Physics-Informed Graph Capsule Generative Autoencoder for Probabilistic AC Optimal Power Flow

Due to the increasing demand for electricity and the inherent uncertainty in power generation, finding efficient solutions to the stochastic alternating current optimal power flow (AC-OPF) problem has become crucial. However, the nonlinear and non-convex nature of AC-OPF, coupled with the growing st...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE transactions on emerging topics in computational intelligence Ročník 8; číslo 5; s. 3382 - 3395
Hlavní autoři:	Saffari, Mohsen, Khodayar, Mahdi, Khodayar, Mohammad E.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	IEEE 01.10.2024
Témata:	AC optimal power flow capsule network dynamic routing Generators graph analysis Heuristic algorithms physics-informed neural network probabilistic estimation Probabilistic logic Reactive power Symmetric matrices Training Transmission line matrix methods
ISSN:	2471-285X, 2471-285X
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	Due to the increasing demand for electricity and the inherent uncertainty in power generation, finding efficient solutions to the stochastic alternating current optimal power flow (AC-OPF) problem has become crucial. However, the nonlinear and non-convex nature of AC-OPF, coupled with the growing stochasticity resulting from the integration of renewable energy sources, presents significant challenges in achieving fast and reliable solutions. To address these challenges, this study proposes a novel graph-based generative methodology that effectively captures the uncertainties in power system measurements, enabling the learning of probability distribution functions for generation dispatch and voltage setpoints. Our approach involves modeling the power system as a weighted graph and utilizing a deep spectral graph convolution network to extract powerful spatial patterns from the input graph measurements. A unique variational approach is introduced to identify the most relevant latent features that accurately describe the setpoints of the AC-OPF problem. Additionally, a capsule network with a new greedy dynamic routing algorithm is proposed to precisely decode the latent features and estimate the probabilistic AC-OPF problem. Further, a set of carefully designed physics-informed loss functions is incorporated in the training procedure of the model to ensure adherence to the fundamental physics rules governing power systems. Notably, the proposed physics-informed loss functions not only enhance the accuracy of AC-OPF estimation by effectively regularizing the deep learning model but also significantly reduce the time complexity. Extensive experimental evaluations conducted on various benchmarks demonstrate our proposed model's superiority over both probabilistic and deterministic approaches in terms of relevant criteria.
AbstractList	Due to the increasing demand for electricity and the inherent uncertainty in power generation, finding efficient solutions to the stochastic alternating current optimal power flow (AC-OPF) problem has become crucial. However, the nonlinear and non-convex nature of AC-OPF, coupled with the growing stochasticity resulting from the integration of renewable energy sources, presents significant challenges in achieving fast and reliable solutions. To address these challenges, this study proposes a novel graph-based generative methodology that effectively captures the uncertainties in power system measurements, enabling the learning of probability distribution functions for generation dispatch and voltage setpoints. Our approach involves modeling the power system as a weighted graph and utilizing a deep spectral graph convolution network to extract powerful spatial patterns from the input graph measurements. A unique variational approach is introduced to identify the most relevant latent features that accurately describe the setpoints of the AC-OPF problem. Additionally, a capsule network with a new greedy dynamic routing algorithm is proposed to precisely decode the latent features and estimate the probabilistic AC-OPF problem. Further, a set of carefully designed physics-informed loss functions is incorporated in the training procedure of the model to ensure adherence to the fundamental physics rules governing power systems. Notably, the proposed physics-informed loss functions not only enhance the accuracy of AC-OPF estimation by effectively regularizing the deep learning model but also significantly reduce the time complexity. Extensive experimental evaluations conducted on various benchmarks demonstrate our proposed model's superiority over both probabilistic and deterministic approaches in terms of relevant criteria.
