Control Method for Multimode Process Control

Modern industrial processes often operate under different modes, which brings challenges to model predictive control (MPC). Recently, most MPC related methods would establish prediction models independently for different modes, which results in their control effect highly relying on switching strate...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE transactions on industrial electronics (1982) Ročník 70; číslo 11; s. 1 - 10
Hlavní autoři:	Huang, Keke, Wei, Ke, Li, Fanbiao, Yang, Chunhua, Gui, Weihua
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York IEEE 01.11.2023 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Adaptation models Behavioral sciences Control methods Deep learning Feasibility studies long short-term memory network Mathematical models model predictive control multimode process Optimization Prediction models Predictive control Predictive models Process controls Stability analysis Switches Switching
ISSN:	0278-0046, 1557-9948
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	Modern industrial processes often operate under different modes, which brings challenges to model predictive control (MPC). Recently, most MPC related methods would establish prediction models independently for different modes, which results in their control effect highly relying on switching strategies. Inspired by the powerful representation capabilities of deep learning, this paper proposed a deep learning based MPC method. Specifically, the LSTM network is applied to predict behaviours of controlled system, which can automatically match different operation modes without switching strategy. Then combined with MPC framework, an adaptive gradient descent method is introduced to handle optimization problem and its constraints. In addition, stability and feasibility analysis have been conducted from the aspect of theory to ensure practical application of the proposed method. Experiments on a numerical simulation process and an industrial process platform show the strength and reliability of the proposed method, which reduces the overshoot by about 10<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> compared to common learning-based MPC methods and improves the control accuracy effectively.
AbstractList	Modern industrial processes often operate under different modes, which brings challenges to model predictive control (MPC). Recently, most MPC related methods would establish prediction models independently for different modes, which results in their control effect highly relying on switching strategies. Inspired by the powerful representation capabilities of deep learning, this paper proposed a deep learning based MPC method. Specifically, the LSTM network is applied to predict behaviours of controlled system, which can automatically match different operation modes without switching strategy. Then combined with MPC framework, an adaptive gradient descent method is introduced to handle optimization problem and its constraints. In addition, stability and feasibility analysis have been conducted from the aspect of theory to ensure practical application of the proposed method. Experiments on a numerical simulation process and an industrial process platform show the strength and reliability of the proposed method, which reduces the overshoot by about 10<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> compared to common learning-based MPC methods and improves the control accuracy effectively. Modern industrial processes often operate under different modes, which brings challenges to model predictive control (MPC). Recently, most MPC related methods would establish prediction models independently for different modes, which results in their control effect highly relying on switching strategies. Inspired by the powerful representation capabilities of deep learning, this article proposed a deep learning based MPC method. Specifically, the LSTM network is applied to predict behaviors of controlled system, which can automatically match different operation modes without switching strategy. Then combined with MPC framework, an adaptive gradient descent method is introduced to handle optimization problem and its constraints. In addition, stability and feasibility analysis have been conducted from the aspect of theory to ensure practical application of the proposed method. Experiments on a numerical simulation process and an industrial process platform show the strength and reliability of the proposed method, which reduces the overshoot by about 10[Formula Omitted] compared to common learning-based MPC methods and improves the control accuracy effectively.
