Multi-Gradient-Descent Federated Learning With Parity for Cooperative Short-Term Load Forecasting
Short-term load forecasting (STLF) is essential in power system research. In scenarios where multiple distribution system operators are involved in cooperative forecasting, it is important to uphold the fairness of multi-party cooperation and drive the parity of model gains among parties under the p...
Uložené v:
| Vydané v: | IEEE transactions on power systems Ročník 40; číslo 2; s. 1199 - 1213 |
|---|---|
| Hlavní autori: | , , , , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
New York
IEEE
01.03.2025
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Predmet: | |
| ISSN: | 0885-8950, 1558-0679 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | Short-term load forecasting (STLF) is essential in power system research. In scenarios where multiple distribution system operators are involved in cooperative forecasting, it is important to uphold the fairness of multi-party cooperation and drive the parity of model gains among parties under the premise of privacy protection. In this context, this paper proposes a short-term load forecasting method namely Multi-Gradient-Descent Federated Learning with Parity Long-short Term Memory (MGD-FLP-LSTM) under a federated learning framework. First, MGD-FLP-LSTM transforms cooperative STLF into a Multi-Objective Optimization (MOO) problem with parity-driven objective. The designed algorithm optimizes multiple objectives simultaneously and incorporates Karush-Kuhn-Tucker (KKT) conditions in solving the dual form of the problem, thus accelerating the global model convergence. Second, MGD-FLP-LSTM constitutes a parity-driven objective with cosine similarity in the MOO iteration. The objective allows all local loss functions to have similar descents on the common gradient descending direction. Third, MGD-FLP-LSTM constructs a Performance Parity Control (PPC) scheme that enables an active moderate trade-off between forecasting accuracy and parity. Experiments on real-measured load dataset highlight the method's enhanced accuracy, efficiency, and performance while maintaining consumer privacy. |
|---|---|
| AbstractList | Short-term load forecasting (STLF) is essential in power system research. In scenarios where multiple distribution system operators are involved in cooperative forecasting, it is important to uphold the fairness of multi-party cooperation and drive the parity of model gains among parties under the premise of privacy protection. In this context, this paper proposes a short-term load forecasting method namely Multi-Gradient-Descent Federated Learning with Parity Long-short Term Memory (MGD-FLP-LSTM) under a federated learning framework. First, MGD-FLP-LSTM transforms cooperative STLF into a Multi-Objective Optimization (MOO) problem with parity-driven objective. The designed algorithm optimizes multiple objectives simultaneously and incorporates Karush-Kuhn-Tucker (KKT) conditions in solving the dual form of the problem, thus accelerating the global model convergence. Second, MGD-FLP-LSTM constitutes a parity-driven objective with cosine similarity in the MOO iteration. The objective allows all local loss functions to have similar descents on the common gradient descending direction. Third, MGD-FLP-LSTM constructs a Performance Parity Control (PPC) scheme that enables an active moderate trade-off between forecasting accuracy and parity. Experiments on real-measured load dataset highlight the method's enhanced accuracy, efficiency, and performance while maintaining consumer privacy. |
