Using stochastic programming to train neural network approximation of nonlinear MPC laws

To facilitate the real-time implementation of nonlinear model predictive control (NMPC), this paper proposes a deep learning-based NMPC scheme, in which the NMPC law is approximated via a deep neural network (DNN). To optimize the DNN controller, a novel “optimize and train” architecture is designed...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Automatica (Oxford) Ročník 146; s. 110665
Hlavní autori: Li, Yun, Hua, Kaixun, Cao, Yankai
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Ltd 01.12.2022
Predmet:
ISSN:0005-1098, 1873-2836
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract To facilitate the real-time implementation of nonlinear model predictive control (NMPC), this paper proposes a deep learning-based NMPC scheme, in which the NMPC law is approximated via a deep neural network (DNN). To optimize the DNN controller, a novel “optimize and train” architecture is designed, where the processes of data generation and neural network training are combined together to result in a single large-scale stochastic optimization problem. Unlike the conventional “optimize then train” approach, our proposed one directly optimizes the closed-loop performance of the DNN controller over a finite horizon for a number of initial states. The important features of our proposed scheme are that it can deal with set-valued optimal MPC input, and a probabilistic guarantee of constraint satisfaction can be concluded for the closed-loop system without simulating the DNN controller. With our proposed scheme, an increased number of training scenarios leads to improved constraint satisfaction of the derived DNN controller, which is not necessarily true for the “optimize then train” approach. Statistical approaches for validating closed-loop control performance are also discussed. Furthermore, computational methods are introduced to efficiently solve the resulting stochastic optimization problem. The effectiveness of the proposed scheme is extensively illustrated with several numerical simulations. Compared with the conventional “optimize then train” approach, our proposed approach exhibits better closed-loop constraint satisfaction for all considered case studies.
AbstractList To facilitate the real-time implementation of nonlinear model predictive control (NMPC), this paper proposes a deep learning-based NMPC scheme, in which the NMPC law is approximated via a deep neural network (DNN). To optimize the DNN controller, a novel “optimize and train” architecture is designed, where the processes of data generation and neural network training are combined together to result in a single large-scale stochastic optimization problem. Unlike the conventional “optimize then train” approach, our proposed one directly optimizes the closed-loop performance of the DNN controller over a finite horizon for a number of initial states. The important features of our proposed scheme are that it can deal with set-valued optimal MPC input, and a probabilistic guarantee of constraint satisfaction can be concluded for the closed-loop system without simulating the DNN controller. With our proposed scheme, an increased number of training scenarios leads to improved constraint satisfaction of the derived DNN controller, which is not necessarily true for the “optimize then train” approach. Statistical approaches for validating closed-loop control performance are also discussed. Furthermore, computational methods are introduced to efficiently solve the resulting stochastic optimization problem. The effectiveness of the proposed scheme is extensively illustrated with several numerical simulations. Compared with the conventional “optimize then train” approach, our proposed approach exhibits better closed-loop constraint satisfaction for all considered case studies.