Author	Gui, Weihua Huang, Keke Wei, Ke Li, Fanbiao Yang, Chunhua
Author_xml	– sequence: 1 givenname: Keke orcidid: 0000-0003-3553-3424 surname: Huang fullname: Huang, Keke organization: School of Automation, Central South University, Changsha, China – sequence: 2 givenname: Ke surname: Wei fullname: Wei, Ke organization: School of Automation, Central South University, Changsha, China – sequence: 3 givenname: Fanbiao surname: Li fullname: Li, Fanbiao organization: School of Automation, Central South University, Changsha, China – sequence: 4 givenname: Chunhua orcidid: 0000-0002-3770-9887 surname: Yang fullname: Yang, Chunhua organization: School of Automation, Central South University, Changsha, China – sequence: 5 givenname: Weihua surname: Gui fullname: Gui, Weihua organization: School of Automation, Central South University, Changsha, China
BookMark	eNp9kTFPwzAQRi0EEm1hR2KxxJxyduLEYSulQKVGVKKIMXLsC7hq42KnSPx7UrUwMDDd8r37dO_65LhxDRJywWDIGOTXi-lkyIHzYcx5HvP4iPSYEFmU54k8Jj3gmYwAkvSU9ENYArBEMNEjr7PnRREV8_ENHdE7xA2dofKNbd7orQpo6Nyjsbq1n0jHrmm9W9EC23dnaO08Lbar1q6dwS7nNIbwEzojJ7VaBTw_zAF5uZ8sxo_R7OlhOh7NIs1z1kZxVUmjGZdSS66MTOoqEQaUBBSxyJSsmJaKg6kBqiRNM26URqzrLE-rOmfxgFzt9268-9hiaMul2_qmqyy5ZCyLQUDapWCf0t6F4LEuN96ulf8qGZQ7fWWnr9zpKw_6OiT9g2jbqtbuzlN29R94uQctIv725N0fMiHib-0ifY0
CODEN	ITIED6
CitedBy_id	crossref_primary_10_1016_j_apm_2025_116348 crossref_primary_10_1016_j_energy_2025_135701 crossref_primary_10_29109_gujsc_1726037 crossref_primary_10_1016_j_asoc_2025_113674 crossref_primary_10_1016_j_apenergy_2024_124164 crossref_primary_10_1038_s41598_025_01290_x crossref_primary_10_3390_electronics13214241 crossref_primary_10_1016_j_neucom_2025_129712 crossref_primary_10_1109_ACCESS_2025_3554168 crossref_primary_10_1109_TCST_2023_3279623 crossref_primary_10_1016_j_isatra_2024_04_019 crossref_primary_10_1109_TASE_2024_3524472 crossref_primary_10_1016_j_chaos_2023_113739 crossref_primary_10_3390_wevj16090524 crossref_primary_10_3390_app151910329 crossref_primary_10_1109_JSEN_2024_3401678 crossref_primary_10_1007_s10489_023_05157_4 crossref_primary_10_1109_TIA_2024_3351959 crossref_primary_10_1007_s42835_024_01982_w crossref_primary_10_3390_s25072014 crossref_primary_10_1080_00207179_2024_2366435 crossref_primary_10_1016_j_trd_2025_104896 crossref_primary_10_1007_s10845_024_02438_8 crossref_primary_10_1109_JAS_2024_124212 crossref_primary_10_1115_1_4067768 crossref_primary_10_1109_TII_2024_3393503 crossref_primary_10_3390_wevj16040206 crossref_primary_10_1109_TITS_2024_3421626 crossref_primary_10_1115_1_4069104 crossref_primary_10_1109_JSYST_2024_3406800 crossref_primary_10_1109_TASE_2025_3529124 crossref_primary_10_1007_s00521_024_09710_1 crossref_primary_10_1016_j_neucom_2025_129418 crossref_primary_10_1109_JAS_2024_124929 crossref_primary_10_3390_electronics14102000 crossref_primary_10_1016_j_neucom_2025_131048 crossref_primary_10_1109_TIA_2023_3335033 crossref_primary_10_3390_a16080387 crossref_primary_10_1007_s10489_023_05234_8 crossref_primary_10_1088_1361_6501_ad8947 crossref_primary_10_1177_01423312241262079 crossref_primary_10_1109_TIE_2023_3323745 crossref_primary_10_1007_s11071_024_10342_2 crossref_primary_10_1109_TII_2025_3574400 crossref_primary_10_1080_02626667_2025_2464949 crossref_primary_10_1109_TCSII_2025_3557419 crossref_primary_10_1016_j_applthermaleng_2025_126202 crossref_primary_10_1109_TII_2025_3567401 crossref_primary_10_1109_TSMC_2023_3341031 crossref_primary_10_1109_TII_2024_3438234 crossref_primary_10_3390_a18080508