| Author | Pan, Zibin Zhao, Junhua Xing, Shuangshuang Wang, Haijin Dong, Zhaoyang Si, Caomingzhe Qiu, Jing |
| Author_xml | – sequence: 1 givenname: Haijin orcidid: 0000-0002-1913-0505 surname: Wang fullname: Wang, Haijin email: wanghj@csg.cn organization: China Southern Power Grid Company, Ltd., Energy Development Research Institute, CSG, Guangdong, China – sequence: 2 givenname: Shuangshuang surname: Xing fullname: Xing, Shuangshuang email: xingshuangshuang@zj.sgcc.com.cn organization: Tonglu Electric Power Company, State Grid Zhejiang Electric Power Company, Ltd., Hangzhou, China – sequence: 3 givenname: Caomingzhe surname: Si fullname: Si, Caomingzhe email: scmz21@mails.tsinghua.edu.cn organization: Department of Electrical Engineering, Tsinghua University, Beijing, China – sequence: 4 givenname: Zibin surname: Pan fullname: Pan, Zibin email: zibinpan@link.cuhk.edu.cn organization: School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China – sequence: 5 givenname: Junhua orcidid: 0000-0001-5446-2655 surname: Zhao fullname: Zhao, Junhua email: zhaojunhua@cuhk.edu.cn organization: School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China – sequence: 6 givenname: Jing orcidid: 0000-0001-8507-0558 surname: Qiu fullname: Qiu, Jing email: qiujing0322@gmail.com organization: School of Electrical and Information Engineering, The University of Sydney, Sydney, NSW, Australia – sequence: 7 givenname: Zhaoyang orcidid: 0000-0001-9659-0858 surname: Dong fullname: Dong, Zhaoyang email: zy.dong@ntu.edu.sg organization: Department of Electrical Engineering, City University of Hong Kong, Kowloon, Hong Kong |
| BookMark | eNp9kMFKAzEQhoNUsK2-gHgIeN462Ww2yVGqVaGi2IrHJd1MNKXd1Gwq-PZurQfx4Ok_zP_NMN-A9JrQICGnDEaMgb6YP748zUY55MWIFzkoxg5InwmhMiil7pE-KCUypQUckUHbLgGg7AZ9Yu63q-Szm2isxyZlV9jWXdIJWowmoaVTNLHxzSt98emNPpro0yd1IdJxCJtdx38gnb2FmLI5xjWdBmPpJESsTZs67pgcOrNq8eQnh-R5cj0f32bTh5u78eU0q3NdpswCSmlFrpVirpaldJAbzsUCSmOBFeCKUhcGFzUrpVigFbUDaZR1TnGtkQ_J-X7vJob3LbapWoZtbLqTFWelzrkEXnQttW_VMbRtRFfVPnU_hCZF41cVg2ontPoWWu2EVj9COzT_g26iX5v4-T90toc8Iv4ChJRCA_8CuU2EqQ |
| CODEN | ITPSEG |
| CitedBy_id | crossref_primary_10_1016_j_ins_2025_122440 |
| Cites_doi | 10.1287/moor.22.2.301 10.1145/3447548.3470814 10.1016/j.apenergy.2017.03.034 10.1145/3375627.3375840 10.1016/j.eswa.2010.11.033 10.1109/TSG.2020.3031007 10.1016/j.asoc.2019.105616 10.1109/TII.2021.3055283 10.1137/1.9781611976700.21 10.1109/TSG.2021.3125677 10.1561/9781680837896 10.1109/TPWRS.2018.2868167 10.1109/tnnls.2022.3216981 10.1109/TPWRS.2007.907583 10.1016/j.apenergy.2022.119915 10.1109/TSG.2022.3215742 10.1016/j.neucom.2008.10.017 10.17775/CSEEJPES.2021.07350 10.1109/TSG.2023.3253855 10.1080/02331934.2010.522710 10.1049/enc2.12055 10.1109/TSG.2019.2933413 10.1109/TSG.2017.2753802 10.1109/JSYST.2016.2594208 10.1109/tnnls.2023.3263594 |
| 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 7SP 7TB 8FD FR3 KR7 L7M |
| DOI | 10.1109/TPWRS.2024.3420811 |
| DatabaseName | IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Xplore CrossRef Electronics & Communications Abstracts Mechanical & Transportation Engineering Abstracts Technology Research Database Engineering Research Database Civil Engineering Abstracts Advanced Technologies Database with Aerospace |
| DatabaseTitle | CrossRef Civil Engineering Abstracts Engineering Research Database Technology Research Database Mechanical & Transportation Engineering Abstracts Advanced Technologies Database with Aerospace Electronics & Communications Abstracts |
| DatabaseTitleList | Civil Engineering Abstracts |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Xplore url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1558-0679 |
