NEURAL NETWORK APPLICATION IN THE DEFECTOSCOPY

J. Grman

Abstract

At present avery perspective solution of indications classification in the defectoscopy is neural network application. One of the fields is classification of indications into classes, that are characterized by the signal shape, eventually by the signatures relating to the signal shape. The nondestructive defectoscopy of steam generator tubes of nuclear power plants by multifrequency eddy current method is the field, in that the use of classifiers, based on neural network is very perspective.<BR />The contribution concentrates on the choice of suitable neural network structures, on the choice of the training set and of the suitable representation of indications. The success of selected solutions is compared on real records of steam generator tubes with artificial defects and with imitation of construction element.

Keywords

eddy-current, neural network, expert system

Download

IMEKO-WC-2000-EXP-P526.pdf

DOI

-

Back to the proceedings