CFD Simulation of Counter-current Flow Limitations in a Full Scale Pressurized Water Reactor (PWR) Hot Leg

The counter-current gas-liquid two-phase flow in the hot leg of a pressurized water reactor (PWR) has received a special attention for safety regulation in the nuclear industry. One hypothetical scenario is a loss-of-coolant-accident (LOCA) in a PWR, which is caused by the damage at any position of the primary circuit. The analytical simulation of this phenomenon is an essential element to understand safety-related issues in nuclear power plants. It is expected that the introduction of computational fluid dynamics (CFD) tools will enhance the accuracy of the simulation predictions compared to the established one-dimensional thermal hydraulic analyses. Here CFD allows substituting geometry-dependent empirical closure relations with more physically justified closure laws that are formulated at the scale of the structures of the gas–liquid interface.
This paper presents a CFD simulation on the counter-current flow limitation (CCFL) phenomena in a full scale PWR hot leg of Upper Plenum Test Facility (UPTF) Test case No 11 by using a commercial CFD code of ANSYS CFX 13.0, based on the finite volume method for an Euler-Euler model. The grid consist 29,100 hexahedral elements and 30,102 nodes. The transient calculations were carried out using a gas/liquid inhomogeneous multiphase flow model coupled with a shear stress transport (SST) turbulence model. A new formulation of an interfacial drag coefficient was implemented inside the Algebraic Interfacial Area Density (AIAD) model (Höhne, 2010) into the three-dimensional (3-D) computational fluid dynamics (CFD) code.
To demonstrate the feasibility of the developed drag coefficient in AIAD model, the computed main parameters of the selected test case were compared with experimental data. The results indicated that the quantitative agreement between calculation and experimental data was obtained. This means that the AIAD model combined with the new drag force formulation is a promising way to simulate the phenomena in the frame of an Euler-Euler approach.