Applications of diffraction tomography on quantitative imaging using ultrasonic guided plate waves

Abstract

Critical infrastructures in transport and civil engineering applications can suffer from structural damages such as fatigue cracks and corrosion due to extreme loading or environmental conditions. Under the commonly used damage tolerant design philosophy, structures will not fail if damages are limited in size and severity. However, if damages are not detected at the early stage and properly controlled, they may grow in terms of their sizes and severities, which can eventually lead to catastrophic failures. Therefore, it is important to monitor the onset and progression of structural damages to ensure the safety of critical infrastructure assets. Structural health monitoring (SHM) technologies have shown potential to improve the management of key structural components. Among the possible SHM techniques, guided wave testing using piezoelectric transducers as transmitter and receiver elements is an attractive methodology, because it enables fast scanning of large areas. To assess the structural integrity, a quantitative imaging method is required. One of the methods that has attracted major attention is guided wave or plate wave diffraction tomography (PWDT) because it enables quantitative characterization of damages including location, size as well as severity. Although the performance of PWDT has been demonstrated to characterize for example remaining plate thickness for corrosion damages, its reliability and versatility using different system configurations have not yet been systematically evaluated. This thesis aims to evaluate the performance of PWDT, and to understand the underlying physical mechanisms affecting its performance. It also investigates the feasibility of extending PWDT to other fundamental plate wave modes beyond the first flexural wave mode. Hence, the three major objectives of the thesis are: i) to quantitatively identify the applicable range of PWDT for reliable and accurate damage reconstructions for different wave modes; ii) to understand the physical mechanisms that affect the performance of PWDT reconstruction; and iii) to investigate the feasibility of developing a baseline-free imaging technique by considering mode conversion effects in the framework of PWDT. To achieve this comprehensive and fundamental understanding of PWDT and systematically evaluate its imaging quality, its performance to image circular blind hole damages in isotropic plates is evaluated. This requires the calculation of the scatter fields corresponding to the three fundamental plate wave modes using analytical models and including all possible mode conversion effects. In the case of incoming and scattered fundamental flexural waves the damage characteristics, including size and severity are accurately estimated as long as the fundamental limitations of the imaging algorithm are fulfilled, i.e. Born approximation and diffraction limit. Hence the accuracy is generally lower for larger and deeper defects. For the fundamental symmetric Lamb wave mode, the imaging quality is superior for large and severe damages. However, the quality of the imaging results shows large variability, which is controlled by the nature of the scatter field. Baseline free reconstruction of damage severity using mode conversion scatter effects is mostly unsuccessful because of the inherent bandpass filter character of the reconstruction algorithm. The PWDT algorithm is extended to allow near field measurements and point source excitations, which are essential to make the method feasible for practical applications. The ability of this modified plate wave diffraction tomography methodology of accurately reconstruct circular and non-circular blind hole damages are demonstrated using numerically generated and experimentally measured scatter field data. The experimental results show that for practical applications additional factors, in particular signal-to-noise ratio effects become very important.