Ultrasonic imaging of microdamage using time reversal techniques based on nonlinear elastic wave spectroscopy.

During the last decade, Nonlinear techniques based on Time Reversed Acoustics (NLTRA) have proven their efficiency and extreme sensitivity for Nondestructive Testing (NDT) of materials containing damage with a nonlinear behavior. NLTRA-techniques combine Nonlinear Elastic Wave Spectroscopy (NEWS), which is a collective term for techniques aiming to detect various kinds of nonlinear behavior in response signals, with the principle of Time Reversal (TR). This combination results in more efficient and sensitive techniques to localize and image incipient damage in the form of microcracks, delaminations, etc.Despite the success of NLTRA, these techniques share an important shortcoming. In a purely experimental application, only near-surface defects can be localized, since it is required to position a detector near the defect region. Detection of embedded defects is therefore only possible in a hybrid experimental/numerical approach. The main obstacle is to achieve a reliable numerical model, with a high accuracy on the material parameters such that the propagation of multiple reverberating signals that are typical for NLTRA can be calculated without losing the focus quality in the TR-process. Apart from the density and the geometry, it is critical to know the velocities and/or velocity distribution of the sample under consideration.In view of future hybrid ultrasonic NDT applications, we first developed a new automatic procedure for the determination of the (isotropic) longitudinal and shear material velocities, based on traditional TR-principles, and the signal dilatation and contraction method developed by Scalerandi et al. The method requires a single recorded signal as input, together with a (crude) estimation of the longitudinal velocity. Density and geometry are assumed to be known exactly, as well as the position of excitation and recording. Iteration of this TR based procedure allows to determine the longitudinal and shear velocities of the sample material with an accuracy better than 0.5%. Results are presented for a 3D numerical 'blind' test case and for two applications involving experimentally recorded data.Further, a NLTRA-technique was investigated that consists of the following two steps. In the first step, TR is used to focus high levels of energy in a small area of a medium in order to activate possible defects, and in a second postprocessing step, information regarding the nonlinear contribution is extracted from the reponse signals. Numerical simulations are reported showing the potential of a combination consisting of dual energy reciprocal time reversal and nonlinearity filtering using the scaling subtraction method. The method is applied to the detection of planar near-surface defects parallel to the surface in a 2D domain. The results are evaluated for sweep excitation at different frequency ranges; for point-like receivers as well as extended transducers, and for in-plane as well as out-of-plane focusing. The observable nonlinear response at the surface is linked to an effective nonlinearitywithin the medium based on the defect geometry and the distribution of the local stresses.The final problem that is considered in this thesis deals with the difficulty of imaging multiple masked scatterers. Inherent limitations of the traditional time reversal process in the case of multiple sources or scatterers make it difficult to distinguish the scattering sources individually. The selective source reduction (SSR) method employs a subtraction technique to selectively suppress in amplitude (and ideally eliminate) a time reversed focal signal that is masking another focus. In previous work, Scalerandi et al. and Anderson et al. successfully applied the SSR method to identify masked primary sources in a fully linear medium. Here, we extend the capabilities of the SSR method to deal with scattering caused by embedded defects. We call this new method SSR-NLTRA (Selective Source Reduction based on Nonlinear Time Reversed Acoustics). In the extended approach, the contribution of all primary and linear sources is first eliminated by means of the scaling subtraction method. Subsequently, the SSR(-TRA) method is applied to the remaining nonlinear content of the signals. We show by means of 2D wave propagation simulations that the new method can be applied iteratively to successfully image multiple masked nonlinear defects.

Publication year:2011

Accessibility:Closed