Proceedings of the 2nd International Conference On Systems Engineering and Modeling (ICSEM 2013)

Research on Fault Diagnosis Based on SVM

Authors
Wanling Li, Zhensheng Wang, Daquan Deng, Jun Tang
Corresponding Author
Wanling Li
Available Online April 2013.
DOI
10.2991/icsem.2013.104How to use a DOI?
Keywords
SVM, fault diagnosis, QPSO, improved QPSO, multi-class SVM
Abstract

In order to improve the fault diagnosis precision of electron system, a method based on wavelet packet transform and SVM was proposed. Fault diagnosis method based on SVM was researched on in this paper because of the complexity of electron system, difficulty of fault diagnosis method and special advantages of SVM. Wavelet packet transform is used to extract fault features from the signal of the circuit output voltage. The specific feature extraction method is introduced. Improved QPSO algorithm was proposed to improve the training speed and class precision of SVM. At last the method mentioned above was applied to a circuit. The result showed that this method was very good.

Copyright
© 2013, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

Volume Title
Proceedings of the 2nd International Conference On Systems Engineering and Modeling (ICSEM 2013)
Series
Advances in Intelligent Systems Research
Publication Date
April 2013
ISBN
978-94-91216-42-8
ISSN
1951-6851
DOI
10.2991/icsem.2013.104How to use a DOI?
Copyright
© 2013, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - CONF
AU  - Wanling Li
AU  - Zhensheng Wang
AU  - Daquan Deng
AU  - Jun Tang
PY  - 2013/04
DA  - 2013/04
TI  - Research on Fault Diagnosis Based on SVM
BT  - Proceedings of the 2nd International Conference On Systems Engineering and Modeling (ICSEM 2013)
PB  - Atlantis Press
SP  - 533
EP  - 536
SN  - 1951-6851
UR  - https://doi.org/10.2991/icsem.2013.104
DO  - 10.2991/icsem.2013.104
ID  - Li2013/04
ER  -