Fault Detection and Diagnosis for Industry Process Based on Support Vector Data Description

DOI

10.2991/eame-18.2018.77How to use a DOI?

Keywords

fault detection; fault diagnosis; support vector data description; mutual information

Abstract

A new approach for fault detection and diagnosis based on Support Vector Data Description (SVDD) has been proposed in this paper. Similar to the T2 and SPE statistic in principal components analysis (PCA), an appropriate nonlinear distance metric measured in feature space and threshold have been developed for fault detection. Once the fault is detected, fault diagnosis is then carried out using SVM based method. The fault diagnosis procedure is based on SVM and mutual information. The idea and effectiveness of the proposed algorithm are illustrated with respect to the simulation data collection from an illustrative example and the well-known Tennessee Eastman benchmark chemical process. Both the results show that the proposed approach works well to capture the underlying nonlinear process correlation thus providing a feasible and promising solution for nonlinear process monitoring.

Copyright

© 2018, 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  - Shuning Zhang
AU  - Hongyong Yang
AU  - Guanlong Deng
PY  - 2018/06
DA  - 2018/06
TI  - Fault Detection and Diagnosis for Industry Process Based on Support Vector Data Description
BT  - Proceedings of the 2018 3rd International Conference on Electrical, Automation and Mechanical Engineering (EAME 2018)
PB  - Atlantis Press
SP  - 364
EP  - 371
SN  - 2352-5401
UR  - https://doi.org/10.2991/eame-18.2018.77
DO  - 10.2991/eame-18.2018.77
ID  - Zhang2018/06
ER  -