Cites_doi	10.1016/j.arcontrol.2004.01.004 10.1016/j.neunet.2019.09.035 10.1016/j.conengprac.2020.104319 10.1016/j.powtec.2019.10.065 10.1109/TNNLS.2013.2261574 10.1109/TNNLS.2016.2561702 10.1109/TIE.2020.2979560 10.1016/S1474-6670(17)58026-X 10.1109/TNNLS.2015.2465174 10.1016/j.oceaneng.2018.06.068 10.1109/TIE.2019.2903743 10.1016/j.conengprac.2010.05.012 10.1109/TNNLS.2020.3015869 10.1002/aic.16729 10.1109/TII.2020.2969729 10.1007/s10489-021-02770-z 10.1109/TIE.2013.2261039 10.1016/j.conengprac.2015.08.009 10.1016/S0005-1098(99)00113-2 10.1109/TCYB.2022.3178878 10.1109/TIE.2011.2169636 10.1109/TII.2022.3161640 10.1016/j.ins.2003.10.011 10.1109/72.279181 10.1109/TIE.2013.2266086 10.1109/TIE.2015.2492946 10.1109/JIOT.2021.3078330 10.1109/TAC.2017.2753460 10.1016/j.jprocont.2020.05.009 10.1016/j.knosys.2019.105006 10.1016/j.jprocont.2017.02.014 10.1109/TIE.2017.2779442
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
DBID	97E RIA RIE AAYXX CITATION 7SP 8FD L7M
DOI	10.1109/TIE.2022.3229323
DatabaseName	IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) Online IEEE Electronic Library (IEL) CrossRef Electronics & Communications Abstracts Technology Research Database Advanced Technologies Database with Aerospace
DatabaseTitle	CrossRef Technology Research Database Advanced Technologies Database with Aerospace Electronics & Communications Abstracts
DatabaseTitleList	Technology Research Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE Xplore url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	1557-9948
EndPage	10
ExternalDocumentID	10_1109_TIE_2022_3229323 9994755
Genre	orig-research
GrantInformation_xml	– fundername: Major Key Project of Peng Cheng Laboratory (PCL) grantid: PCL2021A09 – fundername: National Natural Science Foundation of China grantid: 62073340; 61860206014 funderid: 10.13039/501100001809 – fundername: National Key R&D Program of China grantid: 2022YFB3304900; 2019YFB1705300 – fundername: Science and Technology Innovation Program of Hunan Province grantid: 2022JJ10083; 2021RC3018; 2021RC4054
GroupedDBID	-~X .DC 0R~ 29I 4.4 5GY 6IK 97E AAJGR AASAJ AAWTH ABQJQ ABVLG ACGFO ACGFS ACIWK ACNCT AENEX AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS HZ~ IFIPE IPLJI JAVBF LAI M43 MS~ O9- OCL P2P RIA RIE RNS TAE TN5 TWZ 5VS 9M8 AAYXX ACKIV AETIX AGSQL AI. AIBXA ALLEH CITATION EJD H~9 IBMZZ ICLAB IFJZH VH1 VJK 7SP 8FD AARMG ABAZT L7M
ID	FETCH-LOGICAL-c291t-3bb8dc1288c82ad84fb45d0a80e5357a8b1c8a20df00b46672daceeff796bf913
IEDL.DBID	RIE
ISICitedReferencesCount	61
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000986714600072&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0278-0046
