Accurate Thickness Characterization of Glass Fiber-Reinforced Composites Using Terahertz Dispersion Compensation Method Based on Sparse Representation

Terahertz (THz) technology, a new nondestructive testing (NDT) approach with frequencies generally ranging from 100 GHz to 10 THz, has been widely used by virtue of its excellent penetration and high spatial–temporal resolution for non-metallic materials. However, in the process of characterizing the thickness of glass fiber-reinforced composite (GFRP) laminate along the THz propagation direction, the dispersion effect of the material will severely affect the accuracy of thickness measurement. In this work, in order to reduce the influence of the dispersion effect and improve thickness estimation accuracy, a THz dispersion compensation method based on sparse representation is proposed to accurately estimate the sparse impulse vector from the dispersive THz signal, with thickness measurement accomplished by the dispersion compensation process. Finally, a series of numerical simulations and experiments were implemented to verify the effectiveness of the proposed THz dispersion compensation method in thickness measurement. It is worth noting that the minimum estimated error in thickness is only 0.11%.