Sparse wavenumber analysis of guided wave based on hybrid Lasso regression in composite laminates

Abstract

The guided wave is an efficient and reliable tool for the structural health monitoring (SHM) of the composite laminates. In the guided wave-based SHM methods, extracting the dispersion curves is essential for integrity evaluation. In this study, a sparse wavenumber analysis based on hybrid least absolute shrinkage and selection operator (Lasso) regression is proposed to extract the dispersion curves in the frequency–wavenumber distribution (FKD) for the composite laminate. The hybrid Lasso regression model is constructed based on the guided wave propagation mechanism. Considering that responses of some wave modes are very weak at specific frequencies due to the guided wave attenuation in the composite laminates, the group-sparsity and continuity regularizations are imposed in this model to improve frequency–wavenumber resolution and remove noises. Only few sensors are required for the proposed method to extract the dispersion curves. Both the simulation and the experiment are used to verify the effectiveness of the proposed method. Furthermore, the material property of the composite laminate in the experiment is non-destructively estimated by using the dispersion curves extracted by the proposed method.

Reconstruction of Composite Stiffness Matrix with Array-Guided Wave-Based Genetic Algorithm

“…Zhao et al [23] first converted the experimentally measured group velocities of a GW fundamental mode in the 360-degree circumferential directions into phase velocities, manually selected a specific mode, and then used the error between the theoretical and experimental phase velocities to construct the OF. Hu et al [30] further developed the hybrid Lasso regression to improve the experimental signal-to-noise ratio of a GW along one single direction for the reconstruction of the material stiffness matrix.…”