Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain

Wing Chong Lo; C. K.M. Lee; Chak Nam Wong; Jingyuan Huang

doi:10.1109/ISSSR58837.2023.00072

Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain

Wing Chong Lo, C. K.M. Lee, Chak Nam Wong, Jingyuan Huang

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

Abstract

Time series contributed by sensor signal can be used for fault diagnosis, and machine learning is adopted to identify the causes of failure and the relevant factors in the time-frequency domain. However, the lack of labeled data, incredibly faulty data in various conditions, is one of the significant challenges when applying machine learning approaches. To reduce the barrier of applying those approaches, this study investigated the use of transfer learning. A high accuracy of nearly 95% for classification without the labels in training is found. There is potential research direction in unsupervised domain adaptation and domain generalization.

Original language	English
Title of host publication	Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	442-443
Number of pages	2
ISBN (Electronic)	9798350302479
DOIs	https://doi.org/10.1109/ISSSR58837.2023.00072 https://doi.org/10.1109/ISSSR58837.2023.00072
Publication status	Published - Jun 2023
Event	9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023 - Hangzhou, China Duration: 10 Jun 2023 → 11 Jun 2023

Publication series

Name	Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023

Conference

Conference	9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023
Country/Territory	China
City	Hangzhou
Period	10/06/23 → 11/06/23

Keywords

Domain Generalization
Fault Diagnosis
Time Series
Time-Frequency Analysis
Transfer Learning
Unsupervised Domain Adaptation

ASJC Scopus subject areas

Computer Networks and Communications
Information Systems
Industrial and Manufacturing Engineering
Mechanical Engineering
Safety, Risk, Reliability and Quality

More information

Cite this

Lo, W. C., Lee, C. K. M., Wong, C. N., & Huang, J. (2023). Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain. In Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023 (pp. 442-443). (Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISSSR58837.2023.00072, https://doi.org/10.1109/ISSSR58837.2023.00072

Lo, Wing Chong ; Lee, C. K.M. ; Wong, Chak Nam et al. / Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain. Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 442-443 (Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023).

@inproceedings{aeb9fbc41bb84340b45069c2754a8c39,

title = "Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain",

abstract = "Time series contributed by sensor signal can be used for fault diagnosis, and machine learning is adopted to identify the causes of failure and the relevant factors in the time-frequency domain. However, the lack of labeled data, incredibly faulty data in various conditions, is one of the significant challenges when applying machine learning approaches. To reduce the barrier of applying those approaches, this study investigated the use of transfer learning. A high accuracy of nearly 95% for classification without the labels in training is found. There is potential research direction in unsupervised domain adaptation and domain generalization.",

keywords = "Domain Generalization, Fault Diagnosis, Time Series, Time-Frequency Analysis, Transfer Learning, Unsupervised Domain Adaptation",

author = "Lo, {Wing Chong} and Lee, {C. K.M.} and Wong, {Chak Nam} and Jingyuan Huang",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023 ; Conference date: 10-06-2023 Through 11-06-2023",

year = "2023",

month = jun,

doi = "10.1109/ISSSR58837.2023.00072",

language = "English",

series = "Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "442--443",

booktitle = "Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023",

}

Lo, WC, Lee, CKM, Wong, CN & Huang, J 2023, Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain. in Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023. Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023, Institute of Electrical and Electronics Engineers Inc., pp. 442-443, 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023, Hangzhou, China, 10/06/23. https://doi.org/10.1109/ISSSR58837.2023.00072, https://doi.org/10.1109/ISSSR58837.2023.00072

Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain. / Lo, Wing Chong; Lee, C. K.M.; Wong, Chak Nam et al.
Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 442-443 (Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023).

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain

AU - Lo, Wing Chong

AU - Lee, C. K.M.

AU - Wong, Chak Nam

AU - Huang, Jingyuan

PY - 2023/6

Y1 - 2023/6

N2 - Time series contributed by sensor signal can be used for fault diagnosis, and machine learning is adopted to identify the causes of failure and the relevant factors in the time-frequency domain. However, the lack of labeled data, incredibly faulty data in various conditions, is one of the significant challenges when applying machine learning approaches. To reduce the barrier of applying those approaches, this study investigated the use of transfer learning. A high accuracy of nearly 95% for classification without the labels in training is found. There is potential research direction in unsupervised domain adaptation and domain generalization.

AB - Time series contributed by sensor signal can be used for fault diagnosis, and machine learning is adopted to identify the causes of failure and the relevant factors in the time-frequency domain. However, the lack of labeled data, incredibly faulty data in various conditions, is one of the significant challenges when applying machine learning approaches. To reduce the barrier of applying those approaches, this study investigated the use of transfer learning. A high accuracy of nearly 95% for classification without the labels in training is found. There is potential research direction in unsupervised domain adaptation and domain generalization.

KW - Domain Generalization

KW - Fault Diagnosis

KW - Time Series

KW - Time-Frequency Analysis

KW - Transfer Learning

KW - Unsupervised Domain Adaptation

UR - http://www.scopus.com/inward/record.url?scp=85170084650&partnerID=8YFLogxK

U2 - 10.1109/ISSSR58837.2023.00072

DO - 10.1109/ISSSR58837.2023.00072

M3 - Conference article published in proceeding or book

AN - SCOPUS:85170084650

T3 - Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023

SP - 442

EP - 443

BT - Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023

Y2 - 10 June 2023 through 11 June 2023

ER -

Lo WC, Lee CKM, Wong CN, Huang J. Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain. In Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 442-443. (Proceedings - 2023 9th International Symposium on System Security, Safety, and Reliability, ISSSR 2023). doi: 10.1109/ISSSR58837.2023.00072, 10.1109/ISSSR58837.2023.00072

Fault Diagnosis for Time Series Signal based on Transfer Learning in Time-Frequency Domain

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this