SCADA-based wind turbine anomaly detection using Gaussian Process (GP) models for wind turbine condition monitoring purposes

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

The penetration of wind energy into power systems is steadily increasing; this highlights the importance of operations and maintenance, and also specifically the role of condition monitoring. Wind turbine power curves based on SCADA data provide a cost-effective approach to wind turbine health monitoring.
This paper proposes a Gaussian Process (a non-parametric machine learning approach) based algorithm for condition monitoring. The standard IEC binned power curve together with individual bin probability distributions can be used to identify operational anomalies. The IEC approach can also be modified to create a form of real-time power curve. Both of these approaches will be compared with a Gaussian Process model to assess both speed and accuracy of anomaly detection. Significant yaw misalignment, reflecting a yaw control error or fault, results in a loss of power. Such a fault is quite common and early detection is important to prevent loss of power generation. Yaw control error provides a useful case study to demonstrate the effectiveness of the proposed algorithms and allows the advantages and limitations of the proposed methods to be determined.
LanguageEnglish
Number of pages9
JournalIET Renewable Power Generation
Early online date31 May 2018
DOIs
Publication statusE-pub ahead of print - 31 May 2018

Fingerprint

Condition monitoring
Wind turbines
Bins
Wind power
Probability distributions
Power generation
Learning systems
Health
Monitoring
Costs

Keywords

  • wind energy
  • wind turbine
  • condition monitoring and fault diagnosis system
  • Gaussian Process models
  • SCADA analysis

Cite this

@article{bc7c31758030421187e96ad0b56c4470,
title = "SCADA-based wind turbine anomaly detection using Gaussian Process (GP) models for wind turbine condition monitoring purposes",
abstract = "The penetration of wind energy into power systems is steadily increasing; this highlights the importance of operations and maintenance, and also specifically the role of condition monitoring. Wind turbine power curves based on SCADA data provide a cost-effective approach to wind turbine health monitoring.This paper proposes a Gaussian Process (a non-parametric machine learning approach) based algorithm for condition monitoring. The standard IEC binned power curve together with individual bin probability distributions can be used to identify operational anomalies. The IEC approach can also be modified to create a form of real-time power curve. Both of these approaches will be compared with a Gaussian Process model to assess both speed and accuracy of anomaly detection. Significant yaw misalignment, reflecting a yaw control error or fault, results in a loss of power. Such a fault is quite common and early detection is important to prevent loss of power generation. Yaw control error provides a useful case study to demonstrate the effectiveness of the proposed algorithms and allows the advantages and limitations of the proposed methods to be determined.",
keywords = "wind energy, wind turbine, condition monitoring and fault diagnosis system, Gaussian Process models, SCADA analysis",
author = "Pandit, {Ravi Kumar} and David Infield",
note = "This paper is a postprint of a paper submitted to and accepted for publication in IET Renewable Power Generation and is subject to Institution of Engineering and Technology Copyright. The copy of record is available at IET Digital Library.",
year = "2018",
month = "5",
day = "31",
doi = "10.1049/iet-rpg.2018.0156",
language = "English",
journal = "IET Renewable Power Generation",
issn = "1752-1416",
publisher = "Institution of Engineering and Technology",

}

TY - JOUR

T1 - SCADA-based wind turbine anomaly detection using Gaussian Process (GP) models for wind turbine condition monitoring purposes

AU - Pandit, Ravi Kumar

AU - Infield, David

N1 - This paper is a postprint of a paper submitted to and accepted for publication in IET Renewable Power Generation and is subject to Institution of Engineering and Technology Copyright. The copy of record is available at IET Digital Library.

PY - 2018/5/31

Y1 - 2018/5/31

N2 - The penetration of wind energy into power systems is steadily increasing; this highlights the importance of operations and maintenance, and also specifically the role of condition monitoring. Wind turbine power curves based on SCADA data provide a cost-effective approach to wind turbine health monitoring.This paper proposes a Gaussian Process (a non-parametric machine learning approach) based algorithm for condition monitoring. The standard IEC binned power curve together with individual bin probability distributions can be used to identify operational anomalies. The IEC approach can also be modified to create a form of real-time power curve. Both of these approaches will be compared with a Gaussian Process model to assess both speed and accuracy of anomaly detection. Significant yaw misalignment, reflecting a yaw control error or fault, results in a loss of power. Such a fault is quite common and early detection is important to prevent loss of power generation. Yaw control error provides a useful case study to demonstrate the effectiveness of the proposed algorithms and allows the advantages and limitations of the proposed methods to be determined.

AB - The penetration of wind energy into power systems is steadily increasing; this highlights the importance of operations and maintenance, and also specifically the role of condition monitoring. Wind turbine power curves based on SCADA data provide a cost-effective approach to wind turbine health monitoring.This paper proposes a Gaussian Process (a non-parametric machine learning approach) based algorithm for condition monitoring. The standard IEC binned power curve together with individual bin probability distributions can be used to identify operational anomalies. The IEC approach can also be modified to create a form of real-time power curve. Both of these approaches will be compared with a Gaussian Process model to assess both speed and accuracy of anomaly detection. Significant yaw misalignment, reflecting a yaw control error or fault, results in a loss of power. Such a fault is quite common and early detection is important to prevent loss of power generation. Yaw control error provides a useful case study to demonstrate the effectiveness of the proposed algorithms and allows the advantages and limitations of the proposed methods to be determined.

KW - wind energy

KW - wind turbine

KW - condition monitoring and fault diagnosis system

KW - Gaussian Process models

KW - SCADA analysis

UR - http://digital-library.theiet.org/content/journals/iet-rpg

U2 - 10.1049/iet-rpg.2018.0156

DO - 10.1049/iet-rpg.2018.0156

M3 - Article

JO - IET Renewable Power Generation

T2 - IET Renewable Power Generation

JF - IET Renewable Power Generation

SN - 1752-1416

ER -