Identification of Thermal Model Parameters Using Deep Learning Techniques

dc.contributor.authorŠevčík, Jakub
dc.contributor.authorŠmídl, Václav
dc.contributor.authorVotava, Martin
dc.date.accessioned2023-02-06T11:00:18Z
dc.date.available2023-02-06T11:00:18Z
dc.date.issued2022
dc.description.abstract-translatedIdentification of thermal model parameters using multi-step prediction is proposed. Even in the case of a linear model, the multi-step prediction is a non-linear complex function, hence we use techniques of deep learning for its identification. Specifically, we use stochastic gradient descent optimization with importance sampling of mini-batches. The importance function is designed to match the character of thermal experiments in which the step change is less frequent than steady-state operation. The proposed method is demonstrated on the identification of an IGBT module SK 20 DGDL 065 ET. The maximum error of the model identified by the multi-step approach is almost two times smaller than that of the model identified by the least squares.en
dc.format4 s.cs
dc.format.mimetypeapplication/pdf
dc.identifier.citationŠEVČÍK, J. ŠMÍDL, V. VOTAVA, M. Identification of Thermal Model Parameters Using Deep Learning Techniques. In 2022 IEEE 31st International Symposium on Industrial Electronics (ISIE) : /proceedings/. Piscataway: IEEE, 2022. s. 978-981. ISBN: 978-1-66548-240-0 , ISSN: 2163-5145cs
dc.identifier.doi10.1109/ISIE51582.2022.9831641
dc.identifier.isbn978-1-66548-240-0
dc.identifier.issn2163-5145
dc.identifier.obd43936479
dc.identifier.uri2-s2.0-85135786103
dc.identifier.urihttp://hdl.handle.net/11025/51295
dc.language.isoenen
dc.project.IDEF18_069/0009855/Elektrotechnické technologie s vysokým podílem vestavěné inteligencecs
dc.project.IDSGS-2021-021/Výzkum a vývoj perspektivních technologií v elektrických pohonech a strojích IVcs
dc.publisherIEEEen
dc.relation.ispartofseries2022 IEEE 31st International Symposium on Industrial Electronics (ISIE) : /proceedings/en
dc.rightsPlný text je přístupný v rámci univerzity přihlášeným uživatelům.cs
dc.rights© IEEEen
dc.rights.accessrestrictedAccessen
dc.subject.translateddeep learningen
dc.subject.translatedjunction temperatureen
dc.subject.translatedmultistep predictionen
dc.subject.translatedneural networken
dc.subject.translatedordinary least squaresen
dc.subject.translatedthermal modelen
dc.titleIdentification of Thermal Model Parameters Using Deep Learning Techniquesen
dc.typekonferenční příspěvekcs
dc.typeConferenceObjecten
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen

Files

Original bundle
Showing 1 - 1 out of 1 results
No Thumbnail Available
Name:
Sevcik_Identification.pdf
Size:
1.86 MB
Format:
Adobe Portable Document Format