Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-25296
Full metadata record
DC FieldValueLanguage
dc.contributor.authorZgraggen, Jannik-
dc.contributor.authorPizza, Gianmarco-
dc.contributor.authorGoren Huber, Lilach-
dc.date.accessioned2022-07-14T09:22:44Z-
dc.date.available2022-07-14T09:22:44Z-
dc.date.issued2022-07-04-
dc.identifier.isbn978-1-936263-36-3de_CH
dc.identifier.urihttps://digitalcollection.zhaw.ch/handle/11475/25296-
dc.descriptionBest Paper Award Lizenzangabe: CC BY 3.0 United Statesde_CH
dc.description.abstractQuantifying the predictive uncertainty of a model is an important ingredient in data-driven decision making. Uncertainty quantification has been gaining interest especially for deep learning models, which are often hard to justify or explain. Various techniques for deep learning based uncertainty estimates have been developed primarily for image classification and segmentation, but also for regression and forecasting tasks. Uncertainty quantification for anomaly detection tasks is still rather limited for image data and has not yet been demonstrated for machine fault detection in PHM applications. In this paper we suggest an approach to derive an uncertainty-informed anomaly score for regression models trained with normal data only. The score is derived using a deep ensemble of probabilistic neural networks for uncertainty quantification. Using an example of wind-turbine fault detection, we demonstrate the superiority of the uncertainty-informed anomaly score over the conventional score. The advantage is particularly clear in an "out-of-distribution" scenario, in which the model is trained with limited data which does not represent all normal regimes that are observed during model deployment.de_CH
dc.language.isoende_CH
dc.publisherPHM Societyde_CH
dc.relation.ispartofseriesPHM Society European Conferencede_CH
dc.rightshttp://creativecommons.org/licenses/by/3.0/de_CH
dc.subjectDeep learningde_CH
dc.subjectAnomaly detectionde_CH
dc.subjectPredictive maintenancede_CH
dc.subjectWind turbinesde_CH
dc.subjectUncertainty quantificationde_CH
dc.subjectCondition based maintenancede_CH
dc.subjectMachine learningde_CH
dc.subject.ddc006: Spezielle Computerverfahrende_CH
dc.titleUncertainty informed anomaly scores with deep learning : robust fault detection with limited datade_CH
dc.typeKonferenz: Paperde_CH
dcterms.typeTextde_CH
zhaw.departementSchool of Engineeringde_CH
zhaw.organisationalunitInstitut für Datenanalyse und Prozessdesign (IDP)de_CH
zhaw.publisher.placeState Collegede_CH
dc.identifier.doi10.36001/phme.2022.v7i1.3342de_CH
dc.identifier.doi10.21256/zhaw-25296-
zhaw.conference.details7th European PHM, Turin, Italy, 6-8 July 2022de_CH
zhaw.funding.euNode_CH
zhaw.issue1de_CH
zhaw.originated.zhawYesde_CH
zhaw.pages.end540de_CH
zhaw.pages.start530de_CH
zhaw.parentwork.editorDo, Phuc-
zhaw.parentwork.editorMichau, Gabriel-
zhaw.parentwork.editorEzhilarasu, Cordelia-
zhaw.publication.statuspublishedVersionde_CH
zhaw.series.number7de_CH
zhaw.volume7de_CH
zhaw.publication.reviewPeer review (Publikation)de_CH
zhaw.title.proceedingsProceedings of the 7th European Conference of the Prognostics and Health Management Society 2022de_CH
zhaw.webfeedDatalabde_CH
zhaw.funding.zhawMachine Learning Based Fault Detection for Wind Turbinesde_CH
zhaw.author.additionalNode_CH
zhaw.display.portraitYesde_CH
Appears in collections:Publikationen School of Engineering

Files in This Item:
File Description SizeFormat 
2022_Zgraggen-Pizza-Goren-Huber_Uncertainty-informed_ProPHME22.pdf3.56 MBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.