Perspectives in the Simulation of PTS Scenarios: Form RANS to V-LES & LES and from Phase-averaged to Interface Tracking Methods

Abstract

We report recent LEIS results of steam condensation under violent turbulence and phase- change conditions in the cold leg of the COSI experiment [1], completing earlier efforts on the subject [2, 3]. The LEIS framework, short for Large-Eddy & Interface Simulation [2], is based on interface tracking methods (level set approach) combined with large-scale prediction of turbulence, where super-grid scale turbulence and interfaces are directly solved whereas the sub-scale parts are modelled. Large scale prediction of turbulence is interpreted here broadly, including either the LES or the V-LES (Very Large-Eddy Simulation) variant [4], in which the flow-dependent cut-off filter is larger and independent from the grid, in contrast to LES where it depends on the grid. The interfacial phase-change heat transfer model used is DNS- based, built on the Surface Divergence theory modified to account for high-shear flow conditions using a scale-separation approach. The model is found to return good results: VLES is superior to RANS, but LES seems to remain quite expensive to reach statistically converged results, while it delivers instantaneous local flow-events with valuable details.