Application of the Hilbert-Huang signal processing to ultrasonic non-destructive testing of composite materials

Abstract

The purpose of the present paper is to provide the application of new signal processing method to the defects detection in multi-layered fibre-reinforced plastic pipe. The so-called Hilbert-Huang technique is based on direct extraction of the energy associated with the intrinsic time scales in the signal. To investigate the possibility of application of this method to multi-layered plastic pipe non-destructive testing, we analysed two pipe samples with artificial defects. The samples used in this investigation were one layer homogeneous pipe sample and three layers inhomogeneous pipe sample with internal fibreglass layer. For decomposition of the original signal we used data sifting to generate the intrinsic modes and apply the Hilbert transform to this modes. The decomposition results have shown that in a homogeneous one layer plastic pipe sample the information about defects represents the first mode or the combination of the first and second modes. In the inhomogeneous three-layers pipe sample detection of the defects is not simple. To determine the artificial defect in the second layer we used subtraction from the original signals the signal at the fixed point without artificial defect. This method gives a chance to determine the location of defects in space; measurement of the coordinates of the defect with necessary accuracy is still a problem.