Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning

Shichen Zhang; Zizhen Qiu; Lingxiao Zhao; Xin Huang; Fang Wang; Yang Kang

doi:10.1007/978-3-031-69483-7_27

Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning

Shichen Zhang, Zizhen Qiu^*, Lingxiao Zhao, Xin Huang, Fang Wang, Yang Kang

^*Corresponding author for this work

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

Abstract

This paper introduces a method for fault prediction in electric drive systems, specifically designed to address the challenge of few-shot fault sample availability. Utilizing operational data under normal conditions, the method involves the pre-training of a Long Short-Term Memory (LSTM) network model, followed by the augmentation of scarce fault data through time sliding windows. Transfer learning is then applied to adapt the pre-trained model to fault detection tasks. This innovative approach not only significantly enhances the accuracy of fault prediction but also leverages minimal data to achieve high reliability and efficiency in electric drive assemblies. The methodology demonstrates a substantial improvement over traditional fault prediction models, offering a practical solution to the often resource-intensive field of electric drive system maintenance.

Original language	English
Title of host publication	Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP
Editors	Tongtong Liu, Fan Zhang, Shiqing Huang, Jingjing Wang, Fengshou Gu
Publisher	Springer Science and Business Media B.V.
Pages	285-295
Number of pages	11
ISBN (Print)	9783031694820
DOIs	https://doi.org/10.1007/978-3-031-69483-7_27
Publication status	Published - 2024
Externally published	Yes
Event	TEPEN International Workshop on Fault Diagnostics and Prognostics, TEPEN-IWFDP 2024 - Qingdao, China Duration: 8 May 2024 → 11 May 2024

Publication series

Name	Mechanisms and Machine Science
Volume	169 MMS
ISSN (Print)	2211-0984
ISSN (Electronic)	2211-0992

Conference

Conference	TEPEN International Workshop on Fault Diagnostics and Prognostics, TEPEN-IWFDP 2024
Country/Territory	China
City	Qingdao
Period	8/05/24 → 11/05/24

Keywords

Electric Drive Systems
Few-shot Sample
Transfer Learning

Access to Document

10.1007/978-3-031-69483-7_27

Cite this

Zhang, S., Qiu, Z., Zhao, L., Huang, X., Wang, F., & Kang, Y. (2024). Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning. In T. Liu, F. Zhang, S. Huang, J. Wang, & F. Gu (Eds.), Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP (pp. 285-295). (Mechanisms and Machine Science; Vol. 169 MMS). Springer Science and Business Media B.V.. https://doi.org/10.1007/978-3-031-69483-7_27

Zhang, Shichen ; Qiu, Zizhen ; Zhao, Lingxiao et al. / Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning. Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP. editor / Tongtong Liu ; Fan Zhang ; Shiqing Huang ; Jingjing Wang ; Fengshou Gu. Springer Science and Business Media B.V., 2024. pp. 285-295 (Mechanisms and Machine Science).

@inproceedings{c38914aa4e284067957d6f84cafc942b,

title = "Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning",

abstract = "This paper introduces a method for fault prediction in electric drive systems, specifically designed to address the challenge of few-shot fault sample availability. Utilizing operational data under normal conditions, the method involves the pre-training of a Long Short-Term Memory (LSTM) network model, followed by the augmentation of scarce fault data through time sliding windows. Transfer learning is then applied to adapt the pre-trained model to fault detection tasks. This innovative approach not only significantly enhances the accuracy of fault prediction but also leverages minimal data to achieve high reliability and efficiency in electric drive assemblies. The methodology demonstrates a substantial improvement over traditional fault prediction models, offering a practical solution to the often resource-intensive field of electric drive system maintenance.",

keywords = "Electric Drive Systems, Few-shot Sample, Transfer Learning",

author = "Shichen Zhang and Zizhen Qiu and Lingxiao Zhao and Xin Huang and Fang Wang and Yang Kang",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.; TEPEN International Workshop on Fault Diagnostics and Prognostics, TEPEN-IWFDP 2024 ; Conference date: 08-05-2024 Through 11-05-2024",