ArticleNumber 110665
Author Cao, Yankai
Hua, Kaixun
Li, Yun
Author_xml – sequence: 1
  givenname: Yun
  surname: Li
  fullname: Li, Yun
  email: y.li-39@tudelft.nl
– sequence: 2
  givenname: Kaixun
  surname: Hua
  fullname: Hua, Kaixun
  email: huakaixun@gmail.com
– sequence: 3
  givenname: Yankai
  surname: Cao
  fullname: Cao, Yankai
  email: yankai.cao@ubc.ca
BookMark eNqNkN1KAzEQhYNUsFbfIS-wa366u-mNoMU_qOiFBe_CNJvU1G1SktTq25u1guCNXh1mmPnmzDlGA-edRghTUlJC67NVCdvk15CsgpIRxkpKSV1XB2hIRcMLJng9QENCSFVQMhFH6DjGVS7HVLAhep5H65Y4Jq9eIGYI3gS_DLBe9-3kcQpgHXZ6G6DLknY-vGLY5Kl321_1DnuDs6nOOg0B3z9OcQe7eIIODXRRn37rCM2vr56mt8Xs4eZuejErFKciFdmxanlVZTWCGTrmzYKwRQNAKeembTW0ilWNAVYTVTdiPDGTtlELyitdV5SP0Pmeq4KPMWgjlU1fvnrnnaRE9jnJlfzJSfY5yX1OGSB-ATYhfxY-_rN6uV_V-cE3q4OMymqndGuDVkm23v4N-QTvnIz3
CitedBy_id crossref_primary_10_1002_rnc_70057
crossref_primary_10_3390_act14080402
crossref_primary_10_1007_s40435_024_01426_3
crossref_primary_10_1016_j_patcog_2025_111650
crossref_primary_10_1016_j_jprocont_2024_103228
crossref_primary_10_1016_j_asoc_2025_113882
crossref_primary_10_1016_j_engappai_2024_109009
crossref_primary_10_1016_j_compchemeng_2025_109096
crossref_primary_10_1109_TITS_2023_3342651
crossref_primary_10_1016_j_jprocont_2024_103270
crossref_primary_10_1109_TTE_2025_3543510
crossref_primary_10_1109_TAES_2025_3551283
crossref_primary_10_1016_j_compchemeng_2023_108511
crossref_primary_10_1016_j_ins_2024_120970
crossref_primary_10_1016_j_asr_2024_07_019
crossref_primary_10_1016_j_conengprac_2025_106480
crossref_primary_10_1016_j_jprocont_2024_103302
crossref_primary_10_1002_aic_18644
crossref_primary_10_1016_j_jprocont_2024_103353
crossref_primary_10_1016_j_jprocont_2023_103144
Cites_doi 10.1016/j.ifacol.2018.09.373
10.1002/rnc.5696
10.1016/j.compchemeng.2017.01.021
10.1016/S0005-1098(01)00174-1
10.1016/0893-6080(90)90005-6
10.1109/TAC.2014.2351991
10.1016/j.compchemeng.2014.09.013
10.1016/j.ifacol.2018.11.036
10.1109/TCYB.2020.2999556
10.1016/j.ifacol.2020.12.546
10.1016/j.compchemeng.2018.08.036
10.1016/j.automatica.2014.11.004
10.1109/LCSYS.2018.2843682
10.1109/TAC.2002.805688
10.1109/TAC.2013.2275667
10.1561/2600000008
10.1109/TCST.2020.3024571
10.1016/j.compchemeng.2019.03.009
10.1016/0005-1098(95)00044-W
10.1016/j.ifacol.2020.12.538
10.1016/j.compchemeng.2020.107174
10.1109/TAC.2009.2031207
10.1016/j.jprocont.2020.06.012
10.1007/s10589-015-9813-x
10.1007/s10107-004-0559-y
10.1016/j.compchemeng.2020.107133
10.1109/TAC.2011.2141410
10.1016/j.automatica.2011.02.029
10.1016/j.ces.2007.05.022
ContentType Journal Article
Copyright 2022 Elsevier Ltd
Copyright_xml – notice: 2022 Elsevier Ltd
DBID AAYXX
CITATION
DOI 10.1016/j.automatica.2022.110665
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1873-2836
ExternalDocumentID 10_1016_j_automatica_2022_110665
S0005109822005295
GrantInformation_xml – fundername: Natural Science and Engineering Research Council of Canada
  grantid: RGPIN-2019-05499
  funderid: http://dx.doi.org/10.13039/501100000038
GroupedDBID --K
--M
-~X
.DC
.~1
0R~
1B1
1~.
1~5
23N
3R3
4.4
457
4G.
5GY
5VS
6TJ
7-5
71M
8P~
9JN
9JO
AAAKF
AAAKG
AABNK
AACTN
AAEDT
AAEDW
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AARIN
AAXUO
ABDEX
ABFNM
ABFRF
ABJNI
ABMAC
ABUCO
ABXDB
ABYKQ
ACBEA
ACDAQ
ACGFO
ACGFS
ACNNM
ACRLP
ADBBV
ADEZE
ADIYS
ADMUD
ADTZH
AEBSH
AECPX
AEFWE
AEKER
AENEX
AFFNX
AFKWA
AFTJW
AGHFR
AGUBO
AGYEJ
AHHHB
AHJVU
AHPGS
AI.