| EndPage | 1213 |
| ExternalDocumentID | 10_1109_TPWRS_2024_3420811 10577590 |
| Genre | orig-research |
| GroupedDBID | -~X .DC 0R~ 29I 3EH 4.4 5GY 5VS 6IK 85S 97E AAJGR AARMG AASAJ AAWTH ABAZT ABFSI ABQJQ ABVLG ACGFO ACGFS ACIWK ACKIV AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD HZ~ H~9 ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P RIA RIE RNS TAE TN5 VH1 VJK AAYXX CITATION 7SP 7TB 8FD FR3 KR7 L7M |
| ID | FETCH-LOGICAL-c296t-d0e77d529881fc767f02a335b06ad0140f4694aebc1675bed5cf07a8dff8399e3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 2 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001468243700014&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0885-8950 |
| IngestDate | Fri Jul 25 12:16:47 EDT 2025 Sat Nov 29 08:18:33 EST 2025 Tue Nov 18 22:13:16 EST 2025 Wed Aug 27 01:52:51 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 2 |
| 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-c296t-d0e77d529881fc767f02a335b06ad0140f4694aebc1675bed5cf07a8dff8399e3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| ORCID | 0000-0002-1913-0505 0000-0001-5446-2655 0000-0001-9659-0858 0000-0001-8507-0558 |
| PQID | 3169237034 |
| PQPubID | 85441 |
| PageCount | 15 |
| ParticipantIDs | ieee_primary_10577590 crossref_citationtrail_10_1109_TPWRS_2024_3420811 proquest_journals_3169237034 crossref_primary_10_1109_TPWRS_2024_3420811 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-03-01 |
| PublicationDateYYYYMMDD | 2025-03-01 |
| PublicationDate_xml | – month: 03 year: 2025 text: 2025-03-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | New York |
| PublicationPlace_xml | – name: New York |
| PublicationTitle | IEEE transactions on power systems |
| PublicationTitleAbbrev | TPWRS |
| 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 ref15 ref14 ref31 ref30 ref11 ref10 ref2 ref1 ref19 Li (ref28) Dieterich (ref12) 2016; 7 ref23 ref26 ref25 McMahan (ref18) ref20 Bagdasaryan (ref24) ref22 ref21 ref27 ref29 ref8 ref7 Li (ref17) ref9 ref4 (ref16) 2023 ref3 ref6 ref5 |
| References_xml | – ident: ref11 doi: 10.1287/moor.22.2.301 – ident: ref29 doi: 10.1145/3447548.3470814 – start-page: 2938 volume-title: Proc. Int. Conf. Artif. Intell. Statist. ident: ref24 article-title: How to backdoor federated learning – ident: ref6 doi: 10.1016/j.apenergy.2017.03.034 – start-page: 1273 volume-title: Proc. 20th Int. Conf. Artif. Intell. Statist. ident: ref18 article-title: Communication-efficient learning of deep networks from decentralized data – volume: 7 start-page: 1 issue: 7 year: 2016 ident: ref12 article-title: COMPAS risk scales: Demonstrating accuracy equity and predictive parity publication-title: Northpointe Inc. – ident: ref26 doi: 10.1145/3375627.3375840 – ident: ref4 doi: 10.1016/j.eswa.2010.11.033 – ident: ref19 doi: 10.1109/TSG.2020.3031007 – ident: ref5 doi: 10.1016/j.asoc.2019.105616 – ident: ref14 doi: 10.1109/TII.2021.3055283 – ident: ref30 doi: 10.1137/1.9781611976700.21 – ident: ref13 doi: 10.1109/TSG.2021.3125677 – ident: ref10 doi: 10.1561/9781680837896 – volume-title: arXiv:1905.10497 ident: ref28 article-title: Fair resource allocation in federated learning – ident: ref25 doi: 10.1109/TPWRS.2018.2868167 – ident: ref31 doi: 10.1109/tnnls.2022.3216981 – ident: ref3 doi: 10.1109/TPWRS.2007.907583 – ident: ref9 doi: 10.1016/j.apenergy.2022.119915 – year: 2023 ident: ref16 article-title: The urban power consumption monitoring project in Eastern China – ident: ref8 doi: 10.1109/TSG.2022.3215742 – ident: ref2 doi: 10.1016/j.neucom.2008.10.017 – start-page: 6357 volume-title: Proc. 38th Int. Conf. Mach. Learn. ident: ref17 article-title: Ditto: Fair and robust federated learning through personalization – ident: ref20 doi: 10.17775/CSEEJPES.2021.07350 – ident: ref21 doi: 10.1109/TSG.2023.3253855 – ident: ref15 doi: 10.1080/02331934.2010.522710 – ident: ref22 doi: 10.1049/enc2.12055 – ident: ref23 doi: 10.1109/TSG.2019.2933413 – ident: ref7 doi: 10.1109/TSG.2017.2753802 – ident: ref1 doi: 10.1109/JSYST.2016.2594208 – ident: ref27 doi: 10.1109/tnnls.2023.3263594 |