year = "2024",

doi = "10.1007/978-3-031-69483-7_27",

language = "English",

isbn = "9783031694820",

series = "Mechanisms and Machine Science",

publisher = "Springer Science and Business Media B.V.",

pages = "285--295",

editor = "Tongtong Liu and Fan Zhang and Shiqing Huang and Jingjing Wang and Fengshou Gu",

booktitle = "Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP",

address = "Germany",

}

Zhang, S, Qiu, Z, Zhao, L, Huang, X, Wang, F & Kang, Y 2024, Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning. in T Liu, F Zhang, S Huang, J Wang & F Gu (eds), Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP. Mechanisms and Machine Science, vol. 169 MMS, Springer Science and Business Media B.V., pp. 285-295, TEPEN International Workshop on Fault Diagnostics and Prognostics, TEPEN-IWFDP 2024, Qingdao, China, 8/05/24. https://doi.org/10.1007/978-3-031-69483-7_27

Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning. / Zhang, Shichen; Qiu, Zizhen; Zhao, Lingxiao et al.
Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP. ed. / Tongtong Liu; Fan Zhang; Shiqing Huang; Jingjing Wang; Fengshou Gu. Springer Science and Business Media B.V., 2024. p. 285-295 (Mechanisms and Machine Science; Vol. 169 MMS).

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

TY - GEN

T1 - Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning

AU - Zhang, Shichen

AU - Qiu, Zizhen

AU - Zhao, Lingxiao

AU - Huang, Xin

AU - Wang, Fang

AU - Kang, Yang

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

PY - 2024

Y1 - 2024

N2 - This paper introduces a method for fault prediction in electric drive systems, specifically designed to address the challenge of few-shot fault sample availability. Utilizing operational data under normal conditions, the method involves the pre-training of a Long Short-Term Memory (LSTM) network model, followed by the augmentation of scarce fault data through time sliding windows. Transfer learning is then applied to adapt the pre-trained model to fault detection tasks. This innovative approach not only significantly enhances the accuracy of fault prediction but also leverages minimal data to achieve high reliability and efficiency in electric drive assemblies. The methodology demonstrates a substantial improvement over traditional fault prediction models, offering a practical solution to the often resource-intensive field of electric drive system maintenance.

AB - This paper introduces a method for fault prediction in electric drive systems, specifically designed to address the challenge of few-shot fault sample availability. Utilizing operational data under normal conditions, the method involves the pre-training of a Long Short-Term Memory (LSTM) network model, followed by the augmentation of scarce fault data through time sliding windows. Transfer learning is then applied to adapt the pre-trained model to fault detection tasks. This innovative approach not only significantly enhances the accuracy of fault prediction but also leverages minimal data to achieve high reliability and efficiency in electric drive assemblies. The methodology demonstrates a substantial improvement over traditional fault prediction models, offering a practical solution to the often resource-intensive field of electric drive system maintenance.

KW - Electric Drive Systems

KW - Few-shot Sample

KW - Transfer Learning

UR - http://www.scopus.com/inward/record.url?scp=85204404280&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-69483-7_27

DO - 10.1007/978-3-031-69483-7_27

M3 - Conference contribution

AN - SCOPUS:85204404280

SN - 9783031694820

T3 - Mechanisms and Machine Science

SP - 285

EP - 295

BT - Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP

A2 - Liu, Tongtong

A2 - Zhang, Fan

A2 - Huang, Shiqing

A2 - Wang, Jingjing

A2 - Gu, Fengshou

PB - Springer Science and Business Media B.V.

T2 - TEPEN International Workshop on Fault Diagnostics and Prognostics, TEPEN-IWFDP 2024

Y2 - 8 May 2024 through 11 May 2024

ER -

Zhang S, Qiu Z, Zhao L, Huang X, Wang F, Kang Y. Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning. In Liu T, Zhang F, Huang S, Wang J, Gu F, editors, Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic - TEPEN2024-IWFDP. Springer Science and Business Media B.V. 2024. p. 285-295. (Mechanisms and Machine Science). doi: 10.1007/978-3-031-69483-7_27

Research on Few-Shot Sample Fault Prediction Method for Electric Drive Systems Based on Transfer Learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this