AIEXJ
AIKHN
AITUG
AJBFU
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
APLSM
ASPBG
AVWKF
AXJTR
AZFZN
BJAXD
BKOJK
BLXMC
CS3
EBS
EFJIC
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
HAMUX
HLZ
HVGLF
HZ~
H~9
IHE
J1W
JJJVA
K-O
KOM
LG9
LY7
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RIG
ROL
RPZ
RXW
SBC
SDF
SDG
SDP
SES
SET
SEW
SPC
SPCBC
SSB
SSD
SST
SSZ
T5K
T9H
TAE
TN5
VH1
WH7
WUQ
X6Y
XFK
XPP
ZMT
~G-
77I
9DU
AATTM
AAXKI
AAYWO
AAYXX
ABUFD
ABWVN
ACLOT
ACRPL
ACVFH
ADCNI
ADNMO
AEIPS
AEUPX
AFJKZ
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKRWK
AKYEP
ANKPU
APXCP
CITATION
EFKBS
~HD
ID FETCH-LOGICAL-c318t-202cd355202f82f1437b02b7aa1133fddeadc257fa260c67849f9d7cb135e6513
ISICitedReferencesCount 26
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000878605100002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0005-1098
IngestDate Sat Nov 29 07:32:04 EST 2025
Tue Nov 18 22:39:59 EST 2025
Fri Feb 23 02:39:39 EST 2024
IsPeerReviewed true
IsScholarly true
Keywords Model predictive control
Parallel computation
Deep neural networks
Stochastic optimization
Policy learning
Nonlinear systems
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c318t-202cd355202f82f1437b02b7aa1133fddeadc257fa260c67849f9d7cb135e6513
ParticipantIDs crossref_citationtrail_10_1016_j_automatica_2022_110665
crossref_primary_10_1016_j_automatica_2022_110665
elsevier_sciencedirect_doi_10_1016_j_automatica_2022_110665
PublicationCentury 2000
PublicationDate December 2022
2022-12-00
PublicationDateYYYYMMDD 2022-12-01
PublicationDate_xml – month: 12
  year: 2022
  text: December 2022
PublicationDecade 2020
PublicationTitle Automatica (Oxford)
PublicationYear 2022
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Parisini, Zoppoli (b26) 1995; 31
Patrinos, Bemporad (b27) 2013; 59
Vaupel, Hamacher, Caspari, Mhamdi, Kevrekidis, Mitsos (b34) 2020; 92
Cao, Gopaluni (b11) 2020; 53
Maddalena, Moraes, Waltrich, Jones (b25) 2020; 53
Jerez, Goulart, Richter, Constantinides, Kerrigan, Morari (b19) 2014; 59
Zhang, Bujarbaruah, Borrelli (b39) 2020; 29
Stathopoulos, Shukla, Szuecs, Pu, Jones (b32) 2016; 3
Karg, Alamo, Lucia (b21) 2021; 31
Kang, Cao, Word, Laird (b20) 2014; 71
Bonzanini, Paulson, Makrygiorgos, Mesbah (b7) 2021; 145
Cao, Fuentes-Cortes, Chen, Zavala (b10) 2017; 99
Calafiore, Dabbene, Tempo (b9) 2011; 47
Domahidi, Zeilinger, Morari, Jones (b15) 2011
Karg, Lucia (b22) 2020; 50
Hertneck, Köhler, Trimpe, Allgöwer (b16) 2018; 2
Wächter, Biegler (b36) 2006; 106
Bemporad, Borrelli, Morari (b3) 2002; 47
Hornik, Stinchcombe, White (b17) 1990; 3
Bemporad, Oliveri, Poggi, Storace (b5) 2011; 56
Zavala, Laird, Biegler (b38) 2008; 63
Jalving, Cao, Zavala (b18) 2019; 125
Chan, Paulson, Mesbah (b13) 2021
Bemporad, Morari, Dua, Pistikopoulos (b4) 2002; 38
Birge, Louveaux (b6) 2011
Rodriguez, Nicholson, Laird, Zavala (b30) 2018; 119
Boyd, Parikh, Chu (b8) 2011
Yoo, Kim, Kim, Lee (b37) 2021; 144
Alamo, Tempo, Luque, Ramirez (b2) 2015; 52
Tempo, Bai, Dabbene (b33) 1996
Cao, Laird, Zavala (b12) 2016; 64
Karg, Lucia (b23) 2021
Paulson, Mesbah (b28) 2018; 51
Safran, Shamir (b31) 2017
Von Luxburg, Schölkopf (b35) 2011
Alamo, Tempo, Camacho (b1) 2009; 54
Kumar, Tulsyan, Gopaluni, Loewen (b24) 2018; 51
Chiang, Petra, Zavala (b14) 2014
Raff, Huber, Nagy, Allgower (b29) 2006
Alamo (10.1016/j.automatica.2022.110665_b2) 2015; 52
Karg (10.1016/j.automatica.2022.110665_b23) 2021
Wächter (10.1016/j.automatica.2022.110665_b36) 2006; 106
Patrinos (10.1016/j.automatica.2022.110665_b27) 2013; 59
Maddalena (10.1016/j.automatica.2022.110665_b25) 2020; 53
Paulson (10.1016/j.automatica.2022.110665_b28) 2018; 51
Bemporad (10.1016/j.automatica.2022.110665_b5) 2011; 56
Chiang (10.1016/j.automatica.2022.110665_b14) 2014
Vaupel (10.1016/j.automatica.2022.110665_b34) 2020; 92
Hertneck (10.1016/j.automatica.2022.110665_b16) 2018; 2
Raff (10.1016/j.automatica.2022.110665_b29) 2006
Cao (10.1016/j.automatica.2022.110665_b12) 2016; 64
Tempo (10.1016/j.automatica.2022.110665_b33) 1996
Domahidi (10.1016/j.automatica.2022.110665_b15) 2011
Bemporad (10.1016/j.automatica.2022.110665_b4) 2002; 38
Karg (10.1016/j.automatica.2022.110665_b22) 2020; 50
Von Luxburg (10.1016/j.automatica.2022.110665_b35) 2011
Bonzanini (10.1016/j.automatica.2022.110665_b7) 2021; 145
Yoo (10.1016/j.automatica.2022.110665_b37) 2021; 144
