Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Analysis of SCADA data for early fault detection with application to the maintenance management of wind turbines
KTH, Skolan för elektroteknik och datavetenskap (EECS), Elektroteknisk teori och konstruktion.ORCID-id: 0000-0003-4763-9429
2016 (Engelska)Ingår i: CIGRE Session 46, CIGRE , 2016, s. 1-10Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

During the past decade wind turbines have proven to be a promising source of renewable power. Wind turbines are generally placed in remote locations and are subject to harsh environmental conditions throughout their lifetimes. Consequently, the failures in wind turbines are expensive to repair and cause loss of revenue due to long down times. Asset management in wind turbines can aid in assessing and improving the reliability and availability of wind turbines, thereby making them more competitive. Maintenance policies play an important role in asset management and different maintenance models have been developed for wind turbine applications. Typically, mathematical models for maintenance optimization provide either an age based or a condition based preventive maintenance schedule. Age based preventive maintenance schedules provide the owner with the possibility to financially plan for maintenance activities for the entire lifetime of the wind turbine by providing the expected number of replacements for each component. However, age based preventive maintenance schedule may not consume the operating life of the wind turbine components to the maximum. Condition based maintenance scheduling has the advantage of better utilizing the operating life of the components. This paper proposes a wind turbine maintenance management framework which utilizes operation and maintenance data from different sources to combine the benefits of age based and condition based maintenance scheduling. This paper also presents an artificial neural network (ANN) based condition monitoring method which utilizes data from supervisory control and data acquisition (SCADA) system to detect failures in wind turbine components and systems. The procedures to construct ANN models for condition monitoring application are outlined. In order to demonstrate the effectiveness of the ANN based condition monitoring method it is applied to case studies from real wind turbines. Furthermore, a mathematical model called preventive maintenance schedule with interval costs (PMSPIC) is discussed and its application to a case study within the maintenance management framework is presented. The case study demonstrates the advantage of combining both the age based and condition based maintenance scheduling methods. 

Ort, förlag, år, upplaga, sidor
CIGRE , 2016. s. 1-10
Nyckelord [en]
Artificial neural network (ANN), Asset management, Condition monitoring system (CMS), Data acquisition (SCADA), Maintenance scheduling, Mathematical model, Supervisory control, Wind turbine, Condition monitoring, Fault detection, Information management, Mathematical models, Neural networks, SCADA systems, Scheduling, Turbine components, Wind turbines, Condition monitoring systems, Condition-based maintenance scheduling, Environmental conditions, Operation and maintenance, Reliability and availability, Supervisory Control and Data Acquisition (SCADA) systems, Preventive maintenance
Nationell ämneskategori
Elektroteknik och elektronik
Identifikatorer
URN: urn:nbn:se:kth:diva-236886ISI: 000413615500047Scopus ID: 2-s2.0-85048744376OAI: oai:DiVA.org:kth-236886DiVA, id: diva2:1270168
Konferens
CIGRE Session 46, 21 August 2016 through 26 August 2016
Forskningsfinansiär
Vetenskapsrådet
Anmärkning

QC 20181212

Tillgänglig från: 2018-12-12 Skapad: 2018-12-12 Senast uppdaterad: 2018-12-12Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas i DiVA

Scopus

Personposter BETA

Bertling, Lina

Sök vidare i DiVA

Av författaren/redaktören
Bertling, Lina
Av organisationen
Elektroteknisk teori och konstruktion
Elektroteknik och elektronik

Sök vidare utanför DiVA

GoogleGoogle Scholar

urn-nbn

Altmetricpoäng

urn-nbn
Totalt: 281 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf