Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors

Ibrahem M. Hussein; Zakariya Al-Hamouz

doi:10.1109/ASYU.2018.8554007

Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors

Ibrahem M. Hussein, Zakariya Al-Hamouz

King Fahd University of Petroleum & Minerals

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

8 Scopus citations

Abstract

Early detection of different faults will assist motor operation and prevent it from complete damage. Condition monitoring is extremely important to monitor the motor status and isolate it under failure conditions. This paper will propose a MATLAB® mathematical based neural network model for early detection of static eccentricity fault in line start permanent magnet synchronous (LSPMS) Motors. Under different combinations of fault-load conditions, the motor will be simulated to specify the characteristic of this fault. The line current will be utilized to extract the distinct principal components. The efficient selected components will be used as the input of neural network to recognize the percentage of occurrence as well as the fault's severity. The network will be trained over a specified range of static eccentricity degrees. Besides, it will be tested for unseen cases to qualify the effectiveness of the trained neural network. The testing results show a detection accuracy in a range between 95-98%.

Original language	English
Title of host publication	Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018
Editors	Tulay Yildirim, Buse Melis Ozyildirim
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781538677865
DOIs	https://doi.org/10.1109/ASYU.2018.8554007
State	Published - 29 Nov 2018

Publication series

Name	Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Keywords

Eccentricity faults
LSPMS Motor
Line current
Neural network
Principal Components

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications

Access to Document

10.1109/ASYU.2018.8554007

Cite this

Hussein, I. M., & Al-Hamouz, Z. (2018). Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors. In T. Yildirim, & B. M. Ozyildirim (Eds.), Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018 Article 8554007 (Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ASYU.2018.8554007

Hussein, Ibrahem M. ; Al-Hamouz, Zakariya. / Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors. Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018. editor / Tulay Yildirim ; Buse Melis Ozyildirim. Institute of Electrical and Electronics Engineers Inc., 2018. (Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018).

@inproceedings{caa7f83b3ba145c695773411373966f5,

title = "Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors",

abstract = "Early detection of different faults will assist motor operation and prevent it from complete damage. Condition monitoring is extremely important to monitor the motor status and isolate it under failure conditions. This paper will propose a MATLAB{\textregistered} mathematical based neural network model for early detection of static eccentricity fault in line start permanent magnet synchronous (LSPMS) Motors. Under different combinations of fault-load conditions, the motor will be simulated to specify the characteristic of this fault. The line current will be utilized to extract the distinct principal components. The efficient selected components will be used as the input of neural network to recognize the percentage of occurrence as well as the fault's severity. The network will be trained over a specified range of static eccentricity degrees. Besides, it will be tested for unseen cases to qualify the effectiveness of the trained neural network. The testing results show a detection accuracy in a range between 95-98\%.",

keywords = "Eccentricity faults, LSPMS Motor, Line current, Neural network, Principal Components",

author = "Hussein, \{Ibrahem M.\} and Zakariya Al-Hamouz",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.",

year = "2018",

month = nov,

day = "29",

doi = "10.1109/ASYU.2018.8554007",

language = "English",

series = "Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018",

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

editor = "Tulay Yildirim and Ozyildirim, \{Buse Melis\}",

booktitle = "Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018",

address = "United States",

}

Hussein, IM & Al-Hamouz, Z 2018, Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors. in T Yildirim & BM Ozyildirim (eds), Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018., 8554007, Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018, Institute of Electrical and Electronics Engineers Inc. https://doi.org/10.1109/ASYU.2018.8554007

Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors. / Hussein, Ibrahem M.; Al-Hamouz, Zakariya.
Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018. ed. / Tulay Yildirim; Buse Melis Ozyildirim. Institute of Electrical and Electronics Engineers Inc., 2018. 8554007 (Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018).

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

TY - GEN

T1 - Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors

AU - Hussein, Ibrahem M.

AU - Al-Hamouz, Zakariya

PY - 2018/11/29

Y1 - 2018/11/29

N2 - Early detection of different faults will assist motor operation and prevent it from complete damage. Condition monitoring is extremely important to monitor the motor status and isolate it under failure conditions. This paper will propose a MATLAB® mathematical based neural network model for early detection of static eccentricity fault in line start permanent magnet synchronous (LSPMS) Motors. Under different combinations of fault-load conditions, the motor will be simulated to specify the characteristic of this fault. The line current will be utilized to extract the distinct principal components. The efficient selected components will be used as the input of neural network to recognize the percentage of occurrence as well as the fault's severity. The network will be trained over a specified range of static eccentricity degrees. Besides, it will be tested for unseen cases to qualify the effectiveness of the trained neural network. The testing results show a detection accuracy in a range between 95-98%.

AB - Early detection of different faults will assist motor operation and prevent it from complete damage. Condition monitoring is extremely important to monitor the motor status and isolate it under failure conditions. This paper will propose a MATLAB® mathematical based neural network model for early detection of static eccentricity fault in line start permanent magnet synchronous (LSPMS) Motors. Under different combinations of fault-load conditions, the motor will be simulated to specify the characteristic of this fault. The line current will be utilized to extract the distinct principal components. The efficient selected components will be used as the input of neural network to recognize the percentage of occurrence as well as the fault's severity. The network will be trained over a specified range of static eccentricity degrees. Besides, it will be tested for unseen cases to qualify the effectiveness of the trained neural network. The testing results show a detection accuracy in a range between 95-98%.

KW - Eccentricity faults

KW - LSPMS Motor

KW - Line current

KW - Neural network

KW - Principal Components

UR - https://www.scopus.com/pages/publications/85059980506

U2 - 10.1109/ASYU.2018.8554007

DO - 10.1109/ASYU.2018.8554007

M3 - Conference contribution

AN - SCOPUS:85059980506

T3 - Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018

BT - Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018

A2 - Yildirim, Tulay

A2 - Ozyildirim, Buse Melis

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Hussein IM, Al-Hamouz Z. Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors. In Yildirim T, Ozyildirim BM, editors, Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018. Institute of Electrical and Electronics Engineers Inc. 2018. 8554007. (Proceedings - 2018 Innovations in Intelligent Systems and Applications Conference, ASYU 2018). doi: 10.1109/ASYU.2018.8554007

Neural Network Based Detection Technique for Eccentricity Fault in LSPMS Motors

Abstract

Publication series

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this