SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics

Accurately modelling the nonlinear dynamics of a system from measurement data is a challenging yet vital topic. The sparse identification of nonlinear dynamics (SINDy) algorithm is one approach to discover dynamical systems models from data. Although extensions have been developed to identify implic...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences Mathematical, physical, and engineering sciences, 2020-10, Vol.476 (2242), p.1-25
Hauptverfasser: Kaheman, Kadierdan, Kutz, J. Nathan, Brunton, Steven L.
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 25
container_issue 2242
container_start_page 1
container_title Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences
container_volume 476
creator Kaheman, Kadierdan
Kutz, J. Nathan
Brunton, Steven L.
description Accurately modelling the nonlinear dynamics of a system from measurement data is a challenging yet vital topic. The sparse identification of nonlinear dynamics (SINDy) algorithm is one approach to discover dynamical systems models from data. Although extensions have been developed to identify implicit dynamics, or dynamics described by rational functions, these extensions are extremely sensitive to noise. In this work, we develop SINDy-PI (parallel, implicit), a robust variant of the SINDy algorithm to identify implicit dynamics and rational nonlinearities. The SINDy-PI framework includes multiple optimization algorithms and a principled approach to model selection. We demonstrate the ability of this algorithm to learn implicit ordinary and partial differential equations and conservation laws from limited and noisy data. In particular, we show that the proposed approach is several orders of magnitude more noise robust than previous approaches, and may be used to identify a class of ODE and PDE dynamics that were previously unattainable with SINDy, including for the double pendulum dynamics and simplified model for the Belousov–Zhabotinsky (BZ) reaction.
doi_str_mv 10.1098/rspa.2020.0279
format Article
fullrecord <record><control><sourceid>jstor_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7655768</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><jstor_id>27097307</jstor_id><sourcerecordid>27097307</sourcerecordid><originalsourceid>FETCH-LOGICAL-c363t-44cd679b389f9a8b7f3dec09a6822d2057737ab0f85719c582c85fc8369033dc3</originalsourceid><addsrcrecordid>eNpVkE1Lw0AQhhdRbK1evXn3kjjZye7sXgSpX4WignpeNpuNpqRN3U2F_nsTWgqeZmCe9x14GLvMIM1Aq5sQ1zblwCEFTvqIjbOcsoTrXB73O8o8EcCzETuLcQEAWig6ZSNE3nMSxmz0Pnu53yZvs3N2Utkm-ov9nLDPx4eP6XMyf32aTe_miUOJXZLnrpSkC1S60lYVVGHpHWgrFeclB0GEZAuolKBMO6G4U6JyCqUGxNLhhN3uetebYulL51ddsI1Zh3ppw9a0tjb_L6v623y1v4akECRVX3C9Lwjtz8bHzizr6HzT2JVvN9HwXGJvRgvZo-kOdaGNMfjq8CYDM-gzgz4z6DODvj5wtQssYteGA80JNCEQ_gFkVmkG</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2463098956</pqid></control><display><type>article</type><title>SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics</title><source>JSTOR Archive Collection A-Z Listing</source><source>Alma/SFX Local Collection</source><source>JSTOR Mathematics &amp; Statistics</source><creator>Kaheman, Kadierdan ; Kutz, J. Nathan ; Brunton, Steven L.</creator><creatorcontrib>Kaheman, Kadierdan ; Kutz, J. Nathan ; Brunton, Steven L.</creatorcontrib><description>Accurately modelling the nonlinear dynamics of a system from measurement data is a challenging yet vital topic. The sparse identification of nonlinear dynamics (SINDy) algorithm is one approach to discover dynamical systems models from data. Although extensions have been developed to identify implicit dynamics, or dynamics described by rational functions, these extensions are extremely sensitive to noise. In this work, we develop SINDy-PI (parallel, implicit), a robust variant of the SINDy algorithm to identify implicit dynamics and rational nonlinearities. The SINDy-PI framework includes multiple optimization algorithms and a principled approach to model selection. We demonstrate the ability of this algorithm to learn implicit ordinary and partial differential equations and conservation laws from limited and noisy data. In particular, we show that the proposed approach is several orders of magnitude more noise robust than previous approaches, and may be used to identify a class of ODE and PDE dynamics that were previously unattainable with SINDy, including for the double pendulum dynamics and simplified model for the Belousov–Zhabotinsky (BZ) reaction.</description><identifier>ISSN: 1364-5021</identifier><identifier>EISSN: 1471-2946</identifier><identifier>DOI: 10.1098/rspa.2020.0279</identifier><identifier>PMID: 33214760</identifier><language>eng</language><publisher>Royal Society</publisher><ispartof>Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences, 2020-10, Vol.476 (2242), p.1-25</ispartof><rights>2020 The Authors</rights><rights>2020 The Authors. 2020</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c363t-44cd679b389f9a8b7f3dec09a6822d2057737ab0f85719c582c85fc8369033dc3</citedby><cites>FETCH-LOGICAL-c363t-44cd679b389f9a8b7f3dec09a6822d2057737ab0f85719c582c85fc8369033dc3</cites><orcidid>0000-0003-2279-2793</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.jstor.org/stable/pdf/27097307$$EPDF$$P50$$Gjstor$$H</linktopdf><linktohtml>$$Uhttps://www.jstor.org/stable/27097307$$EHTML$$P50$$Gjstor$$H</linktohtml><link.rule.ids>230,314,776,780,799,828,881,27901,27902,57992,57996,58225,58229</link.rule.ids></links><search><creatorcontrib>Kaheman, Kadierdan</creatorcontrib><creatorcontrib>Kutz, J. Nathan</creatorcontrib><creatorcontrib>Brunton, Steven L.