NUMERICAL SIMULATION OF FUEL PELLET MELTING AND MELT CONVECTION UNDER SUSTAINED PCM SCENARIO IN SFR

B. Thilak
Computational Simulation Section Safety Engineering Division, Homi Bhabha National Institute, Indira Gandhi Centre for Atomic Research, Kalpakkam-603102, India

P. Mangarjuna Rao
Computational Simulation Section, Safety Engineering Division Fast Reactor Technology Group Indira Gandhi Centre for Atomic Research, Kalpakkam − 603102, India; Homi Bhabha National Institute, IGCAR, Kalpakkam, Tamil Nadu, India

B. K. Nashine
Computational Simulation Section, Safety Engineering Division Fast Reactor Technology Group Indira Gandhi Centre for Atomic Research, Kalpakkam − 603102, India

P. Selvaraj
Computational Simulation Section, Safety Engineering Division Fast Reactor Technology Group Indira Gandhi Centre for Atomic Research, Kalpakkam − 603102, India

DOI: 10.1615/IHMTC-2017.2710
pages 1935-1940

要約

The pre-clad failure in-pin molten fuel relocation phenomenon under the postulated sustained Sodium Fast Reactor (SFR) core Power Coolant Mismatch (PCM) scenario is considered to be an important source of negative reactivity. Such negative reactivity insertion could significantly mitigate the consequences of whole core meltdown accidents. In this regard, a numerical model has been developed to simulate the fuel axial motion from molten cavity inside the fuel pin in a single fuel pin. The model evaluates the fuel relocation phenomenon in two subsequent stages. In the first stage, a thermal analysis was carried out to evaluate the fuel melting rate for the given power ramp rate. Then the fuel melting rate is given as an input to the simulation of fuel relocation phenomenon using annular flow approach. The model evaluates transient temperature and velocity distribution due to phase change and the melt convection in the fuel pellet by solving heat conduction equation and incompressible Navier-Stokes equation respectively. The heat generated during the power transient is given as source term (volumetric heat generation rate) in heat conduction equation. The computation domain contains two moving interfaces namely: liquid-gas interface and solid-liquid interface. Both the interfaces are explicitly tracked by Arbitrary Lagrangian Eulerian (ALE) method. The spatial discretization is done using Galerkin Least Square finite element method and fully implicit multistep backward difference method was used for temporal discretization. The numerical model was validated by comparing the in-pile experimental data available in the literature for axial molten fuel distribution and melt fraction for irradiated fuel pellets corresponding to slow power ramp and good agreement was observed.