A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection

Xiangzhao Xia; Xuyun Fu; Shisheng Zhong; Xingjie Zhou; Zhengfeng Bai

doi:10.1109/PHM2022-London52454.2022.00017

A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection

Xiangzhao Xia
, Xuyun Fu^*
, Shisheng Zhong
, Xingjie Zhou
, Zhengfeng Bai

^*Corresponding author for this work

School of Mechatronics Engineering, Harbin Institute of Technology
School of Ocean Engineering, Harbin Institute of Technology Weihai

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

Abstract

Traditional methods for gas path anomaly detection cannot fully extract remarkable shape features that can represent the gas path anomaly mode. Therefore, a feature representation method based on dual segment and entropy evaluation for aeroengine gas path anomaly detection is proposed in this paper. Taking the temporal and spatial correlations of the multivariate time series into consideration, the expression rule of the anomaly mode in the multivariate gas path parameter deviation time series is analyzed, on this basis, time series subsequence segment method is determined. To obtain the features that best fit the anomaly expression rule, a dual segment method based on piecewise optimal fitting is proposed. The entropy evaluation method is introduced to comprehensively evaluate and optimize the primary features while calculating the common shape features of subsequence, and then the remarkable shape feature matrix for anomaly detection is determined. Finally, the early warning for the gas path anomaly is realized by mining the potential anomaly mode of the gas path state using isolation forest model. The experimental results show that this method can improve the accuracy of aeroengine gas path anomaly detection.

Original language	English
Title of host publication	Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022
Editors	Chuan Li, Gianluca Valentino, Ling Kang, Diego Cabrera, Dejan Gjorgjevikj
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	50-57
Number of pages	8
ISBN (Electronic)	9781665479547
DOIs	https://doi.org/10.1109/PHM2022-London52454.2022.00017
State	Published - 2022
Externally published	Yes
Event	2022 Prognostics and Health Management Conference, PHM-London 2022 - London, United Kingdom Duration: 27 May 2022 → 29 May 2022

Publication series

Name	Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022

Conference

Conference	2022 Prognostics and Health Management Conference, PHM-London 2022
Country/Territory	United Kingdom
City	London
Period	27/05/22 → 29/05/22

Keywords

aeroengine
anomaly detection
dual segment
entropy evaluation
feature representation
isolation forest

Access to Document

10.1109/PHM2022-London52454.2022.00017

Cite this

Xia, X., Fu, X., Zhong, S., Zhou, X., & Bai, Z. (2022). A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection. In C. Li, G. Valentino, L. Kang, D. Cabrera, & D. Gjorgjevikj (Eds.), Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022 (pp. 50-57). (Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PHM2022-London52454.2022.00017

Xia, Xiangzhao ; Fu, Xuyun ; Zhong, Shisheng et al. / A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection. Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022. editor / Chuan Li ; Gianluca Valentino ; Ling Kang ; Diego Cabrera ; Dejan Gjorgjevikj. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 50-57 (Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022).

@inproceedings{6d9682567729460ab15d8d4c277753df,

title = "A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection",

abstract = "Traditional methods for gas path anomaly detection cannot fully extract remarkable shape features that can represent the gas path anomaly mode. Therefore, a feature representation method based on dual segment and entropy evaluation for aeroengine gas path anomaly detection is proposed in this paper. Taking the temporal and spatial correlations of the multivariate time series into consideration, the expression rule of the anomaly mode in the multivariate gas path parameter deviation time series is analyzed, on this basis, time series subsequence segment method is determined. To obtain the features that best fit the anomaly expression rule, a dual segment method based on piecewise optimal fitting is proposed. The entropy evaluation method is introduced to comprehensively evaluate and optimize the primary features while calculating the common shape features of subsequence, and then the remarkable shape feature matrix for anomaly detection is determined. Finally, the early warning for the gas path anomaly is realized by mining the potential anomaly mode of the gas path state using isolation forest model. The experimental results show that this method can improve the accuracy of aeroengine gas path anomaly detection.",

keywords = "aeroengine, anomaly detection, dual segment, entropy evaluation, feature representation, isolation forest",

author = "Xiangzhao Xia and Xuyun Fu and Shisheng Zhong and Xingjie Zhou and Zhengfeng Bai",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 Prognostics and Health Management Conference, PHM-London 2022 ; Conference date: 27-05-2022 Through 29-05-2022",