Author	Saffari, Mohsen Khodayar, Mohammad E. Khodayar, Mahdi
Author_xml	– sequence: 1 givenname: Mohsen orcidid: 0000-0002-8336-8542 surname: Saffari fullname: Saffari, Mohsen email: mohsen-saffari@utulsa.edu organization: Department of Computer Science, University of Tulsa, Tulsa, OK, USA – sequence: 2 givenname: Mahdi orcidid: 0000-0003-4683-7810 surname: Khodayar fullname: Khodayar, Mahdi email: mahdi-khodayar@utulsa.edu organization: Department of Computer Science, University of Tulsa, Tulsa, OK, USA – sequence: 3 givenname: Mohammad E. orcidid: 0000-0003-3856-5704 surname: Khodayar fullname: Khodayar, Mohammad E. email: mkhodayar@smu.edu organization: Department of Electrical and Computer Engineering, Southern Methodist University, Dallas, TX, USA
BookMark	eNp9UMFqAjEUDMVCrfUHSg_5gbVJNtlNjiLVCoIeLPRSlmT3LaasmyVZK_59Y_VQeig8eHOYeW9m7tGgdS0g9EjJhFKinrcv29lywgjjkzTN8yynN2jIeE4TJsX74Be-Q-MQPgkhTAmaCj5EH5vdKdgyJMu2dn4PFV543e3wTHfh0ABeQAte9_YL8PTQO2hLV4HHkYs33hltbGNDb0s8neF119u9bvDGHSNl3rjjA7qtdRNgfN0j9DaPZl-T1XqxnE1XScky2ScqF4YLQ5TiKiaADCptuKx0qZmWNGIaR5lMKg2SlLnguaFQm5qLWug0HSF2uVt6F4KHuuh8tOJPBSXFuaPip6Pi3FFx7SiK5B9RafsY1bW917b5X_p0kVoA-PWLS8l5nn4DgER4Ig
CODEN	ITETCU
CitedBy_id	crossref_primary_10_3390_en18184867
Cites_doi	10.1109/SEST50973.2021.9543326 10.1109/TPWRS.2010.2051168 10.1115/1.4050542 10.1109/tnnls.2023.3280078 10.1007/978-3-030-89698-0_55 10.1109/ISGT.2017.8086083 10.1002/9781119480402.ch6 10.1155/2021/1015367 10.1109/TIE.2018.2803732 10.1109/TPWRS.2020.3026379 10.1016/j.ijepes.2021.106908 10.1109/TSG.2019.2895333 10.1080/23789689.2019.1708182 10.1109/TIA.2022.3213008 10.1109/ACCESS.2020.3015473 10.1016/j.epsr.2022.108282 10.1016/j.ijepes.2022.108484 10.1016/j.epsr.2021.107216 10.1016/j.apenergy.2022.119672 10.3390/su11226293 10.1002/etep.2312 10.1561/2200000006 10.1109/OAJPE.2022.3140314 10.1007/s00202-013-0277-7 10.1016/j.cma.2020.113547 10.1007/s00202-021-01223-7 10.1016/j.solener.2019.04.014 10.1109/JSYST.2022.3201041 10.1109/TPWRS.2020.2987292 10.1016/j.ijepes.2020.106492 10.1109/AEEICB.2017.7972430 10.1016/j.epsr.2022.108353 10.1109/TPWRS.2020.3001919 10.1109/TSTE.2018.2847558 10.1109/SmartGridComm47815.2020.9303008 10.1016/j.ejor.2023.01.019 10.1109/ACCESS.2018.2870430 10.1109/TSTE.2019.2927135 10.1109/EEEIC/ICPSEurope51590.2021.9584602 10.1016/j.apenergy.2020.115124 10.1109/MLSP.2019.8918690 10.1109/TPWRS.2018.2848207
ContentType	Journal Article
DBID	97E RIA RIE AAYXX CITATION
DOI	10.1109/TETCI.2024.3377671
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE/IET Electronic Library CrossRef
DatabaseTitle	CrossRef
DatabaseTitleList
Database_xml	– sequence: 1 dbid: RIE name: IEEE/IET Electronic Library url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
EISSN	2471-285X
EndPage	3395
ExternalDocumentID	10_1109_TETCI_2024_3377671 10488447
Genre	orig-research
GroupedDBID	0R~ 97E AAJGR AASAJ AAWTH ABAZT ABJNI ABQJQ ABVLG ACGFS AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD IFIPE JAVBF OCL RIA RIE AAYXX CITATION
ID	FETCH-LOGICAL-c268t-975b45b09949337e6edab48daca2a81ab41b419b689ae80c7547b1efbf45f5a33
IEDL.DBID	RIE
ISICitedReferencesCount	4
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001197902000001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2471-285X
IngestDate	Sat Nov 29 05:12:10 EST 2025 Tue Nov 18 22:23:59 EST 2025 Wed Aug 27 02:18:54 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	5
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-c268t-975b45b09949337e6edab48daca2a81ab41b419b689ae80c7547b1efbf45f5a33
ORCID	0000-0002-8336-8542 0000-0003-4683-7810 0000-0003-3856-5704
PageCount	14
ParticipantIDs	crossref_citationtrail_10_1109_TETCI_2024_3377671 ieee_primary_10488447 crossref_primary_10_1109_TETCI_2024_3377671
