Title :
Location of plural defects in conductive plates via neural networks
Author :
Morabito, Francesco Carlo ; Campolo, Maurizio
Author_Institution :
Dipatimento di Ingegneria Elettronica e Matematica Applicata, Calabria Univ., Italy
fDate :
5/1/1995 12:00:00 AM
Abstract :
This paper treats an inverse electrostatic sample problem which is very similar to a real nondestructive testing (NDT) problem. The focus of the paper is on the use of an artificial neural network (ANN) approach. The method here presented aims at detecting and characterising plural defects. The experimental results show the validity of the proposed processing
Keywords :
electrical engineering computing; electrostatics; neural nets; nondestructive testing; artificial neural network; conductive plates; inverse electrostatic sample problem; neural networks; nondestructive testing; plural defects; Artificial neural networks; Electromagnetics; Electrostatics; Inspection; Intelligent networks; Inverse problems; Magnetic field measurement; Neural networks; Shape measurement; Testing;
Journal_Title :
Magnetics, IEEE Transactions on
