Data-driven Exploration of Ductile Fracture Response

**Data-Driven exploration of Ductile fracture response**: - Réf **ABG-132651** - Sujet de Thèse - 25/06/2025 - Cifre - Ecole Centrale de Nantes - Lieu de travail- Nantes - Pays de la Loire - France - Intitulé du sujet- Data-Driven exploration of Ductile fracture response - Champs scientifiques- Sciences de l’ingénieur - Mots clés- DDI, DIC, ductile fracture, Plasticity **Description du sujet**: **Context and Objective** The French Alternative Energies and Atomic Energy Commission (CEA) plays a crucial role in ensuring the safety of nuclear installations such as Pressurized Water Reactors (PWRs). One key aspect of this mission is the analysis of severe accident scenarios, such as the Loss of Coolant Accident (LOCA). During such events, the sudden depressurization of the primary circuit can lead to structural failures, including pipe whip [1]. To accurately predict these phenomena and improve safety margins, a detailed understanding of the material behavior involved is essential. Stainless steel 316 is widely used in the nuclear industry—particularly for piping systems—due to its excellent mechanical properties and resistance to corrosion and irradiation. Its failure mode is known to be ductile, meaning it is preceded by significant plastic deformation. Such behavior can be influenced by various factors, including temperature, strain rate, and environmental conditions. Hence, the project aims to develop a high-performance computing (HPC)-scalable tool designed to construct a reliable database that will serve as a foundation for assessing and developing constitutive behavior models, particularly for large-strain plasticity and ductile fracture. The development of this database will rely on the recently developed Data-Driven Identification (DDI) paradigm, which enables the estimation of balanced stress fields from Digital Image Correlation (DIC) for arbitrary geometries and loading conditions. By doing so, complex mechanical tests can be analyzed without postulating a specific mathematical expression for the material response, while also enabling the extraction of relevant mechanical information in the immediate vicinity of the fracture. Promising results have already been obtained in several contexts: hyperelastic materials [2], and small-strain (rate-dependent) plasticity [3]. The main goal of this thesis is therefore to extend the DDI framework beyond its current limitations—including finite strains, strain localization, softening behavior, and 3D deformation states—until it becomes applicable to ductile failure experiments on 316L stainless steel. References [1] Serguei Potapov, Pascal Galon. Modelling of Aquitaine II pipe whipping test with the EUROPLEXUS fast dynamics code. Nuclear Engineering and Design, 2005, 235(17-19), pp. 2045-2054. [2] Dalémat, M., et al. "Measuring stress field without constitutive equation." Mechanics of Materials, 136 (2019): 103087. [3] Vinel, A., Seghir, R., Berthe, J., Portemont, G., & Réthoré, J. (2024). Experimental characterization of material strain-rate dependence based on full-field Data-Driven Identification. International Journal of Impact Engineering, 194, 105083. **Prise de fonction**: - 01/10/2025 **Nature du financement**: - Cifre **Précisions sur le financement**: - CEA **Présentation établissement et labo d'accueil**: - Ecole Centrale de Nantes Les chercheurs de Centrale Nantes déploient une approche interdisciplinaire qui s'organise autour de trois enjeux majeurs autour de la croissance et de l'innovation : l'usine du futur, la transition énergétique et l'ingénierie de la santé. **Intitulé du doctorat**: - Doctorat de Mécanique **Pays d'obtention du doctorat**: - France **Etablissement délivrant le doctorat**: - Ecole Centrale de Nantes A Master's degree (or equivalent) in engineering or mechanics (Bac+5) ; Strong knowledge in solid mechanics, continuum mechanics, and finite element methods ; Knowledge of DIC (Digital Image Correlation) and fracture mechanics would be a plus ; A strong interest in programming, particularly in Python ; A keen interest in both numerical and experimental mechanics ; Scientific rigor and curiosity ; Good written and oral proficiency in English.- 30/06/2025