Cao (10.1016/j.automatica.2022.110665_b11) 2020; 53
Birge (10.1016/j.automatica.2022.110665_b6) 2011
Cao (10.1016/j.automatica.2022.110665_b10) 2017; 99
Alamo (10.1016/j.automatica.2022.110665_b1) 2009; 54
Zavala (10.1016/j.automatica.2022.110665_b38) 2008; 63
Stathopoulos (10.1016/j.automatica.2022.110665_b32) 2016; 3
Parisini (10.1016/j.automatica.2022.110665_b26) 1995; 31
Kumar (10.1016/j.automatica.2022.110665_b24) 2018; 51
Chan (10.1016/j.automatica.2022.110665_b13) 2021
Jalving (10.1016/j.automatica.2022.110665_b18) 2019; 125
Calafiore (10.1016/j.automatica.2022.110665_b9) 2011; 47
Kang (10.1016/j.automatica.2022.110665_b20) 2014; 71
Boyd (10.1016/j.automatica.2022.110665_b8) 2011
Bemporad (10.1016/j.automatica.2022.110665_b3) 2002; 47
Jerez (10.1016/j.automatica.2022.110665_b19) 2014; 59
Hornik (10.1016/j.automatica.2022.110665_b17) 1990; 3
Rodriguez (10.1016/j.automatica.2022.110665_b30) 2018; 119
Zhang (10.1016/j.automatica.2022.110665_b39) 2020; 29
Safran (10.1016/j.automatica.2022.110665_b31) 2017
Karg (10.1016/j.automatica.2022.110665_b21) 2021; 31
References_xml – start-page: 651
  year: 2011
  end-page: 706
  ident: b35
  article-title: Statistical learning theory: Models, concepts, and results
  publication-title: Handbook of the history of logic, Vol. 10
– year: 2011
  ident: b8
  article-title: Distributed optimization and statistical learning via the alternating direction method of multipliers
– volume: 53
  start-page: 11362
  year: 2020
  end-page: 11367
  ident: b25
  article-title: A neural network architecture to learn explicit MPC controllers from data
  publication-title: IFAC-PapersOnLine
– volume: 47
  start-page: 1974
  year: 2002
  end-page: 1985
  ident: b3
  article-title: Model predictive control based on linear programming — the explicit solution
  publication-title: IEEE Transactions on Automatic Control
– volume: 47
  start-page: 1279
  year: 2011
  end-page: 1293
  ident: b9
  article-title: Research on probabilistic methods for control system design
  publication-title: Automatica
– volume: 54
  start-page: 2545
  year: 2009
  end-page: 2559
  ident: b1
  article-title: Randomized strategies for probabilistic solutions of uncertain feasibility and optimization problems
  publication-title: IEEE Transactions on Automatic Control
– volume: 64
  start-page: 379
  year: 2016
  end-page: 406
  ident: b12
  article-title: Clustering-based preconditioning for stochastic programs
  publication-title: Computational Optimization and Applications
– volume: 145
  year: 2021
  ident: b7
  article-title: Fast approximate learning-based multistage nonlinear model predictive control using Gaussian processes and deep neural networks
  publication-title: Computers & Chemical Engineering
– start-page: 3475
  year: 2021
  end-page: 3481
  ident: b13
  article-title: Deep learning-based approximate nonlinear model predictive control with offset-free tracking for embedded applications
  publication-title: 2021 American control conference (ACC)
– volume: 59
  start-page: 3238
  year: 2014
  end-page: 3251
  ident: b19
  article-title: Embedded online optimization for model predictive control at megahertz rates
  publication-title: IEEE Transactions on Automatic Control
– start-page: 149
  year: 2021
  end-page: 156
  ident: b23
  article-title: Reinforced approximate robust nonlinear model predictive control
  publication-title: 2021 23rd international conference on process control (PC)
– volume: 59
  start-page: 18
  year: 2013
  end-page: 33
  ident: b27
  article-title: An accelerated dual gradient-projection algorithm for embedded linear model predictive control
  publication-title: IEEE Transactions on Automatic Control
– volume: 63
  start-page: 4834
  year: 2008
  end-page: 4845
  ident: b38
  article-title: Interior-point decomposition approaches for parallel solution of large-scale nonlinear parameter estimation problems
  publication-title: Chemical Engineering Science
– volume: 125
  start-page: 134
  year: 2019
  end-page: 154
  ident: b18
  article-title: Graph-based modeling and simulation of complex systems
  publication-title: Computers & Chemical Engineering
– volume: 2
  start-page: 543
  year: 2018
  end-page: 548
  ident: b16
  article-title: Learning an approximate model predictive controller with guarantees
  publication-title: IEEE Control Systems Letters