IngestDate	Mon Jun 30 10:12:43 EDT 2025 Tue Nov 18 22:24:33 EST 2025 Sat Nov 29 01:31:59 EST 2025 Tue Nov 25 14:44:27 EST 2025
IsPeerReviewed	true
IsScholarly	true
Issue	11
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-c291t-3bb8dc1288c82ad84fb45d0a80e5357a8b1c8a20df00b46672daceeff796bf913
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0003-3553-3424 0000-0002-3770-9887
PQID	2811730506
PQPubID	85464
PageCount	10
ParticipantIDs	crossref_primary_10_1109_TIE_2022_3229323 crossref_citationtrail_10_1109_TIE_2022_3229323 proquest_journals_2811730506 ieee_primary_9994755
PublicationCentury	2000
PublicationDate	2023-11-01
PublicationDateYYYYMMDD	2023-11-01
PublicationDate_xml	– month: 11 year: 2023 text: 2023-11-01 day: 01
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE transactions on industrial electronics (1982)
PublicationTitleAbbrev	TIE
PublicationYear	2023
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	ref12 ref15 ref14 ref31 ref30 huang (ref21) 2021; 52 ref11 ref33 ref10 ref32 ref2 ref1 ref17 ref16 ref19 ref18 han (ref27) 2013; 24 ref24 ref23 ref26 ref25 ref20 ref22 ref28 ref29 ref8 ref7 ref9 ref4 ref3 ref6 ref5 tang (ref13) 0; 29
References_xml	– ident: ref11 doi: 10.1016/j.arcontrol.2004.01.004 – ident: ref16 doi: 10.1016/j.neunet.2019.09.035 – ident: ref26 doi: 10.1016/j.conengprac.2020.104319 – ident: ref25 doi: 10.1016/j.powtec.2019.10.065 – volume: 24 start-page: 1425 year: 2013 ident: ref27 article-title: Real-time model predictive control using a self-organizing neural network publication-title: IEEE Trans Neural Netw Learn Syst doi: 10.1109/TNNLS.2013.2261574 – ident: ref32 doi: 10.1109/TNNLS.2016.2561702 – ident: ref17 doi: 10.1109/TIE.2020.2979560 – ident: ref10 doi: 10.1016/S1474-6670(17)58026-X – ident: ref18 doi: 10.1109/TNNLS.2015.2465174 – ident: ref14 doi: 10.1016/j.oceaneng.2018.06.068 – ident: ref9 doi: 10.1109/TIE.2019.2903743 – ident: ref33 doi: 10.1016/j.conengprac.2010.05.012 – ident: ref15 doi: 10.1109/TNNLS.2020.3015869 – ident: ref28 doi: 10.1002/aic.16729 – ident: ref19 doi: 10.1109/TII.2020.2969729 – volume: 52 start-page: 14773 year: 2021 ident: ref21 article-title: A multi-mode traffic flow prediction method with clustering based attention convolution LSTM publication-title: Appl Intell doi: 10.1007/s10489-021-02770-z – volume: 29 start-page: 622 year: 0 ident: ref13 article-title: Multi-model MPC for nonlinear systems: Case study of a complex PH neutralization process publication-title: Proc 21st Eur Symp Comput Aided Process Eng – ident: ref4 doi: 10.1109/TIE.2013.2261039 – ident: ref5 doi: 10.1016/j.conengprac.2015.08.009 – ident: ref30 doi: 10.1016/S0005-1098(99)00113-2 – ident: ref7 doi: 10.1109/TCYB.2022.3178878 – ident: ref20 doi: 10.1109/TIE.2011.2169636 – ident: ref8 doi: 10.1109/TII.2022.3161640 – ident: ref12 doi: 10.1016/j.ins.2003.10.011 – ident: ref24 doi: 10.1109/72.279181 – ident: ref3 doi: 10.1109/TIE.2013.2266086 – ident: ref2 doi: 10.1109/TIE.2015.2492946 – ident: ref23 doi: 10.1109/JIOT.2021.3078330 – ident: ref29 doi: 10.1109/TAC.2017.2753460 – ident: ref31 doi: 10.1016/j.jprocont.2020.05.009 – ident: ref22 doi: 10.1016/j.knosys.2019.105006 – ident: ref6 doi: 10.1016/j.jprocont.2017.02.014 – ident: ref1 doi: 10.1109/TIE.2017.2779442
