ANN Optimized by ICSA Used in Fault Diagnostics

DOI

10.2991/meic-14.2014.37How to use a DOI?

Keywords

component;Fault diagnostics;Artificial neural network;Immune clonal selection;True detection rate;parameter optimization

Abstract

Fault diagnostics is to distinguish the current state of the equipment is in normal or abnormal. It can be seen as a problem of multi-class classification. To improve the performance of classification, this paper presents a novel method for fault diagnosis. In this method, we synthetically applied immune clonal selection algorithm and artificial neural network technology to the fault diagnosis of steam-turbine generator. The method consists of two stages. Firstly, the parameters of artificial neural network were optimized by immune clonal selection algorithm. Then an artificial neural network classifier with parameter optimization is constructed and used to identify the fault of steam-turbine generator. The experimental result using the a steam turbine fault diagnosis dataset shows that the fault diagnostics based on artificial neural network optimized by immune clonal selection algorithm can give higher recognition accuracy than other traditional methods.

Copyright

© 2014, 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  - Zhenguo Chen
AU  - Xiaoju Wang
AU  - Liqin Tian
PY  - 2014/11
DA  - 2014/11
TI  - ANN Optimized by ICSA Used in Fault Diagnostics
BT  - Proceedings of the 2014 International Conference on Mechatronics, Electronic, Industrial and Control Engineering
PB  - Atlantis Press
SP  - 166
EP  - 169
SN  - 2352-5401
UR  - https://doi.org/10.2991/meic-14.2014.37
DO  - 10.2991/meic-14.2014.37
ID  - Chen2014/11
ER  -