</creatorcontrib><title>SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics</title><title>Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences</title><description>Accurately modelling the nonlinear dynamics of a system from measurement data is a challenging yet vital topic. The sparse identification of nonlinear dynamics (SINDy) algorithm is one approach to discover dynamical systems models from data. Although extensions have been developed to identify implicit dynamics, or dynamics described by rational functions, these extensions are extremely sensitive to noise. In this work, we develop SINDy-PI (parallel, implicit), a robust variant of the SINDy algorithm to identify implicit dynamics and rational nonlinearities. The SINDy-PI framework includes multiple optimization algorithms and a principled approach to model selection. We demonstrate the ability of this algorithm to learn implicit ordinary and partial differential equations and conservation laws from limited and noisy data. In particular, we show that the proposed approach is several orders of magnitude more noise robust than previous approaches, and may be used to identify a class of ODE and PDE dynamics that were previously unattainable with SINDy, including for the double pendulum dynamics and simplified model for the Belousov–Zhabotinsky (BZ) reaction.</description><issn>1364-5021</issn><issn>1471-2946</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><recordid>eNpVkE1Lw0AQhhdRbK1evXn3kjjZye7sXgSpX4WignpeNpuNpqRN3U2F_nsTWgqeZmCe9x14GLvMIM1Aq5sQ1zblwCEFTvqIjbOcsoTrXB73O8o8EcCzETuLcQEAWig6ZSNE3nMSxmz0Pnu53yZvs3N2Utkm-ov9nLDPx4eP6XMyf32aTe_miUOJXZLnrpSkC1S60lYVVGHpHWgrFeclB0GEZAuolKBMO6G4U6JyCqUGxNLhhN3uetebYulL51ddsI1Zh3ppw9a0tjb_L6v623y1v4akECRVX3C9Lwjtz8bHzizr6HzT2JVvN9HwXGJvRgvZo-kOdaGNMfjq8CYDM-gzgz4z6DODvj5wtQssYteGA80JNCEQ_gFkVmkG</recordid><startdate>20201001</startdate><enddate>20201001</enddate><creator>Kaheman, Kadierdan</creator><creator>Kutz, J. Nathan</creator><creator>Brunton, Steven L.</creator><general>Royal Society</general><general>The Royal Society Publishing</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0003-2279-2793</orcidid></search><sort><creationdate>20201001</creationdate><title>SINDy-PI</title><author>Kaheman, Kadierdan ; Kutz, J. Nathan ; Brunton, Steven L.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c363t-44cd679b389f9a8b7f3dec09a6822d2057737ab0f85719c582c85fc8369033dc3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Kaheman, Kadierdan</creatorcontrib><creatorcontrib>Kutz, J. Nathan</creatorcontrib><creatorcontrib>Brunton, Steven L.</creatorcontrib><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kaheman, Kadierdan</au><au>Kutz, J. Nathan</au><au>Brunton, Steven L.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics</atitle><jtitle>Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences</jtitle><date>2020-10-01</date><risdate>2020</risdate><volume>476</volume><issue>2242</issue><spage>1</spage><epage>25</epage><pages>1-25</pages><issn>1364-5021</issn><eissn>1471-2946</eissn><abstract>Accurately modelling the nonlinear dynamics of a system from measurement data is a challenging yet vital topic. The sparse identification of nonlinear dynamics (SINDy) algorithm is one approach to discover dynamical systems models from data. Although extensions have been developed to identify implicit dynamics, or dynamics described by rational functions, these extensions are extremely sensitive to noise. In this work, we develop SINDy-PI (parallel, implicit), a robust variant of the SINDy algorithm to identify implicit dynamics and rational nonlinearities. The SINDy-PI framework includes multiple optimization algorithms and a principled approach to model selection. We demonstrate the ability of this algorithm to learn implicit ordinary and partial differential equations and conservation laws from limited and noisy data. In particular, we show that the proposed approach is several orders of magnitude more noise robust than previous approaches, and may be used to identify a class of ODE and PDE dynamics that were previously unattainable with SINDy, including for the double pendulum dynamics and simplified model for the Belousov–Zhabotinsky (BZ) reaction.</abstract><pub>Royal Society</pub><pmid>33214760</pmid><doi>10.1098/rspa.2020.0279</doi><tpages>25</tpages><orcidid>https://orcid.org/0000-0003-2279-2793</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1364-5021
ispartof Proceedings of the Royal Society. A, Mathematical, physical, and engineering sciences, 2020-10, Vol.476 (2242), p.1-25
issn 1364-5021
1471-2946
language eng
recordid cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_7655768
source JSTOR Archive Collection A-Z Listing; Alma/SFX Local Collection; JSTOR Mathematics & Statistics
title SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-08T04%3A41%3A10IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-jstor_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=SINDy-PI:%20a%20robust%20algorithm%20for%20parallel%20implicit%20sparse%20identification%20of%20nonlinear%20dynamics&rft.jtitle=Proceedings%20of%20the%20Royal%20Society.%20A,%20Mathematical,%20physical,%20and%20engineering%20sciences&rft.au=Kaheman,%20Kadierdan&rft.date=2020-10-01&rft.volume=476&rft.issue=2242&rft.spage=1&rft.epage=25&rft.pages=1-25&rft.issn=1364-5021&rft.eissn=1471-2946&rft_id=info:doi/10.1098/rspa.2020.0279&rft_dat=%3Cjstor_pubme%3E27097307%3C/jstor_pubme%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2463098956&rft_id=info:pmid/33214760&rft_jstor_id=27097307&rfr_iscdi=true