Tuning of generalized k-omega turbulence model for prediction of heat transfer and pressure drop for helium cooled demo FW

Abstract

The HCPB blanket concept for EU-DEMO fusion reactors employs high-pressure (8 MPa) helium gas as coolant for the plasma facing first wall. Up-to-date estimations for the total maximum of heat flux go up to 0.73 MW/m². Higher short-term transient loads are possible. 60°-V-shaped rectangular ribs show high heat transfer and are thus the subject of the presented studies. Long term objectives of the qualification work for numerical simulations for the helium cooled First Wall channels of DEMO are to provide a consistent, fully validated and practically feasible (in terms of computing resources) numerical approach to provide simulations of complete rib enhanced FW channels/ full-size blanket components. The generalized k-omega (GEKO) turbulence model offers higher accuracy predicting heat transfer and pressure drop compared to other types of RANS (Reynolds-Averaged Navier-Stokes) models, even the Reynolds Stress Models (RSM). In this study, the GEKO model is used for simulating predictions for heat transfer and pressure drop. Three short channel segments with 60° V-shape rib configurations with rib heights of 0.4, 0.6 and 0.9 mm are studied at Re = 150,000. High-fidelity LES predictions serve as target data. The effects of the primary GEKO calibration parameters ( C sep , C mix , C nw , C corner ) are investigated. The predicting of Nu number and pressure gradient of the GEKO model is analysed and discussed.