Research on Fault Diagnosis Based on SVM

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)

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  -