Time-series deep learning fault detection with the application of wind turbine benchmark

Reihane Rahimilarki; Zhiwei Gao; Nanlin Jin; Aihua Zhang

doi:10.1109/INDIN41052.2019.8972237

Time-series deep learning fault detection with the application of wind turbine benchmark

Reihane Rahimilarki, Zhiwei Gao, Nanlin Jin, Aihua Zhang

School of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

11 Citations (Scopus)

Abstract

In this paper, a deep learning fault detection approach is proposed based on the convolutional neural network in order to cope with one class of faults in wind turbine systems. Fault detection is very vital in nowadays industries due to the fact that instantly detection can prevent waste of cost and time. Deep learning as one of the powerful approaches in machine learning is a promising method to identify and classify the intrigued problems, which are hard to solve by classical methods. In this case, less than 5% performance reduction in generator torque along with sensor noise, which is challenging to identify by an operator or classical diagnosis methods is studied. The proposed algorithm, which is evolved from convolutional neural network idea, is evaluated in simulation based on a 4.8 MW wind turbine benchmark and the accuracy of the results confirms the persuasive performance of the suggested approach.

Original language	English
Title of host publication	2019 IEEE 17th International Conference on Industrial Informatics (INDIN)
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	1337-1342
Number of pages	6
ISBN (Electronic)	9781728129273
DOIs	https://doi.org/10.1109/INDIN41052.2019.8972237
Publication status	Published - Jul 2019
Event	17th IEEE International Conference on Industrial Informatics, INDIN 2019 - Helsinki-Espoo, Finland Duration: 22 Jul 2019 → 25 Jul 2019

Publication series

Name	IEEE International Conference on Industrial Informatics (INDIN)
Volume	2019-July
ISSN (Print)	1935-4576

Conference

Conference	17th IEEE International Conference on Industrial Informatics, INDIN 2019
Country/Territory	Finland
City	Helsinki-Espoo
Period	22/07/19 → 25/07/19

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Convolutional neural networks
Deep learning
Fault detection
Time-series data
Wind turbines

Access to Document

10.1109/INDIN41052.2019.8972237

Cite this

Rahimilarki, R., Gao, Z., Jin, N., & Zhang, A. (2019). Time-series deep learning fault detection with the application of wind turbine benchmark. In 2019 IEEE 17th International Conference on Industrial Informatics (INDIN) (pp. 1337-1342). Article 8972237 (IEEE International Conference on Industrial Informatics (INDIN); Vol. 2019-July). IEEE. https://doi.org/10.1109/INDIN41052.2019.8972237

@inproceedings{3438d474373248fdaafec6738984f725,

title = "Time-series deep learning fault detection with the application of wind turbine benchmark",

abstract = "In this paper, a deep learning fault detection approach is proposed based on the convolutional neural network in order to cope with one class of faults in wind turbine systems. Fault detection is very vital in nowadays industries due to the fact that instantly detection can prevent waste of cost and time. Deep learning as one of the powerful approaches in machine learning is a promising method to identify and classify the intrigued problems, which are hard to solve by classical methods. In this case, less than 5\% performance reduction in generator torque along with sensor noise, which is challenging to identify by an operator or classical diagnosis methods is studied. The proposed algorithm, which is evolved from convolutional neural network idea, is evaluated in simulation based on a 4.8 MW wind turbine benchmark and the accuracy of the results confirms the persuasive performance of the suggested approach.",

keywords = "Convolutional neural networks, Deep learning, Fault detection, Time-series data, Wind turbines",

author = "Reihane Rahimilarki and Zhiwei Gao and Nanlin Jin and Aihua Zhang",

note = "Funding Information: The work is partially supported by the RDF studentship, at the E\&E faculty in Northumbria University, and the NSFC under grant 61673074. ; 17th IEEE International Conference on Industrial Informatics, INDIN 2019 ; Conference date: 22-07-2019 Through 25-07-2019",

