Abstract
This thesis presents fine-mesh multiphysics methodologies and algorithms for numerical predictions of the behavior of Light Water Reactor (LWR) cores. The multiphysics aspects cover the distribution of neutrons, the fluid flow of the coolant and the conjugate heat transfer between the solid fuel pins and the fluid coolant. The proposed schemes are aimed at fine-mesh coupled effects, directly resolving the interdependencies of the different fields on the finest scales of the computations.
The solver is developed for both steady-state and transient LWR scenarios. For the steady-state simulations, the neutronics is solved both by the lower order, diffusion equation and the higher order, discrete ordinate transport method, and for transient cases by the former. The thermal-hydraulic solver is based on a computational fluid dynamics (CFD) approach. The implementation utilizes a finite volume method (FVM) computational framework, and to achieve feasible computational times, high performance computing (HPC) aspects such as parallelization by domain decomposition are considered.
The implemented tool is applied to cases of parts of a fuel assembly, analyzing systems of up to 15 × 15 fuel pins and succesfully resolving sub-pin resolution of all fields. Furthermore, the transient fine-mesh neutronic solver is verified based on a novel scheme utilizing the system response to a local perturbation.
In addition, the multiphase flow problem encountered in Boiling Water Reactors (BWRs) is studied. First, the transport of bubbles under subcooled boiling conditions is simulated based on a population balance approach. The novel formulation is shown to increase the computational efficiency and to capture a large range of bubbles sizes with few degrees of freedom. Second, the typical EulerianEulerian approach for two-phase flow is studied from a stability and dynamics perspective. The latter investigations highlight the complexity of the two-fluid formulation and indicate the spontaneous emergence of meso-scale void structures under adiabatic conditions.