– volume: 99
  start-page: 185
  year: 2017
  end-page: 197
  ident: b10
  article-title: Scalable modeling and solution of stochastic multiobjective optimization problems
  publication-title: Computers & Chemical Engineering
– volume: 31
  start-page: 8855
  year: 2021
  end-page: 8876
  ident: b21
  article-title: Probabilistic performance validation of deep learning-based robust NMPC controllers
  publication-title: International Journal of Robust and Nonlinear Control
– start-page: 3424
  year: 1996
  end-page: 3428
  ident: b33
  article-title: Probabilistic robustness analysis: Explicit bounds for the minimum number of samples
  publication-title: Proceedings of 35th IEEE conference on decision and control, Vol. 3
– start-page: 2979
  year: 2017
  end-page: 2987
  ident: b31
  article-title: Depth-width tradeoffs in approximating natural functions with neural networks
  publication-title: International conference on machine learning
– volume: 29
  start-page: 2102
  year: 2020
  end-page: 2114
  ident: b39
  article-title: Near-optimal rapid MPC using neural networks: A primal-dual policy learning framework
  publication-title: IEEE Transactions on Control Systems Technology
– start-page: 513
  year: 2011
  end-page: 519
  ident: b15
  article-title: Learning a feasible and stabilizing explicit model predictive control law by robust optimization
  publication-title: 2011 50th IEEE conference on decision and control and European control conference
– volume: 51
  start-page: 523
  year: 2018
  end-page: 534
  ident: b28
  article-title: Nonlinear model predictive control with explicit backoffs for stochastic systems under arbitrary uncertainty
  publication-title: IFAC-PapersOnLine
– volume: 144
  start-page: 107133
  year: 2021
  ident: b37
  article-title: Reinforcement learning based optimal control of batch processes using Monte-Carlo deep deterministic policy gradient with phase segmentation
  publication-title: Computers & Chemical Engineering
– volume: 51
  start-page: 512
  year: 2018
  end-page: 517
  ident: b24
  article-title: A deep learning architecture for predictive control
  publication-title: IFAC-PapersOnLine
– volume: 119
  start-page: 315
  year: 2018
  end-page: 325
  ident: b30
  article-title: Benchmarking ADMM in nonconvex NLPs
  publication-title: Computers & Chemical Engineering
– year: 2011
  ident: b6
  article-title: Introduction to stochastic programming
– volume: 106
  start-page: 25
  year: 2006
  end-page: 57
  ident: b36
  article-title: On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming
  publication-title: Mathematical Programming
– volume: 92
  start-page: 261
  year: 2020
  end-page: 270
  ident: b34
  article-title: Accelerating nonlinear model predictive control through machine learning
  publication-title: Journal of Process Control
– volume: 52
  start-page: 160
  year: 2015
  end-page: 172
  ident: b2
  article-title: Randomized methods for design of uncertain systems: Sample complexity and sequential algorithms
  publication-title: Automatica
– volume: 56
  start-page: 2883
  year: 2011
  end-page: 2897
  ident: b5
  article-title: Ultra-fast stabilizing model predictive control via canonical piecewise affine approximations
  publication-title: IEEE Transactions on Automatic Control
– volume: 3
  start-page: 551
  year: 1990
  end-page: 560
  ident: b17
  article-title: Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks
  publication-title: Neural Networks
– volume: 71
  start-page: 563
  year: 2014
  end-page: 573
  ident: b20
  article-title: An interior-point method for efficient solution of block-structured NLP problems using an implicit Schur-complement decomposition
  publication-title: Computers & Chemical Engineering
– volume: 31
  start-page: 1443
  year: 1995
  end-page: 1451
  ident: b26
  article-title: A receding-horizon regulator for nonlinear systems and a neural approximation
  publication-title: Automatica
– volume: 53
  start-page: 11319
  year: 2020
  end-page: 11324
  ident: b11
  article-title: Deep neural network approximation of nonlinear model predictive control
  publication-title: IFAC-PapersOnLine
– start-page: 1
  year: 2014
  end-page: 7
  ident: b14
  article-title: Structured nonconvex optimization of large-scale energy systems using PIPS-NLP