year = "2019",

month = jul,

doi = "10.1109/INDIN41052.2019.8972237",

language = "English",

series = "IEEE International Conference on Industrial Informatics (INDIN)",

publisher = "IEEE",

pages = "1337--1342",

booktitle = "2019 IEEE 17th International Conference on Industrial Informatics (INDIN)",

address = "United States",

}

Rahimilarki, R, Gao, Z, Jin, N & Zhang, A 2019, Time-series deep learning fault detection with the application of wind turbine benchmark. in 2019 IEEE 17th International Conference on Industrial Informatics (INDIN)., 8972237, IEEE International Conference on Industrial Informatics (INDIN), vol. 2019-July, IEEE, Piscataway, NJ , pp. 1337-1342, 17th IEEE International Conference on Industrial Informatics, INDIN 2019, Helsinki-Espoo, Finland, 22/07/19. https://doi.org/10.1109/INDIN41052.2019.8972237

Time-series deep learning fault detection with the application of wind turbine benchmark. / Rahimilarki, Reihane; Gao, Zhiwei; Jin, Nanlin et al.
2019 IEEE 17th International Conference on Industrial Informatics (INDIN). Piscataway, NJ : IEEE, 2019. p. 1337-1342 8972237 (IEEE International Conference on Industrial Informatics (INDIN); Vol. 2019-July).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Time-series deep learning fault detection with the application of wind turbine benchmark

AU - Rahimilarki, Reihane

AU - Gao, Zhiwei

AU - Jin, Nanlin

AU - Zhang, Aihua

N1 - Funding Information: The work is partially supported by the RDF studentship, at the E&E faculty in Northumbria University, and the NSFC under grant 61673074.

PY - 2019/7

Y1 - 2019/7

N2 - In this paper, a deep learning fault detection approach is proposed based on the convolutional neural network in order to cope with one class of faults in wind turbine systems. Fault detection is very vital in nowadays industries due to the fact that instantly detection can prevent waste of cost and time. Deep learning as one of the powerful approaches in machine learning is a promising method to identify and classify the intrigued problems, which are hard to solve by classical methods. In this case, less than 5% performance reduction in generator torque along with sensor noise, which is challenging to identify by an operator or classical diagnosis methods is studied. The proposed algorithm, which is evolved from convolutional neural network idea, is evaluated in simulation based on a 4.8 MW wind turbine benchmark and the accuracy of the results confirms the persuasive performance of the suggested approach.

AB - In this paper, a deep learning fault detection approach is proposed based on the convolutional neural network in order to cope with one class of faults in wind turbine systems. Fault detection is very vital in nowadays industries due to the fact that instantly detection can prevent waste of cost and time. Deep learning as one of the powerful approaches in machine learning is a promising method to identify and classify the intrigued problems, which are hard to solve by classical methods. In this case, less than 5% performance reduction in generator torque along with sensor noise, which is challenging to identify by an operator or classical diagnosis methods is studied. The proposed algorithm, which is evolved from convolutional neural network idea, is evaluated in simulation based on a 4.8 MW wind turbine benchmark and the accuracy of the results confirms the persuasive performance of the suggested approach.

KW - Convolutional neural networks

KW - Deep learning

KW - Fault detection

KW - Time-series data

KW - Wind turbines

UR - https://www.scopus.com/pages/publications/85079068704

U2 - 10.1109/INDIN41052.2019.8972237

DO - 10.1109/INDIN41052.2019.8972237

M3 - Conference contribution

AN - SCOPUS:85079068704

T3 - IEEE International Conference on Industrial Informatics (INDIN)

SP - 1337

EP - 1342

BT - 2019 IEEE 17th International Conference on Industrial Informatics (INDIN)

PB - IEEE

CY - Piscataway, NJ

T2 - 17th IEEE International Conference on Industrial Informatics, INDIN 2019

Y2 - 22 July 2019 through 25 July 2019

ER -

Time-series deep learning fault detection with the application of wind turbine benchmark

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this