year = "2022",

doi = "10.1109/PHM2022-London52454.2022.00017",

language = "英语",

series = "Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022",

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

pages = "50--57",

editor = "Chuan Li and Gianluca Valentino and Ling Kang and Diego Cabrera and Dejan Gjorgjevikj",

booktitle = "Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022",

address = "美国",

}

Xia, X, Fu, X , Zhong, S, Zhou, X & Bai, Z 2022, A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection. in C Li, G Valentino, L Kang, D Cabrera & D Gjorgjevikj (eds), Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022. Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022, Institute of Electrical and Electronics Engineers Inc., pp. 50-57, 2022 Prognostics and Health Management Conference, PHM-London 2022, London, United Kingdom, 27/05/22. https://doi.org/10.1109/PHM2022-London52454.2022.00017

A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection. / Xia, Xiangzhao; Fu, Xuyun ; Zhong, Shisheng et al.
Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022. ed. / Chuan Li; Gianluca Valentino; Ling Kang; Diego Cabrera; Dejan Gjorgjevikj. Institute of Electrical and Electronics Engineers Inc., 2022. p. 50-57 (Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022).

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

TY - GEN

T1 - A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection

AU - Xia, Xiangzhao

AU - Fu, Xuyun

AU - Zhong, Shisheng

AU - Zhou, Xingjie

AU - Bai, Zhengfeng

PY - 2022

Y1 - 2022

N2 - Traditional methods for gas path anomaly detection cannot fully extract remarkable shape features that can represent the gas path anomaly mode. Therefore, a feature representation method based on dual segment and entropy evaluation for aeroengine gas path anomaly detection is proposed in this paper. Taking the temporal and spatial correlations of the multivariate time series into consideration, the expression rule of the anomaly mode in the multivariate gas path parameter deviation time series is analyzed, on this basis, time series subsequence segment method is determined. To obtain the features that best fit the anomaly expression rule, a dual segment method based on piecewise optimal fitting is proposed. The entropy evaluation method is introduced to comprehensively evaluate and optimize the primary features while calculating the common shape features of subsequence, and then the remarkable shape feature matrix for anomaly detection is determined. Finally, the early warning for the gas path anomaly is realized by mining the potential anomaly mode of the gas path state using isolation forest model. The experimental results show that this method can improve the accuracy of aeroengine gas path anomaly detection.

AB - Traditional methods for gas path anomaly detection cannot fully extract remarkable shape features that can represent the gas path anomaly mode. Therefore, a feature representation method based on dual segment and entropy evaluation for aeroengine gas path anomaly detection is proposed in this paper. Taking the temporal and spatial correlations of the multivariate time series into consideration, the expression rule of the anomaly mode in the multivariate gas path parameter deviation time series is analyzed, on this basis, time series subsequence segment method is determined. To obtain the features that best fit the anomaly expression rule, a dual segment method based on piecewise optimal fitting is proposed. The entropy evaluation method is introduced to comprehensively evaluate and optimize the primary features while calculating the common shape features of subsequence, and then the remarkable shape feature matrix for anomaly detection is determined. Finally, the early warning for the gas path anomaly is realized by mining the potential anomaly mode of the gas path state using isolation forest model. The experimental results show that this method can improve the accuracy of aeroengine gas path anomaly detection.

KW - aeroengine

KW - anomaly detection

KW - dual segment

KW - entropy evaluation

KW - feature representation

KW - isolation forest

UR - https://www.scopus.com/pages/publications/85134895338

U2 - 10.1109/PHM2022-London52454.2022.00017

DO - 10.1109/PHM2022-London52454.2022.00017

M3 - 会议稿件

AN - SCOPUS:85134895338

T3 - Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022

SP - 50

EP - 57

BT - Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022

A2 - Li, Chuan

A2 - Valentino, Gianluca

A2 - Kang, Ling

A2 - Cabrera, Diego

A2 - Gjorgjevikj, Dejan

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 Prognostics and Health Management Conference, PHM-London 2022

Y2 - 27 May 2022 through 29 May 2022

ER -

Xia X, Fu X , Zhong S, Zhou X, Bai Z. A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection. In Li C, Valentino G, Kang L, Cabrera D, Gjorgjevikj D, editors, Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 50-57. (Proceedings - 2022 Prognostics and Health Management Conference, PHM-London 2022). doi: 10.1109/PHM2022-London52454.2022.00017

A Feature Representation Method Based on Dual Segment and Entropy Evaluation for Aeroengine Gas Path Anomaly Detection

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this