Abstract High Performance Computing (HPC) has taken on an increasingly important role in recent years since it enables to model efficiently the complex behaviour of Systems, Components, and Structures (SSCs) and their interactions, and to perform simulations of the operational scenarios faster. Particularly in the nuclear field, HPC is currently used and expected to be utilized in nearly all areas, such as design optimization, safe and long-term safe operation, integrity evaluation, etc. The latter, as an example, requires an exhaustive SSCs assessment, which in turn requires higherfidelity models, both in terms of resolution and physics, increased knowledge of the physics of the phenomena or mechanisms affecting them, parametric studies, and better estimation of margins by reducing uncertainties and optimising operations. In this paper some case studies are presented and described. The first case consists in the modelling of creep behaviour of a fuel element. Finite Element Modelling (FEM) on hardware designed for parallel computation allows to simulate increasingly complex physical systems and reduce the computation time of the performed transient non-linear analyses. Results point out on the reliable prediction of creep behaviour obtained with a surrogate model. Understanding the thermo-mechanical properties of structural and nuclear materials and their changes provided essential for the plant safe operation as well as the Uncertainty Quantification (UQ). The second case deals with a thermal-hydraulics application in support of Generation IV (GEN IV) Liquid Metal Fast Reactors (LMFRs). Results from RANS calculations will be compared with experimental data for a selected steady-state condition occurring in a rod-bundle geometry. UQ analyses are especially performed using deterministic sampling techniques to assess how much the results are affected by uncertainties in the input parameters.
Read full abstract