Description

In the nuclear field, the integrity of welded nuclear components is assessed using ultrasonic non destructive testing (NDT) methods, providing essential information on the morphology and location of defects. However, their implementation remains complex in the case of austenitic stainless steels, whose welds, produced by the deposition of numerous passes, exhibit a heterogeneous and anisotropic microstructure, leading to divergence and attenuation of the wave beam. Numerical modeling of these microstructures, influenced by the nature of the materials and the process parameters, is therefore a major challenge for optimizing NDT methods.

This doctoral research aims to develop virtual microstructures on thick welded parts to enable the other partners of the ATALANTE project to simulate the propagation of ultrasonic waves and improve NDT tools. The challenge will thus be to model the formation of microstructures of interest on thick welds associated with a high number of passes.

Skills required

Engineer or Master's degree holder in the fields of materials, mechanics, or applied mathematics. Student interested in issues related to modeling and numerical simulation of physical phenomena in engineering science.

Bibliography

[1] C. Xue, N. Blanc et al., Structure and texture simulations in fusion welding processes – comparison with experimental data, Materialia 21, 2021, 101305

[2] T. Camus, D. Maisonnette, et al., Three-Dimensional Modeling of Solidification Grain Structures Generated by Laser Powder Bed Fusion, Materialia 30, 2023, 101804

[3] Z. Kong, G. Guillemot et al., Multiphysics simulation and microstructure prediction of coaxial wirelaser additive manufacturing process, Materialia 42, 2025, 102461

[4] Y. Zhang, G. Guillemot et al., Part-Scale Thermomechanical and Grain Structure Modeling for Additive Manufacturing: Status and Perspectives. Metals 2024, 14, 1173

[5] P. Belamri, H. Proudhon et al., Quaternion-based vision-transformer for polycrystalline EBSD scans pre-trained on large-scale synthetic data, Materials & Design 258, 114599, 2025

Keywords

Welding, Solidification, Microstructure, Modeling, Artificial Intelligence, Nuclear sector