Diagnosis of tidal turbine vibration data through deep neural networks

Grant S. Galloway, Victoria M. Catterson, Thomas Fay, Andrew Robb, Craig Love

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

Tidal power is an emerging field of renewable energy, harnessing the power of regular and predictable tidal currents. However, maintenance of submerged equipment is a major challenge. Routine visual inspections of equipment must be performed onshore, requiring the costly removal of turbines from the sea bed and resulting in long periods of downtime. The development of condition monitoring techniques providing automated fault detection can therefore be extremely beneficial to this industry, reducing the dependency on inspections and allowing maintenance to be planned efficiently. This paper investigates the use of deep learning approaches for fault detection within a tidal turbine's generator from vibration data. Training and testing data were recorded over two deployment periods of operation from an accelerometer sensor placed within the nacelle of the turbine, representing ideal and faulty responses over a range of operating conditions. The paper evaluates a deep learning approach through a stacked autoencoder network in comparison to feature-based classification methods, where features have been extracted over varying rotation speeds through the Vold-Kalma filter.
LanguageEnglish
Title of host publicationProceedings of the Third European Conference of the Prognostics and Health Management Society 2016
EditorsIoana Eballard, Anibal Bregon
Pages172-180
Number of pages9
Publication statusPublished - 8 Jul 2016
EventThird European Conference of the Prognostics and Health Management Society 2016 - Bilbao, Spain
Duration: 5 Jul 20168 Jul 2016

Conference

ConferenceThird European Conference of the Prognostics and Health Management Society 2016
CountrySpain
CityBilbao
Period5/07/168/07/16

Fingerprint

Fault detection
Turbines
Inspection
Tidal power
Turbogenerators
Condition monitoring
Accelerometers
Sensors
Testing
Industry
Deep learning
Deep neural networks

Keywords

  • renewable energy
  • tidal currents
  • tidal turbines
  • condition monitoring
  • fault detection

Cite this

Galloway, G. S., Catterson, V. M., Fay, T., Robb, A., & Love, C. (2016). Diagnosis of tidal turbine vibration data through deep neural networks. In I. Eballard, & A. Bregon (Eds.), Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016 (pp. 172-180)
Galloway, Grant S. ; Catterson, Victoria M. ; Fay, Thomas ; Robb, Andrew ; Love, Craig. / Diagnosis of tidal turbine vibration data through deep neural networks. Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016. editor / Ioana Eballard ; Anibal Bregon. 2016. pp. 172-180
@inproceedings{09f369e185d646f09cbd6165b221dce7,
title = "Diagnosis of tidal turbine vibration data through deep neural networks",
abstract = "Tidal power is an emerging field of renewable energy, harnessing the power of regular and predictable tidal currents. However, maintenance of submerged equipment is a major challenge. Routine visual inspections of equipment must be performed onshore, requiring the costly removal of turbines from the sea bed and resulting in long periods of downtime. The development of condition monitoring techniques providing automated fault detection can therefore be extremely beneficial to this industry, reducing the dependency on inspections and allowing maintenance to be planned efficiently. This paper investigates the use of deep learning approaches for fault detection within a tidal turbine's generator from vibration data. Training and testing data were recorded over two deployment periods of operation from an accelerometer sensor placed within the nacelle of the turbine, representing ideal and faulty responses over a range of operating conditions. The paper evaluates a deep learning approach through a stacked autoencoder network in comparison to feature-based classification methods, where features have been extracted over varying rotation speeds through the Vold-Kalma filter.",
keywords = "renewable energy, tidal currents, tidal turbines, condition monitoring, fault detection",
author = "Galloway, {Grant S.} and Catterson, {Victoria M.} and Thomas Fay and Andrew Robb and Craig Love",
year = "2016",
month = "7",
day = "8",
language = "English",
isbn = "9781936263219",
pages = "172--180",
editor = "Ioana Eballard and Anibal Bregon",
booktitle = "Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016",

}

Galloway, GS, Catterson, VM, Fay, T, Robb, A & Love, C 2016, Diagnosis of tidal turbine vibration data through deep neural networks. in I Eballard & A Bregon (eds), Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016. pp. 172-180, Third European Conference of the Prognostics and Health Management Society 2016, Bilbao, Spain, 5/07/16.

Diagnosis of tidal turbine vibration data through deep neural networks. / Galloway, Grant S.; Catterson, Victoria M.; Fay, Thomas; Robb, Andrew ; Love, Craig.

Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016. ed. / Ioana Eballard; Anibal Bregon. 2016. p. 172-180.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Diagnosis of tidal turbine vibration data through deep neural networks

AU - Galloway, Grant S.

AU - Catterson, Victoria M.

AU - Fay, Thomas

AU - Robb, Andrew

AU - Love, Craig

PY - 2016/7/8

Y1 - 2016/7/8

N2 - Tidal power is an emerging field of renewable energy, harnessing the power of regular and predictable tidal currents. However, maintenance of submerged equipment is a major challenge. Routine visual inspections of equipment must be performed onshore, requiring the costly removal of turbines from the sea bed and resulting in long periods of downtime. The development of condition monitoring techniques providing automated fault detection can therefore be extremely beneficial to this industry, reducing the dependency on inspections and allowing maintenance to be planned efficiently. This paper investigates the use of deep learning approaches for fault detection within a tidal turbine's generator from vibration data. Training and testing data were recorded over two deployment periods of operation from an accelerometer sensor placed within the nacelle of the turbine, representing ideal and faulty responses over a range of operating conditions. The paper evaluates a deep learning approach through a stacked autoencoder network in comparison to feature-based classification methods, where features have been extracted over varying rotation speeds through the Vold-Kalma filter.

AB - Tidal power is an emerging field of renewable energy, harnessing the power of regular and predictable tidal currents. However, maintenance of submerged equipment is a major challenge. Routine visual inspections of equipment must be performed onshore, requiring the costly removal of turbines from the sea bed and resulting in long periods of downtime. The development of condition monitoring techniques providing automated fault detection can therefore be extremely beneficial to this industry, reducing the dependency on inspections and allowing maintenance to be planned efficiently. This paper investigates the use of deep learning approaches for fault detection within a tidal turbine's generator from vibration data. Training and testing data were recorded over two deployment periods of operation from an accelerometer sensor placed within the nacelle of the turbine, representing ideal and faulty responses over a range of operating conditions. The paper evaluates a deep learning approach through a stacked autoencoder network in comparison to feature-based classification methods, where features have been extracted over varying rotation speeds through the Vold-Kalma filter.

KW - renewable energy

KW - tidal currents

KW - tidal turbines

KW - condition monitoring

KW - fault detection

UR - https://www.phmsociety.org/

UR - https://www.phmsociety.org/events/conference/europhm/16/proceedings

UR - https://www.phmsociety.org/events/conference/phm/europe/16

M3 - Conference contribution book

SN - 9781936263219

SP - 172

EP - 180

BT - Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016

A2 - Eballard, Ioana

A2 - Bregon, Anibal

ER -

Galloway GS, Catterson VM, Fay T, Robb A, Love C. Diagnosis of tidal turbine vibration data through deep neural networks. In Eballard I, Bregon A, editors, Proceedings of the Third European Conference of the Prognostics and Health Management Society 2016. 2016. p. 172-180