Model based fault detection in process plant

In this paper, the application of neural network in detecting sensor failures is presented. The study was conducted on the Tennessee Eastman test problem. Faulty conditions were generated by imposing sensor failures in the reactor operation. Both single and multiple fault conditions were investigate...

Full description

Saved in:
Bibliographic Details
Main Authors: Leong, Wah Heng, Ahmad, Arshad
Format: Conference or Workshop Item
Language:English
Published: 1997
Subjects:
Online Access:http://eprints.utm.my/id/eprint/4715/1/LeongWahHeng1997_ModelBasedFaultDetection.pdf
http://eprints.utm.my/id/eprint/4715/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, the application of neural network in detecting sensor failures is presented. The study was conducted on the Tennessee Eastman test problem. Faulty conditions were generated by imposing sensor failures in the reactor operation. Both single and multiple fault conditions were investigated. The results revealed that a three layer feedforward network was capable of providing the intended function.