Signal resoration using dynamic neural network model for eddy current nondestructive testing

摘要

In this paper we propose to use a multi-frequency excitation and spectrogram (MFES) eddy current testing method and a neural network model to flaw signal restoration in case of presence of interferences from fasteners and subsurface structures. The extended experiments with various interfering structures were performed in order to verify the usability of the proposed restoration method. The selected results of numerical simulations are presented.