The flywheel fault detection based on Kernel principal component analysis

Gan Hua Li; Jian Cheng Li; Ya Ni Cao; Min Qiang Xu; Ke Qiang Xia; Jun Wei; Bao Jun Lan; Li Dong

doi:10.1109/ITNEC.2019.8729163

The flywheel fault detection based on Kernel principal component analysis

Gan Hua Li^*
, Jian Cheng Li
, Ya Ni Cao
, Min Qiang Xu
, Ke Qiang Xia
, Jun Wei
, Bao Jun Lan
, Li Dong

^*Corresponding author for this work

School of Astronautics

Xi'an Satellite Control Center

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

7 Scopus citations

Abstract

Simulink model is built according to the mathematic model of flywheel system which is a closed-loop system. Considering the fault modes and the corresponding fault parameters of flywheel, the training data and testing data for two faults are collected from the Simulink model. And then the Kernel Principal Component Analysis (KPCA) is proposed to analysis the correlation of five parameters in the flywheel system. The faults of flywheel will cause some abnormal changes of testing data. This method can solve the problem of lack of fault knowledge and complex mathematic modeling. The behaviors of flywheel can be learned from the training data and the correlation is the interactions of five parameters in the flywheel system. However, the variable correlation is classified as two types, such as the nonlinear and linear relationship. The Principal Component analysis (PCA) is used to build the linear model of training data. The chosen five variables is a nonlinear relationship. In order to demonstrate the effectiveness of proposed algorithm, it is necessary to compare the detections results of KPCA with the results of PCA. Numerical simulation results show that the SPEKPCA index can detect the faults of flywheel without complex mathematical modeling, and better than the detection results of PCA model.

Original language	English
Title of host publication	Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019
Editors	Bing Xu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	425-432
Number of pages	8
ISBN (Electronic)	9781538662434
DOIs	https://doi.org/10.1109/ITNEC.2019.8729163
State	Published - Mar 2019
Event	3rd IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019 - Chengdu, China Duration: 15 Mar 2019 → 17 Mar 2019

Publication series

Name	Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019

Conference

Conference	3rd IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019
Country/Territory	China
City	Chengdu
Period	15/03/19 → 17/03/19

Keywords

Close loop control
Fault detection
Flywheel System
Kernel Principal Component Analysis
PCA

Access to Document

10.1109/ITNEC.2019.8729163

Cite this

Li, G. H., Li, J. C., Cao, Y. N., Xu, M. Q., Xia, K. Q., Wei, J., Lan, B. J., & Dong, L. (2019). The flywheel fault detection based on Kernel principal component analysis. In B. Xu (Ed.), Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019 (pp. 425-432). Article 8729163 (Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ITNEC.2019.8729163

Li, Gan Hua ; Li, Jian Cheng ; Cao, Ya Ni et al. / The flywheel fault detection based on Kernel principal component analysis. Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019. editor / Bing Xu. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 425-432 (Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019).

@inproceedings{fb4def6b2d6c417aa7844c64ab3b89f0,

title = "The flywheel fault detection based on Kernel principal component analysis",

abstract = "Simulink model is built according to the mathematic model of flywheel system which is a closed-loop system. Considering the fault modes and the corresponding fault parameters of flywheel, the training data and testing data for two faults are collected from the Simulink model. And then the Kernel Principal Component Analysis (KPCA) is proposed to analysis the correlation of five parameters in the flywheel system. The faults of flywheel will cause some abnormal changes of testing data. This method can solve the problem of lack of fault knowledge and complex mathematic modeling. The behaviors of flywheel can be learned from the training data and the correlation is the interactions of five parameters in the flywheel system. However, the variable correlation is classified as two types, such as the nonlinear and linear relationship. The Principal Component analysis (PCA) is used to build the linear model of training data. The chosen five variables is a nonlinear relationship. In order to demonstrate the effectiveness of proposed algorithm, it is necessary to compare the detections results of KPCA with the results of PCA. Numerical simulation results show that the SPEKPCA index can detect the faults of flywheel without complex mathematical modeling, and better than the detection results of PCA model.",

keywords = "Close loop control, Fault detection, Flywheel System, Kernel Principal Component Analysis, PCA",

author = "Li, \{Gan Hua\} and Li, \{Jian Cheng\} and Cao, \{Ya Ni\} and Xu, \{Min Qiang\} and Xia, \{Ke Qiang\} and Jun Wei and Lan, \{Bao Jun\} and Li Dong",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 3rd IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019 ; Conference date: 15-03-2019 Through 17-03-2019",

