Comparison of anomaly detection techniques for wind turbine gearbox SCADA data

Research output: Contribution to conferenceAbstract

36 Downloads (Pure)

Abstract

This analysis looks at the use of anomaly detection to assess the condition of wind turbine gearboxes based on data from a number of operational turbines. A comparison is made between various methods of anomaly detection, these being one class support vector machine (OCSVM), random forests, and nonlinear autoregressive neural networks with exogenous inputs (NARX).
Original languageEnglish
Number of pages1
Publication statusPublished - 17 Jun 2019
EventWind Energy Science conference 2019 - University College Cork, Cork, Ireland
Duration: 17 Jun 201920 Jun 2019
Conference number: 2019
https://www.wesc2019.org/

Conference

ConferenceWind Energy Science conference 2019
Abbreviated titleWESC
CountryIreland
CityCork
Period17/06/1920/06/19
Internet address

Keywords

  • anomaly detection
  • operations and maintenance (O&M)
  • wind turbines
  • one class support vector machine
  • neural networks with exogenous inputs
  • wind energy

Fingerprint Dive into the research topics of 'Comparison of anomaly detection techniques for wind turbine gearbox SCADA data'. Together they form a unique fingerprint.

  • Cite this

    Mckinnon, C., Carroll, J., McDonald, A., Koukoura, S., & Soraghan, C. (2019). Comparison of anomaly detection techniques for wind turbine gearbox SCADA data. Abstract from Wind Energy Science conference 2019, Cork, Ireland.