1. ZHAW digitalcollection
  2. School of Engineering
  3. Publikationen
  4. Publikationen School of Engineering
Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-20824
Full metadata record
DC FieldValueLanguage
dc.contributor.authorUlmer, Markus-
dc.contributor.authorJarlskog, Eskil-
dc.contributor.authorPizza, Gianmarco-
dc.contributor.authorGoren Huber, Lilach-
dc.date.accessioned2020-11-12T13:49:42Z-
dc.date.available2020-11-12T13:49:42Z-
dc.date.issued2020-11-03-
dc.identifier.isbn978-1-936263-33-2de_CH
dc.identifier.urihttps://digitalcollection.zhaw.ch/handle/11475/20824-
dc.description.abstractMachine learning algorithms for early fault detection of wind turbines using 10-minute SCADA data are attracting attention in the wind energy community due to their cost-effectiveness. It has been recently shown that convolutional neural networks (CNNs) can significantly improve the performance of such algorithms. One practical aspect in the deployment of these algorithms is that they require a large amount of historical SCADA data for training. These are not always available, for example in the case of newly installed turbines. Here we suggest a cross-turbine training scheme for CNNs: we train a CNN model on a turbine with abundant data and use the trained network to detect faults in a different wind turbine for which only little data are available. We show that this scheme is able to considerably improve the fault detection performance compared to the scarce data training. Moreover, it is shown to detect faults with an accuracy and robustness which are very similar to the single-turbine scheme, in which training and detection are both done on the same turbine with a large and representative training set. We demonstrate this for two different fault types: abrupt and slowly evolving faults and perform a sensitivity analysis in order to compare the performance of the two training schemes. We show that the cross-turbine scheme works successfully also when training on turbines from another farm and with different measured variables than the target turbine.de_CH
dc.language.isoende_CH
dc.publisherPHM Societyde_CH
dc.rightshttps://creativecommons.org/licenses/by/3.0/de_CH
dc.subjectFault Detectionde_CH
dc.subjectPredictive Maintenancede_CH
dc.subjectDeep Learningde_CH
dc.subjectWind Turbinesde_CH
dc.subjectMachine Learningde_CH
dc.subjectDomain adaptationde_CH
dc.subjectConvolutional Neural Networksde_CH
dc.subjectSCADA datade_CH
dc.subject.ddc620: Ingenieurwesende_CH
dc.titleCross-turbine training of convolutional neural networks for SCADA-based fault detection in wind turbinesde_CH
dc.typeKonferenz: Paperde_CH
dcterms.typeTextde_CH
zhaw.departementSchool of Engineeringde_CH
zhaw.organisationalunitInstitut für Datenanalyse und Prozessdesign (IDP)de_CH
dc.identifier.doi10.36001/phmconf.2020.v12i1.1205de_CH
dc.identifier.doi10.21256/zhaw-20824-
zhaw.conference.details12th Annual Conference of the PHM Society, virtual, 9-13 November 2020de_CH
zhaw.funding.euNode_CH
zhaw.issue1de_CH
zhaw.originated.zhawYesde_CH
zhaw.publication.statuspublishedVersionde_CH
zhaw.volume12de_CH
zhaw.publication.reviewPeer review (Publikation)de_CH
zhaw.title.proceedingsProceedings of the Annual Conference of the PHM Society 2020de_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 
2020_Ulmer_Cross-Turbine_Training_of_Convolutional_Neural_Networks_PHM_Society.pdf977.17 kBAdobe PDFThumbnail
View/Open
Show simple item record
Ulmer, M., Jarlskog, E., Pizza, G., & Goren Huber, L. (2020). Cross-turbine training of convolutional neural networks for SCADA-based fault detection in wind turbines [Conference paper]. Proceedings of the Annual Conference of the PHM Society 2020, 12(1). https://doi.org/10.36001/phmconf.2020.v12i1.1205
Ulmer, M. et al. (2020) ‘Cross-turbine training of convolutional neural networks for SCADA-based fault detection in wind turbines’, in Proceedings of the Annual Conference of the PHM Society 2020. PHM Society. Available at: https://doi.org/10.36001/phmconf.2020.v12i1.1205.
M. Ulmer, E. Jarlskog, G. Pizza, and L. Goren Huber, “Cross-turbine training of convolutional neural networks for SCADA-based fault detection in wind turbines,” in Proceedings of the Annual Conference of the PHM Society 2020, Nov. 2020, vol. 12, no. 1. doi: 10.36001/phmconf.2020.v12i1.1205.
ULMER, Markus, Eskil JARLSKOG, Gianmarco PIZZA und Lilach GOREN HUBER, 2020. Cross-turbine training of convolutional neural networks for SCADA-based fault detection in wind turbines. In: Proceedings of the Annual Conference of the PHM Society 2020. Conference paper. PHM Society. 3 November 2020. ISBN 978-1-936263-33-2
Ulmer, Markus, Eskil Jarlskog, Gianmarco Pizza, and Lilach Goren Huber. 2020. “Cross-Turbine Training of Convolutional Neural Networks for SCADA-Based Fault Detection in Wind Turbines.” Conference paper. In Proceedings of the Annual Conference of the PHM Society 2020. Vol. 12. PHM Society. https://doi.org/10.36001/phmconf.2020.v12i1.1205.
Ulmer, Markus, et al. “Cross-Turbine Training of Convolutional Neural Networks for SCADA-Based Fault Detection in Wind Turbines.” Proceedings of the Annual Conference of the PHM Society 2020, vol. 12, no. 1, PHM Society, 2020, https://doi.org/10.36001/phmconf.2020.v12i1.1205.


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