Luiz Antonio Negro Martin Lopez
Mauá Institute of Technology, Brazil
Title: Eddy current signal processing applied to aircraft fuel tubes maintenance
Biography
Biography: Luiz Antonio Negro Martin Lopez
Abstract
Large aircraft have a hundred meters long of fuel tubes that must be inspected during maintenance work services. Th e layout of those tubes is noticeable complex and the access for inspection is frequently very diffi cult. Eddy current non-destructive testing is one of the most important inspection methods and widely used to inspect the tubes by inserting inspection probes. Eddy current signals generated during the maintenance inspection work, in general, contain noise, which hinders the signal analysis and reduces the reliability of the inspection conclusion. Namely one of the noises present in the signals is the probe wobble eff ect, which is caused by the necessary slack between the probe and the fuel tube. In this work, Wavelet Transforms (WT) are used for de-noising the probe wobble Eddy current signals. WT provides defect localization across the tube length making it better than other analysis methods such as the Fourier Transform. However, WT involves wavelet and coeffi cients selection in the user level of analysis. Th is work presents probe wobble signal de-noising examples with several wavelet and coeffi cients applied to Eddy current signals generated by a Zetec MIZ-17ET equipment on an inspection of a 19.05mm diameter stainless steel fuel tube with known artifi cial defects. Th e probe wobble signal de-noising off ers reliable results and is
a promising method as it allows a fast removal of Eddy current noise maintaining the essential signal information.