Guided Wave Scattering Analysis and Damage Diagnosis of Advanced Structural Materials

Ensuring the safety of structures is an important and challenging mission for engineers and infrastructure managers. One of the main reasons causing structural collapse is continuous accumulation of damage during the service life of the structure. Hence, damages are required to be detected as early as possible to prevent disasters. The thesis contributes to the development of using guided waves in damage detection for one- and two-dimensional waveguides, especially for advanced structural materials such as composite laminates. The thesis aims at improving the understanding of guided wave scattering characteristics at damages, and developing practical and reliable damage detection techniques. A probabilistic guided wave based damage detection technique is developed to locate and characterise a damage in one-dimensional waveguides. Instead of pinpointing a numerical value for damage detection, the probabilistic approach employs the Bayesian method, which quantifies the uncertainty associated with the damage detection results, to provide extra information for engineers in making decisions on condition based maintenance. For two-dimensional waveguides, comprehensive numerical and experimental studies are carried out to investigate guided wave scattering characteristics at delaminations and through holes in composite laminates using three-dimensional explicit finite element method. The studies provide improved insight into the guided wave scattering phenomena at damages. The findings are important for developing guided wave based damage detection techniques, and increasing the reliability of the damage detection through the optimisation of the excitation frequency and transducer network. Based on the understanding of guided wave scattering characteristics at damages in composite laminates, two guided wave based damage imaging methods are developed for real-time in situ monitoring of composite laminates. A damage localisation imaging method is developed to locate a damage in composite laminates. The method accounts for the effect of the angular group velocity distribution so it is general for composite laminates with different stacking sequences. In addition the damage localisation image indicates the potential damage area, which provides more information for the engineers in the damage detection. The proposed method is verified by a series of comprehensive numerical and experimental studies. It is demonstrated that the method is robust and reliable for damage localisation. Finally, the thesis demonstrates the feasibility of extending Lamb wave diffraction tomography for damage characterisation imaging in quasi-isotropic composite laminates. A two-phase imaging approach is developed to characterise a damage. In phase one the damage localisation imaging method is further refined and used to locate the damage with a higher degree of accuracy. Lamb wave diffraction tomography is then applied to characterise the damage in phase two. Examples are presented to demonstrate the performance of the proposed two-phase imaging approach in damage characterisation