Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data

Anomaly detection for wind turbine condition monitoring is an active area of research within the wind energy operations and maintenance (O & M) community. In this paper three models were compared for multi-megawatt operational wind turbine SCADA data. The models used for comparison were One-Clas...

Full description

Bibliographic Details
Main Authors:	Conor McKinnon, James Carroll, Alasdair McDonald, Sofia Koukoura, David Infield, Conaill Soraghan
Format:	Article
Language:	English
Published:	MDPI AG 2020-10-01
Series:	Energies
Subjects:	anomaly detection gearbox SCADA condition monitoring Isolation Forest One Class Support Vector Machine
Online Access:	https://www.mdpi.com/1996-1073/13/19/5152

id	doaj-08d76c7201d249008a8d7bbd5446527a
record_format	Article
spelling	doaj-08d76c7201d249008a8d7bbd5446527a2020-11-25T03:42:58ZengMDPI AGEnergies1996-10732020-10-01135152515210.3390/en13195152Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA DataConor McKinnon0James Carroll1Alasdair McDonald2Sofia Koukoura3David Infield4Conaill Soraghan5Wind and Marine Energy Systems CDT, University of Strathclyde, Glasgow G1 1XQ, UKWind and Marine Energy Systems CDT, University of Strathclyde, Glasgow G1 1XQ, UKWind and Marine Energy Systems CDT, University of Strathclyde, Glasgow G1 1XQ, UKWind and Marine Energy Systems CDT, University of Strathclyde, Glasgow G1 1XQ, UKWind and Marine Energy Systems CDT, University of Strathclyde, Glasgow G1 1XQ, UKOffshore Renewable Energy Catapult, Glasgow G1 1XQ, UKAnomaly detection for wind turbine condition monitoring is an active area of research within the wind energy operations and maintenance (O & M) community. In this paper three models were compared for multi-megawatt operational wind turbine SCADA data. The models used for comparison were One-Class Support Vector Machine (OCSVM), Isolation Forest (IF), and Elliptical Envelope (EE). Each of these were compared for the same fault, and tested under various different data configurations. IF and EE have not previously been used for fault detection for wind turbines, and OCSVM has not been used for SCADA data. This paper presents a novel method of condition monitoring that only requires two months of data per turbine. These months were separated by a year, the first being healthy and the second unhealthy. The number of anomalies is compared, with a greater number in the unhealthy month being considered correct. It was found that for accuracy IF and OCSVM had similar performances in both training regimes presented. OCSVM performed better for generic training, and IF performed better for specific training. Overall, IF and OCSVM had an average accuracy of 82% for all configurations considered, compared to 77% for EE.https://www.mdpi.com/1996-1073/13/19/5152anomaly detectiongearboxSCADAcondition monitoringIsolation ForestOne Class Support Vector Machine
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Conor McKinnon James Carroll Alasdair McDonald Sofia Koukoura David Infield Conaill Soraghan
spellingShingle	Conor McKinnon James Carroll Alasdair McDonald Sofia Koukoura David Infield Conaill Soraghan Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data Energies anomaly detection gearbox SCADA condition monitoring Isolation Forest One Class Support Vector Machine
author_facet	Conor McKinnon James Carroll Alasdair McDonald Sofia Koukoura David Infield Conaill Soraghan
author_sort	Conor McKinnon
title	Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data
title_short	Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data
title_full	Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data
title_fullStr	Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data
title_full_unstemmed	Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data
title_sort	comparison of new anomaly detection technique for wind turbine condition monitoring using gearbox scada data
publisher	MDPI AG
series	Energies
issn	1996-1073
publishDate	2020-10-01
description	Anomaly detection for wind turbine condition monitoring is an active area of research within the wind energy operations and maintenance (O & M) community. In this paper three models were compared for multi-megawatt operational wind turbine SCADA data. The models used for comparison were One-Class Support Vector Machine (OCSVM), Isolation Forest (IF), and Elliptical Envelope (EE). Each of these were compared for the same fault, and tested under various different data configurations. IF and EE have not previously been used for fault detection for wind turbines, and OCSVM has not been used for SCADA data. This paper presents a novel method of condition monitoring that only requires two months of data per turbine. These months were separated by a year, the first being healthy and the second unhealthy. The number of anomalies is compared, with a greater number in the unhealthy month being considered correct. It was found that for accuracy IF and OCSVM had similar performances in both training regimes presented. OCSVM performed better for generic training, and IF performed better for specific training. Overall, IF and OCSVM had an average accuracy of 82% for all configurations considered, compared to 77% for EE.
topic	anomaly detection gearbox SCADA condition monitoring Isolation Forest One Class Support Vector Machine
url	https://www.mdpi.com/1996-1073/13/19/5152
work_keys_str_mv	AT conormckinnon comparisonofnewanomalydetectiontechniqueforwindturbineconditionmonitoringusinggearboxscadadata AT jamescarroll comparisonofnewanomalydetectiontechniqueforwindturbineconditionmonitoringusinggearboxscadadata AT alasdairmcdonald comparisonofnewanomalydetectiontechniqueforwindturbineconditionmonitoringusinggearboxscadadata AT sofiakoukoura comparisonofnewanomalydetectiontechniqueforwindturbineconditionmonitoringusinggearboxscadadata AT davidinfield comparisonofnewanomalydetectiontechniqueforwindturbineconditionmonitoringusinggearboxscadadata AT conaillsoraghan comparisonofnewanomalydetectiontechniqueforwindturbineconditionmonitoringusinggearboxscadadata
_version_	1724522303567429632

Comparison of New Anomaly Detection Technique for Wind Turbine Condition Monitoring Using Gearbox SCADA Data

Similar Items