Overview of normal behavior modeling approaches for SCADA-based wind turbine condition monitoring demonstrated on data from operational wind farms

Chesterman, Xavier; Verstraeten, Timothy; Daems, Pieter-Jan; Nowé, Ann; Helsen, Jan

doi:https://doi.org/10.5194/wes-8-893-2023

Articles | Volume 8, issue 6

https://doi.org/10.5194/wes-8-893-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/wes-8-893-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 8, issue 6

Review article

|

05 Jun 2023

Review article |

| 05 Jun 2023

Overview of normal behavior modeling approaches for SCADA-based wind turbine condition monitoring demonstrated on data from operational wind farms

Xavier Chesterman, Timothy Verstraeten, Pieter-Jan Daems, Ann Nowé, and Jan Helsen

Download

Final revised paper (published on 05 Jun 2023)
Preprint (discussion started on 21 Feb 2023)

Interactive discussion

Status: closed

RC1:
'Comment on wes-2022-120', Anonymous Referee #1, 09 Mar 2023

This is an interesting paper on an important topic. It is well written. I have some critcal comments listed below.
The literature review is reasonable but not complete. Useful research has been published on the application of Copula models and also Gaussian Process models in wind turbine normal behaviour modeling. These should be included in the review.
I'm not clear why the data filtering reduced the normall 10 minute data to hourly. Better results may have been obtained using the 10 minute data directly (after error and gap removal). If noise reduction was the main reason this should be demonstrated statistically.
The mathematics of the modelling and error prediction should be presented in more detail. The terms in equation 2 should be explained.
Cooling transients are a major cause of modelling difficulty. However they do not represent an operational turbine and I'm surprised such data was not excluded from the analysis by using status or power data from the SCADA,

Citation: https://doi.org/10.5194/wes-2022-120-RC1
- AC1: 'Reply on RC1', Xavier Chesterman, 06 Apr 2023
  
  Dear reviewer,
  Thank you for the questions and remarks. You can find our replies in the pdf document attached to this message.
  Yours sincerely
  
  Citation: https://doi.org/10.5194/wes-2022-120-AC1
RC2:
'Comment on wes-2022-120', Anonymous Referee #2, 10 Mar 2023

Please, see pdf file.

Citation: https://doi.org/10.5194/wes-2022-120-RC2
- AC2: 'Reply on RC2', Xavier Chesterman, 06 Apr 2023
  
  Dear reviewer,
  Thank you for your questions and remarks. Please find attached to this message our replies.
  Yours sincerely
  
  Citation: https://doi.org/10.5194/wes-2022-120-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

AR by Xavier Chesterman on behalf of the Authors (21 Apr 2023) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (21 Apr 2023) by Weifei Hu

RR by Anonymous Referee #1 (28 Apr 2023)

RR by Anonymous Referee #2 (29 Apr 2023)

ED: Publish as is (29 Apr 2023) by Weifei Hu

ED: Publish as is (04 May 2023) by Paul Veers (Chief editor)

AR by Xavier Chesterman on behalf of the Authors (08 May 2023)

Short summary

This paper reviews and implements several techniques that can be used for condition monitoring and failure prediction for wind turbines using SCADA data. The focus lies on techniques that respond to requirements of the industry, e.g., robustness, transparency, computational efficiency, and maintainability. The end result of this research is a pipeline that can accurately detect three types of failures, i.e., generator bearing failures, generator fan failures, and generator stator failures.