Damage identification of laminated beams based on wavelet multiscale fusion of high-frequency vibration response

Damage such as cracking and delamination will inevitably occur in laminated composite structures. It has been shown that higher-order mode shapes are more sensitive to localized damage, but the current research on damage identification of laminated beams based on higher-order mode shapes is insufficient. To address this issue, this paper proposes a damage detection method based on higher-order mode shapes, which can accurately recognize multiple damages of laminated beams under noise conditions. The enhanced spectral element method proposed in this study is used to separately calculate the higher-order mode shapes of cantilevered laminated beams with cracks and those with delamination under Gaussian white noise conditions. The stationary wavelet packet transform (SWPT) is introduced to decompose the higher-order mode shapes into eight uniform frequency sub-bands with shift invariance at low and high frequencies, and then the multiplication operation is performed on the neighboring SWPT sub-bands, which is able to significantly strengthen the damage-induced peak signals and suppress the noise. By normalizing the multiscale product curves derived from different mode shapes to be at the same data level, the node effect is successfully avoided through the multicurve data fusion strategy. Numerical validation on the above two types of damage cantilever laminated beam models shows that the proposed method not only effectively mitigates the interference of nodal effect, but also accurately presents stable singular peaks at the damage sites. This enables precise damage localization and provides a fast and effective solution for detecting multiple damages in laminated beams under noisy conditions, demonstrating promising application prospects.