year = "2019",

month = mar,

doi = "10.1109/ITNEC.2019.8729163",

language = "英语",

series = "Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019",

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

pages = "425--432",

editor = "Bing Xu",

booktitle = "Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019",

address = "美国",

}

Li, GH, Li, JC, Cao, YN, Xu, MQ, Xia, KQ, Wei, J, Lan, BJ & Dong, L 2019, The flywheel fault detection based on Kernel principal component analysis. in B Xu (ed.), Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019., 8729163, Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019, Institute of Electrical and Electronics Engineers Inc., pp. 425-432, 3rd IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019, Chengdu, China, 15/03/19. https://doi.org/10.1109/ITNEC.2019.8729163

The flywheel fault detection based on Kernel principal component analysis. / Li, Gan Hua; Li, Jian Cheng; Cao, Ya Ni et al.
Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019. ed. / Bing Xu. Institute of Electrical and Electronics Engineers Inc., 2019. p. 425-432 8729163 (Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019).

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

TY - GEN

T1 - The flywheel fault detection based on Kernel principal component analysis

AU - Li, Gan Hua

AU - Li, Jian Cheng

AU - Cao, Ya Ni

AU - Xu, Min Qiang

AU - Xia, Ke Qiang

AU - Wei, Jun

AU - Lan, Bao Jun

AU - Dong, Li

PY - 2019/3

Y1 - 2019/3

N2 - Simulink model is built according to the mathematic model of flywheel system which is a closed-loop system. Considering the fault modes and the corresponding fault parameters of flywheel, the training data and testing data for two faults are collected from the Simulink model. And then the Kernel Principal Component Analysis (KPCA) is proposed to analysis the correlation of five parameters in the flywheel system. The faults of flywheel will cause some abnormal changes of testing data. This method can solve the problem of lack of fault knowledge and complex mathematic modeling. The behaviors of flywheel can be learned from the training data and the correlation is the interactions of five parameters in the flywheel system. However, the variable correlation is classified as two types, such as the nonlinear and linear relationship. The Principal Component analysis (PCA) is used to build the linear model of training data. The chosen five variables is a nonlinear relationship. In order to demonstrate the effectiveness of proposed algorithm, it is necessary to compare the detections results of KPCA with the results of PCA. Numerical simulation results show that the SPEKPCA index can detect the faults of flywheel without complex mathematical modeling, and better than the detection results of PCA model.

AB - Simulink model is built according to the mathematic model of flywheel system which is a closed-loop system. Considering the fault modes and the corresponding fault parameters of flywheel, the training data and testing data for two faults are collected from the Simulink model. And then the Kernel Principal Component Analysis (KPCA) is proposed to analysis the correlation of five parameters in the flywheel system. The faults of flywheel will cause some abnormal changes of testing data. This method can solve the problem of lack of fault knowledge and complex mathematic modeling. The behaviors of flywheel can be learned from the training data and the correlation is the interactions of five parameters in the flywheel system. However, the variable correlation is classified as two types, such as the nonlinear and linear relationship. The Principal Component analysis (PCA) is used to build the linear model of training data. The chosen five variables is a nonlinear relationship. In order to demonstrate the effectiveness of proposed algorithm, it is necessary to compare the detections results of KPCA with the results of PCA. Numerical simulation results show that the SPEKPCA index can detect the faults of flywheel without complex mathematical modeling, and better than the detection results of PCA model.

KW - Close loop control

KW - Fault detection

KW - Flywheel System

KW - Kernel Principal Component Analysis

KW - PCA

UR - https://www.scopus.com/pages/publications/85067846728

U2 - 10.1109/ITNEC.2019.8729163

DO - 10.1109/ITNEC.2019.8729163

M3 - 会议稿件

AN - SCOPUS:85067846728

T3 - Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019

SP - 425

EP - 432

BT - Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019

A2 - Xu, Bing

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019

Y2 - 15 March 2019 through 17 March 2019

ER -

Li GH, Li JC, Cao YN, Xu MQ, Xia KQ, Wei J et al. The flywheel fault detection based on Kernel principal component analysis. In Xu B, editor, Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 425-432. 8729163. (Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2019). doi: 10.1109/ITNEC.2019.8729163

The flywheel fault detection based on Kernel principal component analysis

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this