PublicationCentury	2000
PublicationDate	2024-10-01
PublicationDateYYYYMMDD	2024-10-01
PublicationDate_xml	– month: 10 year: 2024 text: 2024-10-01 day: 01
PublicationDecade	2020
PublicationTitle	IEEE transactions on emerging topics in computational intelligence
PublicationTitleAbbrev	TETCI
PublicationYear	2024
Publisher	IEEE
Publisher_xml	– name: IEEE
References	ref13 ref35 Kipf (ref38) 2016 ref12 ref34 ref15 ref14 ref31 Chatzos (ref23) 2020 ref30 ref11 ref33 ref10 ref32 ref2 ref1 Miller (ref36) 2010 Knoblauch (ref37) 2022; 23 ref17 ref16 ref19 ref18 ref24 ref46 ref45 ref26 ref48 ref25 ref47 ref20 ref42 ref41 ref22 ref44 Navidi (ref8) 2016 ref21 ref43 Sabour (ref39) 2017 ref28 ref27 ref29 ref7 ref9 ref4 ref3 ref6 ref5 ref40
References_xml	– ident: ref24 doi: 10.1109/SEST50973.2021.9543326 – ident: ref42 doi: 10.1109/TPWRS.2010.2051168 – ident: ref32 doi: 10.1115/1.4050542 – ident: ref31 doi: 10.1109/tnnls.2023.3280078 – ident: ref48 doi: 10.1007/978-3-030-89698-0_55 – ident: ref7 doi: 10.1109/ISGT.2017.8086083 – ident: ref34 doi: 10.1002/9781119480402.ch6 – ident: ref22 doi: 10.1155/2021/1015367 – ident: ref41 doi: 10.1109/TIE.2018.2803732 – volume: 23 start-page: 1 issue: 132 year: 2022 ident: ref37 article-title: An optimization-centric view on Bayes rule: Reviewing and generalizing variational inference publication-title: J. Mach. Learn. Res. – ident: ref11 doi: 10.1109/TPWRS.2020.3026379 – ident: ref13 doi: 10.1016/j.ijepes.2021.106908 – volume-title: Inverse Transform Sampling year: 2010 ident: ref36 – ident: ref35 doi: 10.1109/TSG.2019.2895333 – ident: ref4 doi: 10.1080/23789689.2019.1708182 – ident: ref28 doi: 10.1109/TIA.2022.3213008 – ident: ref19 doi: 10.1109/ACCESS.2020.3015473 – ident: ref14 doi: 10.1016/j.epsr.2022.108282 – ident: ref29 doi: 10.1016/j.ijepes.2022.108484 – ident: ref46 doi: 10.1016/j.epsr.2021.107216 – ident: ref26 doi: 10.1016/j.apenergy.2022.119672 – ident: ref3 doi: 10.3390/su11226293 – ident: ref5 doi: 10.1002/etep.2312 – year: 2016 ident: ref38 article-title: Semi-supervised classification with graph convolutional networks – ident: ref40 doi: 10.1561/2200000006 – ident: ref12 doi: 10.1109/OAJPE.2022.3140314 – ident: ref21 doi: 10.1007/s00202-013-0277-7 – start-page: 3859 volume-title: Proc. Adv. Neural Inf. Process. Syst. year: 2017 ident: ref39 article-title: Dynamic routing between capsules – ident: ref33 doi: 10.1016/j.cma.2020.113547 – ident: ref30 doi: 10.1007/s00202-021-01223-7 – ident: ref43 doi: 10.1016/j.solener.2019.04.014 – ident: ref15 doi: 10.1109/JSYST.2022.3201041 – ident: ref17 doi: 10.1109/TPWRS.2020.2987292 – ident: ref18 doi: 10.1016/j.ijepes.2020.106492 – year: 2016 ident: ref8 article-title: Predicting solutions to the optimal power flow problem – ident: ref20 doi: 10.1109/AEEICB.2017.7972430 – year: 2020 ident: ref23 article-title: High-fidelity machine learning approximations of large-scale optimal power flow – ident: ref10 doi: 10.1016/j.epsr.2022.108353 – ident: ref1 doi: 10.1109/TPWRS.2020.3001919 – ident: ref45 doi: 10.1109/TSTE.2018.2847558 – ident: ref16 doi: 10.1109/SmartGridComm47815.2020.9303008 – ident: ref6 doi: 10.1016/j.ejor.2023.01.019 – ident: ref47 doi: 10.1109/ACCESS.2018.2870430 – ident: ref2 doi: 10.1109/TSTE.2019.2927135 – ident: ref27 doi: 10.1109/EEEIC/ICPSEurope51590.2021.9584602 – ident: ref25 doi: 10.1016/j.apenergy.2020.115124 – ident: ref9 doi: 10.1109/MLSP.2019.8918690 – ident: ref44 doi: 10.1109/TPWRS.2018.2848207
SSID	ssj0002951354
Score	2.2964911
Snippet	Due to the increasing demand for electricity and the inherent uncertainty in power generation, finding efficient solutions to the stochastic alternating...
