Neural network recognition of nuclear power plant transients

Abstract

The objective of this report is to describe results obtained during the first year of funding that will lead to the development of an artificial neural network (ANN) fault - diagnostic system for the real - time classification of operational transients at nuclear power plants. The ultimate goal of this three-year project is to design, build, and test a prototype diagnostic adviser for use in the control room or technical support center at Duane Arnold Energy Center (DAEC); such a prototype could be integrated into the plant process computer or safety - parameter display system. The adviser could then warn and inform plant operators and engineers of plant component failures in a timely manner. This report describes the work accomplished in the first of three scheduled years for the project. Included herein is a summary of the first year's results as, well as individual descriptions of each of the major topics undertaken by the researchers. Also included are reprints of the articles written under this funding as well as those that were published during the funded period.

Keywords

This publication has 0 references indexed in Scilit: