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

Abstract

Accurate measurement of the material parameters of composite in a nondestructive manner is of great significance for evaluating mechanical performance. This study proposes to use a genetic algorithm (GA) to reconstruct the stiffness matrix of carbon fiber reinforced polymer (CFRP) with array-guided wave (GW)-based GA. By comparing the numerically calculated GW dispersion curves with the experimental wave number-frequency contour calculated with a two-dimensional Fourier transform (2D-FFT), the matching coefficient is directly obtained as the objective function of the GA, avoiding the overhead of sorting out the respective GW modes. Then the measured stiffness matrix with tensile testing and the longitudinal wave in the unidirectional CFRP is compared with the reconstructed parameters from unidirectional, cross-ply, and quasi-isotropic CFRPs with the GA. For the four independent parameters, excluding C12, an average value of 11.62% for the maximum deviation is achieved among the CFRPs with three stacking sequences, and an average deviation of 11.03% in unidirectional CFRPs is achieved for the parameters measured with different methods. A further correction of fiber orientation results in a relative deviation of only 2.72% for the elastic modulus along the tensile direction, and an expansion of the GW frequency range for the GA narrows down the relative deviation of C12 to 3.9%. The proposed GW-based GA opens up a way of in situ and nondestructive measurement for the composite stiffness matrix.