Research and Review Journal of Nondestructive Testing (Aug 2023)

Determining ultrasonic propagation effective properties in complex heterogeneous media through microstructure-scale simulation

  • Vincent Dorval,
  • Nicolas Leymarie,
  • Alexandre Imperiale,
  • Edouard Demaldent,
  • Zakaria Aghenzour,
  • Pierre-Emile Lhuillier

DOI
https://doi.org/10.58286/28132
Journal volume & issue
Vol. 1, no. 1

Abstract

Read online

Ray-based methods allow for fast computations of ultrasonic propagation in large components, and can be coupled with diffraction models to provide full simulations of NDE inspections. However, they do not account for complex interactions with highly heterogeneous propagation media, which can have a significant impact on inspection performances. In contrast, microstructure-scale finite element simulations consider these interactions but are too computationally intensive for large-scale simulations. The work presented in this communication aims at combining the advantages of these two approaches. On the one hand, finite element simulations for small volumes of the microstructure are used to determine parameters such as effective velocities, attenuations, and scattering coefficients. On the other hand, a ray-based model uses these data to compute wave propagation over distances that are large compared to both wavelengths and characteristic microstructure sizes. A dedicated simulation module has been implemented in a development version of the CIVA software. It generates random realizations of microstructures for given set of parameters, runs finite element computations, and post-processes their results to yield estimations of the properties of the macroscopic effective medium. The volumes considered by the finite element model are small enough to allow for 3D computations. Results were obtained for various types of microstructures, describing metals or concrete. This communication focuses on steel applications that were the focus of collaborative studies between CEA and EDF. The approach is promising, and contributes to bridging the gap between microstructure-scale modelling and larger scale simulations.