Session: Research Posters

Paper Number: 119648

119648 - Defect Imaging With Adjoint Tomography Theory Based Reverse-Time Migration Through Impedance Gradient 

Ultrasound imaging (UI) is a popular technique for nondestructive evaluation (NDE) of components. One imaging method, reverse-time migration (RTM), locates inclusions, voids, and cracks by highlighting imperfections based on a predetermined imaging condition. This technique exploits forward propagated wavefields from the source and backward propagated wavefields from the receivers and evaluates the wavefield local space-time similarity to identify the presence of medium discontinuities. This similarity is determined by the aforementioned imaging condition. Different imaging conditions directly affect RTM’s capabilities to characterize defects within a domain. This study demonstrates multiple RTM imaging conditions based on adjoint tomography theory (ATT) proposed for NDE of metal specimens. These imaging conditions are formed based on the gradient maps of density, impedance, longitudinal velocity, and shear velocity using ATT. The results for each imaging condition are compared for well they highlight the contour of defects. 

Benchmarking is performed by creating a model based on an existing steel block specimen with varying hole sizes. From this model, RTM is performed on numerically generated source and receiver wavefields, which are replicated computationally by a spectral-element solver for seismic wave propagation. The data acquisition in this model is simulated for a 32-element transducer using a full matrix capture (FMC) in reflective mode. Data acquired in this format reflect the potential real-world limitations of only being able to measure data from one side of the investigated object. A Python-based computational framework organizes RTM processes for each transducer element's excitation and stacks together their images to produce a composite image. The key conclusion from the analysis of the resulting images is that using ATT based on the gradient of impedance produces a more highly resolved defect location and size characterization compared to the imaging conditions based on density, longitudinal velocity, and shear velocity.  

After the demonstration of the imaging technique using the numerical model, the technique is also applied to experimental data. The results from the numerical model are validated by the defect characterization of several holes in an aluminum block. The holes are positioned at varying distances from the top surface, forming the shape "AOS". The scanning setup consisted of a 128-element linear array transducer, with each element having a center frequency of 3.0 MHz. This transducer has a 99 mm aperture and is positioned at the top of the block. From this setup, the source and receiver wavefields are obtained for application for the ATT imaging condition. The source wavefield is generated by creating a numerical representation of the background wave velocity without the presence of defects in the material. The scattered receiver wavefield is obtained by backpropagation from data acquired with the described scanning setup. After generating the gradient maps using the source and receiver wavefields, the results are compared. It is concluded from the results that the gradient map of impedance produces best highlights the defects compared to other material property gradient maps. 

Presenting Author: John Day The University of Alabama

Presenting Author Biography: John is a Graduate Student at the University of Alabama. He is pursuing his Master’s Degree and PhD in Aerospace Engineering Mechanics. John has obtained a Bachelor’s in Aerospace Engineering and Mechanics in May 2022, during this time he was also able to obtain additional graduate credit hours through the University’s Accelerated Master’s Program. John joined the CISS lab to gain experience in data acquisition, simulation, programming, and working in a lab environment; his focus is structural health monitoring. He enjoys hobby programming, fishing, and spending time outdoors during his free time.

Authors:

John Day The University of Alabama
Jiaze He The University of Alabama
Jeffrey Shragge Colorado School of Mines
Paul Sava Colorado School of Mines
Erin Lanigan NASA Marshall Space Flight Center
Delphine Duquette NASA Marshall Space Flight Center
Gavin Dao Advanced OEM Solutions