Use of Autoassociative Neural Networks for Signal Validation

Authors:
J. Wesley Hines;Robert E. Uhrig;Darryl J. Wrest
Affiliations:
Nuclear Engineering Department, The University of Tennessee, Knoxville, Tennessee 37996, USA/ e-mail: hines@utkux.utk.edu ruhrig@utk.edu;Nuclear Engineering Department, The University of Tennessee, Knoxville, Tennessee 37996, USA/ e-mail: hines@utkux.utk.edu ruhrig@utk.edu;Systems Health Management Group, Honeywell Technology Center, Minneapolis, Minnesota 55418, USA/ e-mail: dwrest@htc.honeywell.com
Venue:
Journal of Intelligent and Robotic Systems
Year:
1998

Citing 2
Cited 2

Practical neural network recipes in C++

Practical neural network recipes in C++
Neuro-Adaptive Process Control: A Practical Approach

Neuro-Adaptive Process Control: A Practical Approach

Consensus self-organized models for fault detection (COSMO)

Engineering Applications of Artificial Intelligence
Multiple sensor fault diagnosis by evolving data-driven approach

Information Sciences: an International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recently, the use of Autoassociative Neural Networks (AANNs) to performon-line calibration monitoring of process sensors has been shown to not onlybe feasible, but practical as well. This paper summarizes the results ofapplying AANNs to instrument surveillance and calibration monitoring atFlorida Power Corporation’s Crystal River #3 Nuclear Power Plant andat the Oak Ridge National Laboratory High Flux Isotope Reactor. In bothcases sensor drifts are detectable at a nominal level of 0.5%instrument’s full scale range. This paper will discuss the selectionof a five layer neural network architecture, a robust training paradigm, theinput selection criteria, and a retuning algorithm.