Dynamic recrystallization full-field simulations: material parameters identification - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Dynamic recrystallization full-field simulations: material parameters identification

Baptiste Flipon
Marc Bernacki

Résumé

Accurate predictions of microstructural evolutions during complex thermomechanical processes is one of the major concerns in todays metallic parts elaboration. Microstructural properties, and in particular grain size distribution or average grain size are directly related to in-use mechanical properties. Knowing the successive microstructural states through the forming steps should then enable a better understanding of the relation between thermomechanical conditions and microstructure evolutions. In this context, microstructural predictions may help to head over a targeted microstructual state by optimizing the thermomechanical conditions or to have a better insight of the effect of certain process steps on the microstructure. A full-field finite element framework has been developped in order to model grain growth and recrystallization. The tracking of moving interfaces (here grain boundaries) is done by the use of level-set functions and phenomenological laws describe the hardening, recovery and nucleation mechanisms. This formulation enables a good prediction of microstructural evolutions at the polycrystal scale while keeping a limited computational cost. The material dependency of the models is taken into account through several material parameters. The later need to be accurately defined to be able to describe microstructural evolutions over the range of temperatures and strain rates occuring in the considered thermomechanical paths. A detailed and optimized material parameters identification procedure is presented and validated against several sets of thermomechanical conditions. From experimental datasets necessary for identification to the refinement of parameters through inverse analyses, the material parameters are discussed regarding sensitivity and statistical studies. Future evolutions of both models’ equations and of the identification procedure are presented as perspectives.
Fichier non déposé

Dates et versions

hal-03881498 , version 1 (01-12-2022)

Identifiants

  • HAL Id : hal-03881498 , version 1

Citer

Baptiste Flipon, Nathalie Bozzolo, Marc Bernacki. Dynamic recrystallization full-field simulations: material parameters identification. WCCM 2020 - 14th World Congress in Computational Mechanics, Jan 2021, Paris, France. ⟨hal-03881498⟩
11 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More