  publication-title: Power systems computation conference (PSCC)
– volume: 3
  start-page: 249
  year: 2016
  end-page: 362
  ident: b32
  article-title: Operator splitting methods in control
  publication-title: Foundations and Trends in Systems and Control
– volume: 38
  start-page: 3
  year: 2002
  end-page: 20
  ident: b4
  article-title: The explicit linear quadratic regulator for constrained systems
  publication-title: Automatica
– volume: 50
  start-page: 3866
  year: 2020
  end-page: 3878
  ident: b22
  article-title: Efficient representation and approximation of model predictive control laws via deep learning
  publication-title: IEEE Transactions on Cybernetics
– start-page: 237
  year: 2006
  end-page: 242
  ident: b29
  article-title: Nonlinear model predictive control of a four tank system: An experimental stability study
  publication-title: 2006 IEEE conference on computer aided control system design, 2006 IEEE international conference on control applications, 2006 IEEE international symposium on intelligent control
– volume: 51
  start-page: 512
  issue: 18
  year: 2018
  ident: 10.1016/j.automatica.2022.110665_b24
  article-title: A deep learning architecture for predictive control
  publication-title: IFAC-PapersOnLine
  doi: 10.1016/j.ifacol.2018.09.373
– start-page: 3424
  year: 1996
  ident: 10.1016/j.automatica.2022.110665_b33
  article-title: Probabilistic robustness analysis: Explicit bounds for the minimum number of samples
– year: 2011
  ident: 10.1016/j.automatica.2022.110665_b8
– start-page: 3475
  year: 2021
  ident: 10.1016/j.automatica.2022.110665_b13
  article-title: Deep learning-based approximate nonlinear model predictive control with offset-free tracking for embedded applications
– volume: 31
  start-page: 8855
  issue: 18
  year: 2021
  ident: 10.1016/j.automatica.2022.110665_b21
  article-title: Probabilistic performance validation of deep learning-based robust NMPC controllers
  publication-title: International Journal of Robust and Nonlinear Control
  doi: 10.1002/rnc.5696
– start-page: 149
  year: 2021
  ident: 10.1016/j.automatica.2022.110665_b23
  article-title: Reinforced approximate robust nonlinear model predictive control
– volume: 99
  start-page: 185
  year: 2017
  ident: 10.1016/j.automatica.2022.110665_b10
  article-title: Scalable modeling and solution of stochastic multiobjective optimization problems
  publication-title: Computers & Chemical Engineering
  doi: 10.1016/j.compchemeng.2017.01.021
– start-page: 513
  year: 2011
  ident: 10.1016/j.automatica.2022.110665_b15
  article-title: Learning a feasible and stabilizing explicit model predictive control law by robust optimization
– volume: 38
  start-page: 3
  issue: 1
  year: 2002
  ident: 10.1016/j.automatica.2022.110665_b4
  article-title: The explicit linear quadratic regulator for constrained systems
  publication-title: Automatica
  doi: 10.1016/S0005-1098(01)00174-1
– volume: 3
  start-page: 551
  issue: 5
  year: 1990
  ident: 10.1016/j.automatica.2022.110665_b17
  article-title: Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks
  publication-title: Neural Networks
  doi: 10.1016/0893-6080(90)90005-6
– volume: 59
  start-page: 3238
  issue: 12
  year: 2014
  ident: 10.1016/j.automatica.2022.110665_b19
  article-title: Embedded online optimization for model predictive control at megahertz rates
  publication-title: IEEE Transactions on Automatic Control
  doi: 10.1109/TAC.2014.2351991
– volume: 71
  start-page: 563
  year: 2014
  ident: 10.1016/j.automatica.2022.110665_b20
  article-title: An interior-point method for efficient solution of block-structured NLP problems using an implicit Schur-complement decomposition
  publication-title: Computers & Chemical Engineering
  doi: 10.1016/j.compchemeng.2014.09.013
– volume: 51
  start-page: 523
  issue: 20
  year: 2018
  ident: 10.1016/j.automatica.2022.110665_b28
  article-title: Nonlinear model predictive control with explicit backoffs for stochastic systems under arbitrary uncertainty
  publication-title: IFAC-PapersOnLine
  doi: 10.1016/j.ifacol.2018.11.036
– volume: 50
  start-page: 3866
  issue: 9
  year: 2020
  ident: 10.1016/j.automatica.2022.110665_b22
  article-title: Efficient representation and approximation of model predictive control laws via deep learning
  publication-title: IEEE Transactions on Cybernetics
