Data-driven online monitoring of wind turbines

Thomas Kenbeek, Stella Kapodistria, Alessandro Di Bucchianico

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

2 Citations (Scopus)
2 Downloads (Pure)

Abstract

Condition based maintenance is a modern approach to maintenance which has been successfully used in several industrial sectors. A specific problem in wind turbine maintenance is that failures of certain parts may be caused by the malperformance or failure of other parts. This mandates for approaches that can produce timely warnings by combining sensor data from different sources. More concretely, in this paper, we present a hybrid statistical approach to condition based maintenance by combining regression analysis with tools from statistical process control. Our approach improves the wind turbine maintenance practice by using adaptive alarm thresholds for the monitored parameters, whilst correcting for environmental factors or for other relevant parameters. We illustrate our approach with a case study demonstrating that we are able to predict upcoming failures much earlier than the current practice.

Original languageEnglish
Title of host publicationProceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools, VALUETOOLS 2019
PublisherAssociation for Computing Machinery, Inc
Pages143-150
Number of pages8
ISBN (Electronic)9781450365963
DOIs
Publication statusPublished - 12 Mar 2019
Event12th EAI International Conference on Performance Evaluation Methodologies and Tools, (VALUETOOLS 2019) - Palma de Mallorca, Spain
Duration: 13 Mar 201915 Mar 2019

Publication series

NameACM International Conference Proceeding Series

Conference

Conference12th EAI International Conference on Performance Evaluation Methodologies and Tools, (VALUETOOLS 2019)
Abbreviated titleVALUETOOLS2019
Country/TerritorySpain
CityPalma de Mallorca
Period13/03/1915/03/19

Keywords

  • Condition based monitoring
  • Control charts
  • Regression analysis
  • Wind turbine

Fingerprint

Dive into the research topics of 'Data-driven online monitoring of wind turbines'. Together they form a unique fingerprint.

Cite this