A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation

Yiran Sun; Feng Jin; Nan Yang; Lu Yang; Guigang Zhang; Jian Wang

doi:10.1109/PHM-BEIJING63284.2024.10874628

A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation

Yiran Sun
, Feng Jin
, Nan Yang^*
, Lu Yang
, Guigang Zhang
, Jian Wang

^*Corresponding author for this work

School of Automation

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

Abstract

Facing the problem of insufficient labeled data, which is common in the field of fault diagnosis, this study proposes an approach based on the CDAN domain adaptation model. The performance of the fault diagnosis model is improved by fusing data from different fault types. Experiments conducted on the Case Western Reserve University sliding bearing dataset validate the effectiveness of this method compared to traditional model training and direct data augmentation strategies, which achieves higher data utilization and fault diagnosis accuracy. Feature space visualization analysis further confirms the effectiveness of feature alignment between source and target domains.

Original language	English
Title of host publication	15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024
Editors	Huimin Wang, Steven Li
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350354010
DOIs	https://doi.org/10.1109/PHM-BEIJING63284.2024.10874628
Publication status	Published - 2024
Event	15th IEEE Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024 - Beijing, China Duration: 11 Oct 2024 → 13 Oct 2024

Publication series

Name	15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024

Conference

Conference	15th IEEE Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024
Country/Territory	China
City	Beijing
Period	11/10/24 → 13/10/24

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

Conditional adversarial neural network (CDAN)
domain adaptive
fault diagnosis

Access to Document

10.1109/PHM-BEIJING63284.2024.10874628

Cite this

Sun, Y., Jin, F., Yang, N., Yang, L., Zhang, G., & Wang, J. (2024). A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation. In H. Wang, & S. Li (Eds.), 15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024 (15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PHM-BEIJING63284.2024.10874628

Sun, Yiran ; Jin, Feng ; Yang, Nan et al. / A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation. 15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024. editor / Huimin Wang ; Steven Li. Institute of Electrical and Electronics Engineers Inc., 2024. (15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024).

@inproceedings{9aa920bf0d1040028e1437d1067b7511,

title = "A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation",

abstract = "Facing the problem of insufficient labeled data, which is common in the field of fault diagnosis, this study proposes an approach based on the CDAN domain adaptation model. The performance of the fault diagnosis model is improved by fusing data from different fault types. Experiments conducted on the Case Western Reserve University sliding bearing dataset validate the effectiveness of this method compared to traditional model training and direct data augmentation strategies, which achieves higher data utilization and fault diagnosis accuracy. Feature space visualization analysis further confirms the effectiveness of feature alignment between source and target domains.",

keywords = "Conditional adversarial neural network (CDAN), domain adaptive, fault diagnosis",

author = "Yiran Sun and Feng Jin and Nan Yang and Lu Yang and Guigang Zhang and Jian Wang",

note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 15th IEEE Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024 ; Conference date: 11-10-2024 Through 13-10-2024",

year = "2024",

doi = "10.1109/PHM-BEIJING63284.2024.10874628",

language = "English",

series = "15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024",

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

editor = "Huimin Wang and Steven Li",

booktitle = "15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024",

address = "United States",

}

Sun, Y, Jin, F, Yang, N, Yang, L, Zhang, G & Wang, J 2024, A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation. in H Wang & S Li (eds), 15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024. 15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024, Institute of Electrical and Electronics Engineers Inc., 15th IEEE Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024, Beijing, China, 11/10/24. https://doi.org/10.1109/PHM-BEIJING63284.2024.10874628

A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation. / Sun, Yiran; Jin, Feng; Yang, Nan et al.
15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024. ed. / Huimin Wang; Steven Li. Institute of Electrical and Electronics Engineers Inc., 2024. (15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024).

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

TY - GEN

T1 - A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation

AU - Sun, Yiran

AU - Jin, Feng

AU - Yang, Nan

AU - Yang, Lu

AU - Zhang, Guigang

AU - Wang, Jian

PY - 2024

Y1 - 2024

N2 - Facing the problem of insufficient labeled data, which is common in the field of fault diagnosis, this study proposes an approach based on the CDAN domain adaptation model. The performance of the fault diagnosis model is improved by fusing data from different fault types. Experiments conducted on the Case Western Reserve University sliding bearing dataset validate the effectiveness of this method compared to traditional model training and direct data augmentation strategies, which achieves higher data utilization and fault diagnosis accuracy. Feature space visualization analysis further confirms the effectiveness of feature alignment between source and target domains.

AB - Facing the problem of insufficient labeled data, which is common in the field of fault diagnosis, this study proposes an approach based on the CDAN domain adaptation model. The performance of the fault diagnosis model is improved by fusing data from different fault types. Experiments conducted on the Case Western Reserve University sliding bearing dataset validate the effectiveness of this method compared to traditional model training and direct data augmentation strategies, which achieves higher data utilization and fault diagnosis accuracy. Feature space visualization analysis further confirms the effectiveness of feature alignment between source and target domains.

KW - Conditional adversarial neural network (CDAN)

KW - domain adaptive

KW - fault diagnosis

UR - https://www.scopus.com/pages/publications/85219641047

U2 - 10.1109/PHM-BEIJING63284.2024.10874628

DO - 10.1109/PHM-BEIJING63284.2024.10874628

M3 - Conference contribution

AN - SCOPUS:85219641047

T3 - 15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024

BT - 15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024

A2 - Wang, Huimin

A2 - Li, Steven

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 15th IEEE Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024

Y2 - 11 October 2024 through 13 October 2024

ER -

Sun Y, Jin F, Yang N, Yang L, Zhang G, Wang J. A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation. In Wang H, Li S, editors, 15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024. Institute of Electrical and Electronics Engineers Inc. 2024. (15th Global Reliability and Prognostics and Health Management Conference, PHM-Beijing 2024). doi: 10.1109/PHM-BEIJING63284.2024.10874628

A Low Resource Fault Diagnosis Model Based on CDAN Domain Adaptation

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this