  doi: 10.1109/TCYB.2020.2999556
– volume: 53
  start-page: 11362
  issue: 2
  year: 2020
  ident: 10.1016/j.automatica.2022.110665_b25
  article-title: A neural network architecture to learn explicit MPC controllers from data
  publication-title: IFAC-PapersOnLine
  doi: 10.1016/j.ifacol.2020.12.546
– volume: 119
  start-page: 315
  year: 2018
  ident: 10.1016/j.automatica.2022.110665_b30
  article-title: Benchmarking ADMM in nonconvex NLPs
  publication-title: Computers & Chemical Engineering
  doi: 10.1016/j.compchemeng.2018.08.036
– start-page: 2979
  year: 2017
  ident: 10.1016/j.automatica.2022.110665_b31
  article-title: Depth-width tradeoffs in approximating natural functions with neural networks
– volume: 52
  start-page: 160
  year: 2015
  ident: 10.1016/j.automatica.2022.110665_b2
  article-title: Randomized methods for design of uncertain systems: Sample complexity and sequential algorithms
  publication-title: Automatica
  doi: 10.1016/j.automatica.2014.11.004
– volume: 2
  start-page: 543
  issue: 3
  year: 2018
  ident: 10.1016/j.automatica.2022.110665_b16
  article-title: Learning an approximate model predictive controller with guarantees
  publication-title: IEEE Control Systems Letters
  doi: 10.1109/LCSYS.2018.2843682
– volume: 47
  start-page: 1974
  issue: 12
  year: 2002
  ident: 10.1016/j.automatica.2022.110665_b3
  article-title: Model predictive control based on linear programming — the explicit solution
  publication-title: IEEE Transactions on Automatic Control
  doi: 10.1109/TAC.2002.805688
– volume: 59
  start-page: 18
  issue: 1
  year: 2013
  ident: 10.1016/j.automatica.2022.110665_b27
  article-title: An accelerated dual gradient-projection algorithm for embedded linear model predictive control
  publication-title: IEEE Transactions on Automatic Control
  doi: 10.1109/TAC.2013.2275667
– volume: 3
  start-page: 249
  issue: 3
  year: 2016
  ident: 10.1016/j.automatica.2022.110665_b32
  article-title: Operator splitting methods in control
  publication-title: Foundations and Trends in Systems and Control
  doi: 10.1561/2600000008
– start-page: 1
  year: 2014
  ident: 10.1016/j.automatica.2022.110665_b14
  article-title: Structured nonconvex optimization of large-scale energy systems using PIPS-NLP
– volume: 29
  start-page: 2102
  issue: 5
  year: 2020
  ident: 10.1016/j.automatica.2022.110665_b39
  article-title: Near-optimal rapid MPC using neural networks: A primal-dual policy learning framework
  publication-title: IEEE Transactions on Control Systems Technology
  doi: 10.1109/TCST.2020.3024571
– volume: 125
  start-page: 134
  year: 2019
  ident: 10.1016/j.automatica.2022.110665_b18
  article-title: Graph-based modeling and simulation of complex systems
  publication-title: Computers & Chemical Engineering
  doi: 10.1016/j.compchemeng.2019.03.009
– volume: 31
  start-page: 1443
  issue: 10
  year: 1995
  ident: 10.1016/j.automatica.2022.110665_b26
  article-title: A receding-horizon regulator for nonlinear systems and a neural approximation
  publication-title: Automatica
  doi: 10.1016/0005-1098(95)00044-W
– volume: 53
  start-page: 11319
  issue: 2
  year: 2020
  ident: 10.1016/j.automatica.2022.110665_b11
  article-title: Deep neural network approximation of nonlinear model predictive control
  publication-title: IFAC-PapersOnLine
  doi: 10.1016/j.ifacol.2020.12.538
– year: 2011
  ident: 10.1016/j.automatica.2022.110665_b6
– volume: 145
  year: 2021
  ident: 10.1016/j.automatica.2022.110665_b7
  article-title: Fast approximate learning-based multistage nonlinear model predictive control using Gaussian processes and deep neural networks
  publication-title: Computers & Chemical Engineering
  doi: 10.1016/j.compchemeng.2020.107174
– start-page: 651
  year: 2011
  ident: 10.1016/j.automatica.2022.110665_b35
  article-title: Statistical learning theory: Models, concepts, and results
– volume: 54
  start-page: 2545
  issue: 11
  year: 2009
  ident: 10.1016/j.automatica.2022.110665_b1
  article-title: Randomized strategies for probabilistic solutions of uncertain feasibility and optimization problems
  publication-title: IEEE Transactions on Automatic Control
  doi: 10.1109/TAC.2009.2031207
– start-page: 237
  year: 2006
  ident: 10.1016/j.automatica.2022.110665_b29
