Evaluation of Anomaly Detection of an Autoencoder Based on Maintenace Information and Scada-Data

Marc-Alexander Lutz, Stephan Vogt, Volker Berkhout, Stefan Faulstich, Steffen Dienst, Urs Steinmetz, Christian Gück and Andres Ortega
Additional contact information
Marc-Alexander Lutz: Fraunhofer Institute for Energy Economics and Energy System Technology, Königstor 59, 34119 Kassel, Germany
Stephan Vogt: Fraunhofer Institute for Energy Economics and Energy System Technology, Königstor 59, 34119 Kassel, Germany
Volker Berkhout: Fraunhofer Institute for Energy Economics and Energy System Technology, Königstor 59, 34119 Kassel, Germany
Stefan Faulstich: Fraunhofer Institute for Energy Economics and Energy System Technology, Königstor 59, 34119 Kassel, Germany
Steffen Dienst: Trianel Windpark Borkum GmbH und Co. KG, Zirkusweg 2, 20359 Hamburg, Germany
Urs Steinmetz: STEAG Energy Services GmbH, Rüttenscheider Str. 1-3, 45128 Essen, Germany
Christian Gück: Fraunhofer Institute for Energy Economics and Energy System Technology, Königstor 59, 34119 Kassel, Germany
Andres Ortega: Fraunhofer Institute for Energy Economics and Energy System Technology, Königstor 59, 34119 Kassel, Germany

Energies, 2020, vol. 13, issue 5, 1-18

Abstract: The usage of machine learning techniques is widely spread and has also been implemented in the wind industry in the last years. Many of these techniques have shown great success but need to constantly prove the expectation of functionality. This paper describes a new method to monitor the health of a wind turbine using an undercomplete autoencoder. To evaluate the health monitoring quality of the autoencoder, the number of anomalies before an event has happened are to be considered. The results show that around 35% of all historical events that have resulted into a failure show many anomalies. Furthermore, the wind turbine subsystems which are subject to good detectability are the rotor system and the control system. If only one third of the service duties can be planned in advance, and thereby the scheduling time can be reduced, huge cost saving potentials can be seen.

Keywords: wind turbine; maintenance; autoencoder; machine learning; reliability; data driven model; service; performance (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://www.mdpi.com/1996-1073/13/5/1063/pdf (application/pdf)
https://www.mdpi.com/1996-1073/13/5/1063/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:13:y:2020:i:5:p:1063-:d:326419

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().