Damage detection using sideband peak count in spectral correlation domain

Nonlinear ultrasonic techniques have been proven to be more sensitive to the presence of an early-stage damage than linear techniques. Among various nonlinear techniques, laser nonlinear wave modulation spectroscopy (LNWMS) utilizes a pulse laser to exert a broadband input and a damage on the target structure exhibits nonlinear wave modulation among various input frequency components. A sideband peak count (SPC) technique in the spectral frequency domain was proposed to estimate the damage-induced nonlinearity. In this study, the SPC operation is conducted in the spectral correlation domain so that noise has less influence on damage detection performance and a higher sensitivity to damage can be achieved. In addition, through spatial comparison of SPC over an inspection area, damage can be detected without relying on the baseline data obtained from a pristine condition. The performance of the proposed technique is validated using a numerical simulation performed on an aluminum plate with a simulated crack, and experiments performed on an aluminum plate with a fatigue crack and a carbon fiber reinforced polymer plate with delamination. (C) 2017 Elsevier Ltd. All rights reserved.