Improved Unsupervised Learning Method for Material-Properties Identification Based on Mode Separation of Ultrasonic Guided Waves

Abstract

Numerical methods, including machine-learning methods, are now actively used in applications related to elastic guided wave propagation phenomena. The method proposed in this study for material-properties characterization is based on an algorithm of the clustering of multivariate data series obtained as a result of the application of the matrix pencil method to the experimental data. In the developed technique, multi-objective optimization is employed to improve the accuracy of the identification of particular parameters. At the first stage, the computationally efficient method based on the calculation of the Fourier transform of Green’s matrix is employed iteratively and the obtained solution is used for filter construction with decreasing bandwidths providing nearly noise-free classified data (with mode separation). The filter provides data separation between all guided waves in a natural way, which is needed at the second stage, where a more laborious method based on the minimization of the slowness residuals is applied to the data. The method might be further employed for material properties identification in plates with thin coatings/interlayers, multi-layered anisotropic laminates, etc.

Monitoring of Engineered Stones Used in Artwork Reproductions: Mechanical Characterization by Laser Vibrometry

“…The accurate Young's modulus measurement based on Rayleigh wave velocity using an empirical Poisson's ratio may also be possible, as described by Li and Feng [27]. In the case of orthotropic material characterization, other techniques using full wavefield data may require the use of 3D scanning laser Doppler vibrometry and improved machine learning methods for the identification of properties such as elastic stiffness [21,22].…”

“…Recent technological developments have led to performant 3D scanning laser Doppler vibrometers, which give access to both out-of-plane and in-plane vibrational velocity components, which combined with inverse problem approaches, allow for the identification of orthotropic elastic stiffness using 3D guided wavefield data [21]. Likewise, numerical methods, including machine-learning methods (such as clustering of multivariate data series), are also actively used in applications related to material property characterization by means of elastic guided wave propagation phenomena [22].…”

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

“…Eremin et al [26] manually picked the fundamental A 0 and S 0 modes, whose wavelengths were input into a GA for parameter reconstruction of unidirectional and cross-ply carbon fiber reinforced polymer (CFRP) laminate. Further, an unsupervised learning method based on the mode separation of a GW was proposed to characterize the material parameters of multiple isotropic materials [27]. Similarly, Okumura et al [28] adopted a diagonal loading technique to quickly extract a GW curve in an isotropic material.…”