| SSID | ssj0006679 |
| Score | 2.4818442 |
| Snippet | Short-term load forecasting (STLF) is essential in power system research. In scenarios where multiple distribution system operators are involved in cooperative... |
| SourceID | proquest crossref ieee |
| SourceType | Aggregation Database Enrichment Source Index Database Publisher |
| StartPage | 1199 |
| SubjectTerms | Accuracy Algorithms Cooperative STLF Data models Federated learning Forecasting Kuhn-Tucker method Load forecasting Load modeling Long short term memory multi-gradient-descent Multiple objective analysis Parity performance parity Privacy privacy-preserving |
| Title | Multi-Gradient-Descent Federated Learning With Parity for Cooperative Short-Term Load Forecasting |
| URI | https://ieeexplore.ieee.org/document/10577590 https://www.proquest.com/docview/3169237034 |
| Volume | 40 |
| WOSCitedRecordID | wos001468243700014&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: 1558-0679 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0006679 issn: 0885-8950 databaseCode: RIE dateStart: 19860101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NS8QwEA0qHvTgt7i6Sg7eJLtJP_JxFHX1IMuiK3oraTJRQbayW_39JmlXFFHwVmgCpS-dvGlm3kPo2GbGiNJSUhqXkExZSSQHQxinVga-AC42Cl-L4VA-PKhR26wee2EAIBafQS9cxrN8W5m38KusHzxpRa58hr4oBG-atT7DLueNsJ6UOZEqp_MOGar649H9za3PBZOsl4bjZMa-7ULRVuVHLI4bzGD9n4-2gdZaJolPG-g30QJMttDqF33BbaRjey25nMa6rpqcN9pNeBAUJDzJtLiVV33E98_1Ex7pYGWHPY_FZ1X1Co0qOL598hydjH0Mx9eVtjjYeRo9CwXTO-hucDE-uyKtpwIxieI1sRSEsHmipGTOCC4cTXSa5iXl2oZsy_l8OdNQGuZTiRJsbhwVWlrnPJVSkO6ipUk1gT2EtdIGNON5Clmmrb8rwDmdsmBGDyLtIDZ_x4VpBceD78VLERMPqoqISxFwKVpcOujkc85rI7fx5-idgMSXkQ0IHdSdY1m0n-SsSBn3ZNYHuGz_l2kHaCUJ7r6xwqyLlurpGxyiZfNeP8-mR3G1fQCcl9La |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1dSxwxFA1ihdaHalvFba3Ng28lu8l85OOxWFel02XRLfoWMsmNCrIju6O_3yQzK5bSQt8GJoFhTubm3Mm95yB06AprRe0oqa3PSKGcJJKDJYxTJyNfAJ8ahSsxmcirKzXtm9VTLwwApOIzGMbLdJbvGvsQf5WNoietKFXI0F9F66y-Xes58HLeSetJWRKpSrrqkaFqNJtenl-EbDArhnk8UGbst30oGav8EY3TFjPe-s-H20Zvey6Jv3Xgv0NrMH-PNl8oDH5AJjXYkpNFquxqyfdOvQmPo4ZEoJkO9wKr1_jytr3BUxPN7HBgsvioae6h0wXHFzeBpZNZiOK4aozD0dDTmmUsmd5Bv8bHs6NT0rsqEJsp3hJHQQhXZkpK5q3gwtPM5HlZU25czLd8yJgLA7VlIZmowZXWU2Gk8z6QKQX5LlqfN3PYQ9goY8EwXuZQFMaFuwK8NzmLdvQg8gFiq3esbS85Hp0v7nRKPajSCRcdcdE9LgP09XnOfSe48c_ROxGJFyM7EAZof4Wl7j_Kpc4ZD3Q2hLji41-mfUGvT2c_K12dTX58Qm-y6PWb6s320Xq7eIDPaMM-trfLxUFaeU_CS9Yj |
| 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=Multi-Gradient-Descent+Federated+Learning+With+Parity+for+Cooperative+Short-Term+Load+Forecasting&rft.jtitle=IEEE+transactions+on+power+systems&rft.au=Wang%2C+Haijin&rft.au=Xing%2C+Shuangshuang&rft.au=Si%2C+Caomingzhe&rft.au=Pan%2C+Zibin&rft.date=2025-03-01&rft.issn=0885-8950&rft.eissn=1558-0679&rft.volume=40&rft.issue=2&rft.spage=1199&rft.epage=1213&rft_id=info:doi/10.1109%2FTPWRS.2024.3420811&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TPWRS_2024_3420811 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0885-8950&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0885-8950&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0885-8950&client=summon |