Just-in-time Kernel Classifier for Online Process Diagnosis

DOI

10.2991/iccsee.2013.335How to use a DOI?

Keywords

fault detection and diagnosis, mode identification, kernel classifier, just-in-time learning

Abstract

A novel just-in-time kernel modeling method is proposed to online fault detection and diagnosis for chemical processes. The model parameters can be suitably selected using a fast cross-validation strategy. For a query sample, an online kernel classifier is constructed adaptively in a just-in-time manner for mode identification, i.e., fault detection and diagnosis, using the most relevant samples around it. The superiority of the proposed kernel classifier is demonstrated through a simulated chemical example, compared with the related method with fixed parameters.

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  - Yi Liu
AU  - Wenlu Chen
PY  - 2013/03
DA  - 2013/03
TI  - Just-in-time Kernel Classifier for Online Process Diagnosis
BT  - Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013)
PB  - Atlantis Press
SP  - 1337
EP  - 1340
SN  - 1951-6851
UR  - https://doi.org/10.2991/iccsee.2013.335
DO  - 10.2991/iccsee.2013.335
ID  - Liu2013/03
ER  -