An Ensemble Approach for Fault Diagnosis via Continuous Learning

Dapeng Zhang; Zhiwei Gao

doi:10.1109/INDIN45523.2021.9557388

An Ensemble Approach for Fault Diagnosis via Continuous Learning

Dapeng Zhang, Zhiwei Gao

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

5 Citations (Scopus)

Abstract

The great success of deep neural network (DNN) in image field stimulates its application in fault detection and diagnose. However due to the limitation of system security, it is impossible to obtain complete fault data as the training database for neural network, so that it is challenging to identify a fault that never occurred before. In this paper, an ensemble approach is proposed to adapt to a new fault by adding output branches of the neural network. Firstly, the time series are transferred to numerous imaging matrixes. The intrinsic characteristics of the matrixes are then extracted using deep neural network which are used to judge whether it is a new fault according to the distance criterion. For a new fault, the DNN will retrain by transferring learning in order to reduce the computation and training time. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.

Original language	English
Title of host publication	Proceedings - 2021 IEEE 19th International Conference on Industrial Informatics (INDIN)
Place of Publication	Piscataway, US
Publisher	IEEE
Pages	426-430
Number of pages	5
ISBN (Electronic)	9781728143958
ISBN (Print)	9781728143965
DOIs	https://doi.org/10.1109/INDIN45523.2021.9557388
Publication status	Published - 21 Jul 2021
Event	19th IEEE International Conference on Industrial Informatics, INDIN 2021 - Mallorca, Spain Duration: 21 Jul 2021 → 23 Jul 2021

Publication series

Name	IEEE International Conference on Industrial Informatics (INDIN)
Volume	2021-July
ISSN (Print)	1935-4576

Conference

Conference	19th IEEE International Conference on Industrial Informatics, INDIN 2021
Country/Territory	Spain
City	Mallorca
Period	21/07/21 → 23/07/21

UN SDGs

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

SDG 7 Affordable and Clean Energy

Keywords

Deep learning
Fault detection and diagnose
Time series
Wind turbine benchmark model

Access to Document

10.1109/INDIN45523.2021.9557388

Cite this

@inproceedings{6d85bd36a90e49a9916a69987a4e9a89,

title = "An Ensemble Approach for Fault Diagnosis via Continuous Learning",

abstract = "The great success of deep neural network (DNN) in image field stimulates its application in fault detection and diagnose. However due to the limitation of system security, it is impossible to obtain complete fault data as the training database for neural network, so that it is challenging to identify a fault that never occurred before. In this paper, an ensemble approach is proposed to adapt to a new fault by adding output branches of the neural network. Firstly, the time series are transferred to numerous imaging matrixes. The intrinsic characteristics of the matrixes are then extracted using deep neural network which are used to judge whether it is a new fault according to the distance criterion. For a new fault, the DNN will retrain by transferring learning in order to reduce the computation and training time. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.",

keywords = "Deep learning, Fault detection and diagnose, Time series, Wind turbine benchmark model",

author = "Dapeng Zhang and Zhiwei Gao",

note = "Funding Information: The authors would like to thank the research support from the National Nature Science Foundation of China (NNSFC) under Grant 61673074, School of Electrical and Information Engineering at Tianjin University, China, and the E\&E faculty at Northumbria University, Newcastle upon Tyne, Tyne and Wear, NE1 8ST, United Kingdom.; 19th IEEE International Conference on Industrial Informatics, INDIN 2021 ; Conference date: 21-07-2021 Through 23-07-2021",

year = "2021",

month = jul,

day = "21",

doi = "10.1109/INDIN45523.2021.9557388",

language = "English",

isbn = "9781728143965",

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

publisher = "IEEE",

pages = "426--430",

booktitle = "Proceedings - 2021 IEEE 19th International Conference on Industrial Informatics (INDIN)",

address = "United States",

}

Zhang, D & Gao, Z 2021, An Ensemble Approach for Fault Diagnosis via Continuous Learning. in Proceedings - 2021 IEEE 19th International Conference on Industrial Informatics (INDIN)., 9557388, IEEE International Conference on Industrial Informatics (INDIN), vol. 2021-July, IEEE, Piscataway, US, pp. 426-430, 19th IEEE International Conference on Industrial Informatics, INDIN 2021, Mallorca, Spain, 21/07/21. https://doi.org/10.1109/INDIN45523.2021.9557388

An Ensemble Approach for Fault Diagnosis via Continuous Learning. / Zhang, Dapeng; Gao, Zhiwei.
Proceedings - 2021 IEEE 19th International Conference on Industrial Informatics (INDIN). Piscataway, US: IEEE, 2021. p. 426-430 9557388 (IEEE International Conference on Industrial Informatics (INDIN); Vol. 2021-July).

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

TY - GEN

T1 - An Ensemble Approach for Fault Diagnosis via Continuous Learning

AU - Zhang, Dapeng

AU - Gao, Zhiwei

N1 - Funding Information: The authors would like to thank the research support from the National Nature Science Foundation of China (NNSFC) under Grant 61673074, School of Electrical and Information Engineering at Tianjin University, China, and the E&E faculty at Northumbria University, Newcastle upon Tyne, Tyne and Wear, NE1 8ST, United Kingdom.

PY - 2021/7/21

Y1 - 2021/7/21

N2 - The great success of deep neural network (DNN) in image field stimulates its application in fault detection and diagnose. However due to the limitation of system security, it is impossible to obtain complete fault data as the training database for neural network, so that it is challenging to identify a fault that never occurred before. In this paper, an ensemble approach is proposed to adapt to a new fault by adding output branches of the neural network. Firstly, the time series are transferred to numerous imaging matrixes. The intrinsic characteristics of the matrixes are then extracted using deep neural network which are used to judge whether it is a new fault according to the distance criterion. For a new fault, the DNN will retrain by transferring learning in order to reduce the computation and training time. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.

AB - The great success of deep neural network (DNN) in image field stimulates its application in fault detection and diagnose. However due to the limitation of system security, it is impossible to obtain complete fault data as the training database for neural network, so that it is challenging to identify a fault that never occurred before. In this paper, an ensemble approach is proposed to adapt to a new fault by adding output branches of the neural network. Firstly, the time series are transferred to numerous imaging matrixes. The intrinsic characteristics of the matrixes are then extracted using deep neural network which are used to judge whether it is a new fault according to the distance criterion. For a new fault, the DNN will retrain by transferring learning in order to reduce the computation and training time. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.

KW - Deep learning

KW - Fault detection and diagnose

KW - Time series

KW - Wind turbine benchmark model

UR - https://www.scopus.com/pages/publications/85125586768

U2 - 10.1109/INDIN45523.2021.9557388

DO - 10.1109/INDIN45523.2021.9557388

M3 - Conference contribution

AN - SCOPUS:85125586768

SN - 9781728143965

T3 - IEEE International Conference on Industrial Informatics (INDIN)

SP - 426

EP - 430

BT - Proceedings - 2021 IEEE 19th International Conference on Industrial Informatics (INDIN)

PB - IEEE

CY - Piscataway, US

T2 - 19th IEEE International Conference on Industrial Informatics, INDIN 2021

Y2 - 21 July 2021 through 23 July 2021

ER -

An Ensemble Approach for Fault Diagnosis via Continuous Learning

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this