Comparison of the Predictions Relying on Coupled/Uncoupled Damage-Viscoplasticity Models for Creep Test Analyses

Abstract

To enhance the assessment of reactor pressure vessel life under severe accident, a cooperative research program is conducted between CEA, EDF, and Framatome, parts of which are supported by the European Commission. Within the framework of this program, a model based on Lemaitre and Chaboche with predictive capability in the field of viscoplastic flow, material damage, and failure has been generalized and implemented in CASTEM 2000 finite element code. In order to validate the model in mechanical situations featuring the same basic characteristics as the severe accident scenarios, analytical experiments are conducted on the RUPTHER facility: a thin shell tube is loaded with an internal pressure and submitted to an axial thermal gradient with elevated temperatures. The specific question of the interest of the coupled damage approach in the failure prediction is addressed. Comparisons of the predictions relying on coupled and uncoupled damage-viscoplasticity models show that, at higher temperatures, failure could be predicted through uncoupled damage evaluation provided the main nonlinear effects like large displacements and updated pressure are taken into account, while at lower temperatures coupling between damage and deformation is necessary. Examples based on some RUPTHER creep test predictions are given.

Microstructural features effect on the evolution of cyclic damage for polycrystalline metals using a multiscale approach

“…First, the damage is calculated at the Gauss integration points of the finite element discretization in the post-processing calculations of elasto-plasticity computation of quantities (stress and strains) required for the integration of the damage evolution law (equation 2). This type of coupling between plasticity and damage is weak, where the plasticity affects damage but damage does not induce any change in the plastic behavior, which is used in several works (Jiang and Sehitoglu, 1999; Lemaitre and Desmorat, 2005; Nicolas et al., 2001).…”

Experimental study of the fracture kinetics of a tubular 16MnNiMo5 steel specimen under biaxial loading at 900 and 1000°C. Application to the rupture of a vessel bottom head during a core meltdown accident in a pressurized water reactor

“…This weakening depends primarily on the sulfur concentration of the steel grade, but also on the thermal history (initial grain size produced by possible thermal treatments (Sestier, 1998), testing temperature) and, sometimes, on the loading rate or on the stress applied for the creep tests. In the context of the thermal treatments applied during the manufacturing of the vessels, in-depth characterization of the behavior of two such steel grades was carried out both in healthy conditions (Vereecke, 2004;Gentzbittel et al, 2007;Gentzbittel and Calapez, 2008;Tardif, 2009) and at rupture (Nicolas et al, 2001;Koundy et al, 2008a,b;Tardif et al, 2010). The Rupther grade is ductile at high temperature for all test temperatures and rates of solicitation.…”