The fabrication and subsequent in-service excitation of materials/structures invariably generates significant defects including cracks, voids and inclusions, spanning many length scales. The various technologies available for detecting and qualitatively describing such defects suffer from the introduction of various degrees inaccuracy when it comes to the quantification of defect geometry. This can have an adverse impact on modeling and understanding how the material/part/structure performance is effected by these defects. The main cause of this shortcoming is that aspects of the physical processes used to interrogate the material system, using monochromatic or polychromatic waves such as X-ray, mm-wave, or ultrasound, are not taken into account. These waves interact with the multiscale defect ensemble in a complex fashion that inevitably produces spurious “artifacts” in the resulting data, which cannot be removed via conventional data post-processing. These artifacts then introduce unacceptable levels of error when reconstructing defect geometry and computing the remaining lifespan of defect-bearing materials/structures. The present work introduces preliminary efforts towards a multiscale modeling and simulation framework for capturing the interactions of waves (such as X-rays) with materials bearing defect ensembles. It is shown that conventional approaches such as ray tracing are not adequate, and a more robust solution to the relevant wave equations utilizing the Finite Element discretization is employed. A general parameterization of defect geometries based on superquadratic functions is also introduced, and the interactions of defects modeled in this fashion with X-rays are investigated. It is also shown that this combination of parameterization and modeling techniques allows the recovery of true, artifact-free defect geometry utilizing classical inverse methods. The methodology is demonstrated using synthetic tomographic data, and the path forward to a more complete realization of this technology is outlined.

This content is only available via PDF.
You do not currently have access to this content.