SourceID	crossref ieee
SourceType	Enrichment Source Index Database Publisher
StartPage	3382
SubjectTerms	AC optimal power flow capsule network dynamic routing Generators graph analysis Heuristic algorithms physics-informed neural network probabilistic estimation Probabilistic logic Reactive power Symmetric matrices Training Transmission line matrix methods
Title	Physics-Informed Graph Capsule Generative Autoencoder for Probabilistic AC Optimal Power Flow
URI	https://ieeexplore.ieee.org/document/10488447
Volume	8
WOSCitedRecordID	wos001197902000001&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/IET Electronic Library customDbUrl: eissn: 2471-285X dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002951354 issn: 2471-285X databaseCode: RIE dateStart: 20170101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS8MwFA9uePDiB078Jgdv0tk0SdMcx3DqZe4wYRcp-XgFYa7Sdfrvm6TdmAcFoYdQXmjzy9d7yXu_h9CNMsQAKWTEDOPeQOGRYhIioBAbASzWRIdkE2I8zmYzOWmD1UMsDAAE5zPo-2K4y7elWfmjMjfD3XBjTHRQR4i0CdbaHKgkTlegnK0DY2J5N72fDp-cCZiwPqWetYb82Hy2sqmEzWR08M_fOET7rdaIB003H6EdWByj1-C9aZZRE1IEFj949mk8VM7ynQNuGKX9coYHq7r0jJUWKuxk8aRy09i7xXqWZjwY4me3cry7L0x80jQ8mpdfPfQyci17jNpsCZFJ0qyOpOCace00PiZdcyEFqzTLrDIqURlxZeIeqdNMKshcR3AmNIFCF4wXXFF6grqLcgGnCCsDhcxSCkRbpilVcZGAUDSx0mhp2Rkiaxhz01KJ-4wW8zyYFLHMA_S5hz5voT9Dt5s6Hw2Rxp_SPY_7lmQD-fkv7y_Qnq_eeNldom5dreAK7ZrP-m1ZXYeB8g3i374a
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS8MwFA86Bb34gYrzMwdv0tk0L2tzHMO54Zw7TNhFSpK-gjA32Yf--yZpN-ZBQeghlNc2-eXrvfS93yPkRhlmkOUyAAPCGSgiUCAxQI6hiRFCzbRPNhH3eslwKPtlsLqPhUFE73yGNVf0__KziVm4ozI7w-1wA4g3yZYAiMIiXGt1pBJZbYELWIbGhPJucD9odqwRGEGNc8dbw35sP2v5VPx20tr_Z0UOyF6pN9JG0dGHZAPHR-TV-2-aWVAEFWFGHxz_NG0qa_uOkBac0m5Bo43FfOI4KzOcUitL-1M7kZ1jrONppo0mfbZrx7v9Qt-lTaOt0eTrmLy0bMvaQZkvITBRPZkHMhYahLY6H0jbXKxjpjQkmTIqUgmzZWYvqeuJVJjYrhAQa4a5zkHkQnF-QirjyRhPCVUGc5nUOTKdgeZchXmEseJRJo2WGVQJW8KYmpJM3OW0GKXeqAhl6qFPHfRpCX2V3K6e-SioNP6UPna4r0kWkJ_9cv-a7LQHT9202-k9npNd96rC5-6CVObTBV6SbfM5f5tNr_yg-Qbw_sFh
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=Physics-Informed+Graph+Capsule+Generative+Autoencoder+for+Probabilistic+AC+Optimal+Power+Flow&rft.jtitle=IEEE+transactions+on+emerging+topics+in+computational+intelligence&rft.au=Saffari%2C+Mohsen&rft.au=Khodayar%2C+Mahdi&rft.au=Khodayar%2C+Mohammad+E.&rft.date=2024-10-01&rft.issn=2471-285X&rft.eissn=2471-285X&rft.volume=8&rft.issue=5&rft.spage=3382&rft.epage=3395&rft_id=info:doi/10.1109%2FTETCI.2024.3377671&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TETCI_2024_3377671
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2471-285X&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2471-285X&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2471-285X&client=summon