Title :
Neural network models of eddy current multi-frequency system for nondestructive testing
Author :
Chady, Tomasz ; Enokizono, Masato ; Sikora, Ryszard
Author_Institution :
Oita Ind. Res. Inst., Japan
fDate :
7/1/2000 12:00:00 AM
Abstract :
The purpose of this paper is to present new neural network models used for eddy current multi-frequency system simulation and flaws identification. Forward dynamic neural network models are proposed as the solution of this problem. The extended experimental verification of the developed technique is done and the results confirming usability of the models are presented
Keywords :
eddy current testing; flaw detection; inverse problems; neural nets; MFES method; computer simulation; eddy current multi-frequency system; flaw reconstruction; forward dynamic neural network model; inverse problem; nondestructive testing; Coils; Eddy currents; Frequency; Inspection; Inverse problems; Neural networks; Nondestructive testing; Signal processing; Spectrogram; Usability;
Journal_Title :
Magnetics, IEEE Transactions on
