Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches

Yichuan Fu; Zhiwei Gao; Haimeng Wu; Xiuxia Yin; Aihua Zhang

doi:10.1109/PEAS53589.2021.9628697

Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches

Yichuan Fu, Zhiwei Gao, Haimeng Wu, Xiuxia Yin, Aihua Zhang

Mathematics, Physics and Electrical Engineering

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

7 Citations (Scopus)

47 Downloads (Pure)

Abstract

Data-driven fault classification for power converter systems has been taking more into considerations in power electronics, machine drives, and electric vehicles. It is challenging to classify the different topologies of faults in the real time monitoring control systems. In this paper, a data-driven and supervised machine learning-based fault classification technique is adopted by combining and consolidating with Expectation Maximisation Principal Component Analysis (EMPCA) and Support Vector Machine (SVM) to substantiate the availability of fault classification. The proposed methodology is applied to the non-inverting Buck–Boost DC–DC power converter systems subjected to the incipient fault and serious fault, respectively. Finally, the feasibility of the approach is validated by intensive simulations and comparison studies.

Original language	English
Title of host publication	2021 IEEE 1st International Power Electronics and Application Symposium (PEAS)
Place of Publication	Piscataway, US
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	9781665413602
ISBN (Print)	9781665413619, 9781665413596
DOIs	https://doi.org/10.1109/PEAS53589.2021.9628697
Publication status	Published - 13 Nov 2021
Event	1st IEEE International Power Electronics and Application Symposium - Shanghai, China Duration: 12 Nov 2021 → 15 Nov 2021

Conference

Conference	1st IEEE International Power Electronics and Application Symposium
Abbreviated title	IEEE PEAS'2021
Country/Territory	China
City	Shanghai
Period	12/11/21 → 15/11/21

Keywords

Data-driven
Fault classification
expectation maximisation principal component analysis
support vector machine
Buck–Boost DC–DC power converters

Access to Document

10.1109/PEAS53589.2021.9628697

Conference-IEEE-PEAS-2021-Final-Submission-Yichuan-Fu-University-of-Northumbria-at-NewcastleAccepted author manuscript, 989 KB

Cite this

Fu, Y., Gao, Z., Wu, H., Yin, X., & Zhang, A. (2021). Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches. In 2021 IEEE 1st International Power Electronics and Application Symposium (PEAS) Article 9628697 IEEE. https://doi.org/10.1109/PEAS53589.2021.9628697

@inproceedings{cae74015626b4573931c11747cac3b8e,

title = "Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches",

abstract = "Data-driven fault classification for power converter systems has been taking more into considerations in power electronics, machine drives, and electric vehicles. It is challenging to classify the different topologies of faults in the real time monitoring control systems. In this paper, a data-driven and supervised machine learning-based fault classification technique is adopted by combining and consolidating with Expectation Maximisation Principal Component Analysis (EMPCA) and Support Vector Machine (SVM) to substantiate the availability of fault classification. The proposed methodology is applied to the non-inverting Buck–Boost DC–DC power converter systems subjected to the incipient fault and serious fault, respectively. Finally, the feasibility of the approach is validated by intensive simulations and comparison studies.",

keywords = "Data-driven, Fault classification, expectation maximisation principal component analysis, support vector machine, Buck–Boost DC–DC power converters",

author = "Yichuan Fu and Zhiwei Gao and Haimeng Wu and Xiuxia Yin and Aihua Zhang",

note = "The authors would like to thank the research support from the National Nature Science Foundation of China (NNSFC) under Grant 61673074, and the Faculty of Engineering and Environment, Northumbria University, Newcastle upon Tyne, Tyne and Wear, NE1 8ST, United Kingdom.; 1st IEEE International Power Electronics and Application Symposium , IEEE PEAS'2021 ; Conference date: 12-11-2021 Through 15-11-2021",

year = "2021",

month = nov,

day = "13",

doi = "10.1109/PEAS53589.2021.9628697",

language = "English",

isbn = "9781665413619",

booktitle = "2021 IEEE 1st International Power Electronics and Application Symposium (PEAS)",

publisher = "IEEE",

address = "United States",

}

Fu, Y, Gao, Z , Wu, H, Yin, X & Zhang, A 2021, Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches. in 2021 IEEE 1st International Power Electronics and Application Symposium (PEAS)., 9628697, IEEE, Piscataway, US, 1st IEEE International Power Electronics and Application Symposium , Shanghai, China, 12/11/21. https://doi.org/10.1109/PEAS53589.2021.9628697

Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches. / Fu, Yichuan; Gao, Zhiwei ; Wu, Haimeng et al.
2021 IEEE 1st International Power Electronics and Application Symposium (PEAS). Piscataway, US: IEEE, 2021. 9628697.

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

TY - GEN

T1 - Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches

AU - Fu, Yichuan

AU - Gao, Zhiwei

AU - Wu, Haimeng

AU - Yin, Xiuxia

AU - Zhang, Aihua

N1 - The authors would like to thank the research support from the National Nature Science Foundation of China (NNSFC) under Grant 61673074, and the Faculty of Engineering and Environment, Northumbria University, Newcastle upon Tyne, Tyne and Wear, NE1 8ST, United Kingdom.

PY - 2021/11/13

Y1 - 2021/11/13

N2 - Data-driven fault classification for power converter systems has been taking more into considerations in power electronics, machine drives, and electric vehicles. It is challenging to classify the different topologies of faults in the real time monitoring control systems. In this paper, a data-driven and supervised machine learning-based fault classification technique is adopted by combining and consolidating with Expectation Maximisation Principal Component Analysis (EMPCA) and Support Vector Machine (SVM) to substantiate the availability of fault classification. The proposed methodology is applied to the non-inverting Buck–Boost DC–DC power converter systems subjected to the incipient fault and serious fault, respectively. Finally, the feasibility of the approach is validated by intensive simulations and comparison studies.

AB - Data-driven fault classification for power converter systems has been taking more into considerations in power electronics, machine drives, and electric vehicles. It is challenging to classify the different topologies of faults in the real time monitoring control systems. In this paper, a data-driven and supervised machine learning-based fault classification technique is adopted by combining and consolidating with Expectation Maximisation Principal Component Analysis (EMPCA) and Support Vector Machine (SVM) to substantiate the availability of fault classification. The proposed methodology is applied to the non-inverting Buck–Boost DC–DC power converter systems subjected to the incipient fault and serious fault, respectively. Finally, the feasibility of the approach is validated by intensive simulations and comparison studies.

KW - Data-driven

KW - Fault classification

KW - expectation maximisation principal component analysis

KW - support vector machine

KW - Buck–Boost DC–DC power converters

U2 - 10.1109/PEAS53589.2021.9628697

DO - 10.1109/PEAS53589.2021.9628697

M3 - Conference contribution

SN - 9781665413619

SN - 9781665413596

BT - 2021 IEEE 1st International Power Electronics and Application Symposium (PEAS)

PB - IEEE

CY - Piscataway, US

T2 - 1st IEEE International Power Electronics and Application Symposium

Y2 - 12 November 2021 through 15 November 2021

ER -

Fu Y, Gao Z , Wu H, Yin X, Zhang A. Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches. In 2021 IEEE 1st International Power Electronics and Application Symposium (PEAS). Piscataway, US: IEEE. 2021. 9628697 doi: 10.1109/PEAS53589.2021.9628697

Data-Driven Fault Classification for Non-Inverting Buck–Boost DC–DC Power Converters Based on Expectation Maximisation Principal Component Analysis and Support Vector Machine Approaches

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this