SSID	ssj0014515
Score	2.619821
Snippet	Modern industrial processes often operate under different modes, which brings challenges to model predictive control (MPC). Recently, most MPC related methods...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	1
SubjectTerms	Adaptation models Behavioral sciences Control methods Deep learning Feasibility studies long short-term memory network Mathematical models model predictive control multimode process Optimization Prediction models Predictive control Predictive models Process controls Stability analysis Switches Switching
Title	LSTM-MPC: A Deep Learning Based Predictive Control Method for Multimode Process Control
URI	https://ieeexplore.ieee.org/document/9994755 https://www.proquest.com/docview/2811730506
Volume	70
WOSCitedRecordID	wos000986714600072&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 Xplore customDbUrl: eissn: 1557-9948 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014515 issn: 0278-0046 databaseCode: RIE dateStart: 19820101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH5sw4Me_DXF6ZQcvAh2a9O0SbzNuaHgxsCJu5WmSUSQbeyHf79J2hVFEbz18ALlvfR9L3353gdwSVMcMelov1x5JCQmD2pFPKFNrS9Dg2Apc2ITdDhkkwkfVeC65MIopdzlM9Wyj66XL2fZ2v4qa5tihtAoqkKV0jjnapUdAxLlagXYTow1h75NS9Ln7fFDzxwEMW6ZzWvKlfAbBDlNlR-J2KFLf-9_77UPu0UViTp52A-goqaHsPNltmAdXh6fxgNvMOreoA66U2qOilGqr-jWIJdEo4Xt0dhsh7r5fXU0cHLSyNSxyBFzrUwOKqgEG6MjeO73xt17r1BR8DLMg5UXCsFkZmCIZQynkhEtSCT9lPkqCiOaMhFkLMW-1L4vSBxTLFODnFpTHgvNg_AYatPZVJ0ASjVRIQ9VQIQkBvIY5pGOpYxNXSgpDxrQ3jg2yYoR41bp4j1xRw2fJyYUiQ1FUoSiAVflink-XuMP27p1fWlXeL0BzU3skuL7WybY8mdNKvPj099XncG2FY7PWYVNqK0Wa3UOW9nH6m25uHBb6xPOBsmi
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dS8MwED_mB6gPfk1xfubBF8FuaZq0iW86NyauY2BF30rbJCLIJnP695uk3VAUwbc-XKDcpfe79PK7H8BplBHGpaP9CuXRgJo8qBX1cm1qfRkYBMu4E5uIBgP--CiGNTifc2GUUu7ymWraR9fLl-Pi3f4qa5lihkaMLcASo5Tgkq017xlQVuoVEDsz1hz7Zk1JLFrJTcccBQlpmu1rCpbgGwg5VZUfqdjhS3fjf2-2CetVHYkuy8BvQU2NtmHty3TBOjz075LYi4ftC3SJrpV6RdUw1Sd0ZbBLouHEdmlsvkPt8sY6ip2gNDKVLHLUXCuUgyoywcxoB-67naTd8yodBa8gwp96QZ5zWRgg4gUnmeRU55RJnHGsWMCijOd-wTOCpcY4p2EYEZkZ7NQ6EmGuhR_swuJoPFJ7gDJNVSAC5dNcUgN6nAimQylDUxnKSPgNaM0cmxbVkHGrdfGSusMGFqkJRWpDkVahaMDZfMVrOWDjD9u6df3crvJ6Aw5nsUurL_AtJZZBa5IZDvd_X3UCK70k7qf9m8HtAaxaGfmSY3gIi9PJuzqC5eJj-vw2OXbb7BN21Mzp
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=LSTM-MPC%3A+A+Deep+Learning+Based+Predictive+Control+Method+for+Multimode+Process+Control&rft.jtitle=IEEE+transactions+on+industrial+electronics+%281982%29&rft.au=Huang%2C+Keke&rft.au=Wei%2C+Ke&rft.au=Li%2C+Fanbiao&rft.au=Yang%2C+Chunhua&rft.date=2023-11-01&rft.pub=IEEE&rft.issn=0278-0046&rft.spage=1&rft.epage=10&rft_id=info:doi/10.1109%2FTIE.2022.3229323&rft.externalDocID=9994755
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0278-0046&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0278-0046&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0278-0046&client=summon