  article-title: Nonlinear model predictive control of a four tank system: An experimental stability study
– volume: 92
  start-page: 261
  year: 2020
  ident: 10.1016/j.automatica.2022.110665_b34
  article-title: Accelerating nonlinear model predictive control through machine learning
  publication-title: Journal of Process Control
  doi: 10.1016/j.jprocont.2020.06.012
– volume: 64
  start-page: 379
  year: 2016
  ident: 10.1016/j.automatica.2022.110665_b12
  article-title: Clustering-based preconditioning for stochastic programs
  publication-title: Computational Optimization and Applications
  doi: 10.1007/s10589-015-9813-x
– volume: 106
  start-page: 25
  issue: 1
  year: 2006
  ident: 10.1016/j.automatica.2022.110665_b36
  article-title: On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming
  publication-title: Mathematical Programming
  doi: 10.1007/s10107-004-0559-y
– volume: 144
  start-page: 107133
  year: 2021
  ident: 10.1016/j.automatica.2022.110665_b37
  article-title: Reinforcement learning based optimal control of batch processes using Monte-Carlo deep deterministic policy gradient with phase segmentation
  publication-title: Computers & Chemical Engineering
  doi: 10.1016/j.compchemeng.2020.107133
– volume: 56
  start-page: 2883
  issue: 12
  year: 2011
  ident: 10.1016/j.automatica.2022.110665_b5
  article-title: Ultra-fast stabilizing model predictive control via canonical piecewise affine approximations
  publication-title: IEEE Transactions on Automatic Control
  doi: 10.1109/TAC.2011.2141410
– volume: 47
  start-page: 1279
  issue: 7
  year: 2011
  ident: 10.1016/j.automatica.2022.110665_b9
  article-title: Research on probabilistic methods for control system design
  publication-title: Automatica
  doi: 10.1016/j.automatica.2011.02.029
– volume: 63
  start-page: 4834
  issue: 19
  year: 2008
  ident: 10.1016/j.automatica.2022.110665_b38
  article-title: Interior-point decomposition approaches for parallel solution of large-scale nonlinear parameter estimation problems
  publication-title: Chemical Engineering Science
  doi: 10.1016/j.ces.2007.05.022
SSID ssj0004182
Score 2.539714
Snippet To facilitate the real-time implementation of nonlinear model predictive control (NMPC), this paper proposes a deep learning-based NMPC scheme, in which the...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 110665
SubjectTerms Deep neural networks
Model predictive control
Nonlinear systems
Parallel computation
Policy learning
Stochastic optimization
Title Using stochastic programming to train neural network approximation of nonlinear MPC laws
URI https://dx.doi.org/10.1016/j.automatica.2022.110665
Volume 146
WOSCitedRecordID wos000878605100002&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: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1873-2836
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0004182
  issn: 0005-1098
  databaseCode: AIEXJ
  dateStart: 19950101
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1bS8MwFA66-aAP4hXnjTz4Nio23ZoWn8ZQVEQEJ8ynkqUNbmonrpP-fE9y0nZeQEWE0pZAljXn6-nX9JzvEHLgAklmqsUc4Lax09JnQdz2HAXPHqY_NHEpTbEJfnUV9Pvhta3ZOjHlBHiaBnkePv-rqaENjK1TZ39h7vJHoQHOweiwB7PD_keGxyAA4HTyXmgR5iIE68kkRo2xKERT61iCdVKMAkdp8Xz4VBLIFCU0tMrPdbf5KCz3LvRqp9nYaL0KI1eaY4R8uaZwaUIE7qZpBRqbeTbMq8auMMu0dyJ9EMPZ1QfGPkRylGkxVQwSulktb4rlpQ8T9KwB9xzgMv4714vLj5_cOK4ojHQQj72aQz24zlnwsbTEB5HsG8NFj7QYIX67nCd1xtsh-Ll65_ykf1HlyroBKsjbv2ijuzDm7-vxvqYsMzSkt0KW7fsD7aDdV8lckq6RpRlVyXXSNwigFQLoDAJoNqYGARQRQC0C6DsE0LGiJQIoIIBqBGyQ29OTXvfMsQU0HAmuOoPbgskYCCUcVcAUUGM-OGIDLoTrep6CJ5uIJfhsJeCtVgJtaYUqjLkcuF478duut0lqMFiyRaiuH8Jizw9icOCCJcKTXPqKJ4KHsLUahBdzFEmrLq-v5jEqwghHUTW7kZ7dCGe3Qdyy5zMqrPygz3FhhsgyRWSAESDo297bf-q9QxarG2GX1LKXabJHFuRrNpy87Fu4vQHcqZMA
linkProvider Elsevier
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=Using+stochastic+programming+to+train+neural+network+approximation+of+nonlinear+MPC+laws&rft.jtitle=Automatica+%28Oxford%29&rft.au=Li%2C+Yun&rft.au=Hua%2C+Kaixun&rft.au=Cao%2C+Yankai&rft.date=2022-12-01&rft.pub=Elsevier+Ltd&rft.issn=0005-1098&rft.eissn=1873-2836&rft.volume=146&rft_id=info:doi/10.1016%2Fj.automatica.2022.110665&rft.externalDocID=S0005109822005295
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0005-1098&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0005-1098&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0005-1098&client=summon