Coarse-mesh-based Reduced-order Package for Multiphysics Simulation of Nuclear Thermal Propulsion Reactor Core

Among the various proposed space propulsion technologies, the nuclear thermal propulsion (NTP) system has been identified as the technology of choice for NASA’s mission to Mars due to its high efficiency in fuel performance. Most of the legacy NTP engine designs from the NASA’s historical Rover/NERVA program relied on highly enriched (above 90% enrichment) uranium (HEU) fuel. Recent efforts focused on designing a high-efficiency engine that relies on high-assay low-enriched uranium fuel (HALEU with ~19.75% enrichment). To achieve high specific impulse and thrust to weight values, these new HALEU-based generations of NTP would require geometrical modifications associated with the elements’ thickness and pitch, as well as the core’s length and configuration. The evolution of current designs mandates setting up new experiments to alleviate some uncertainties; however, these are costly and not readily available. Therefore, there is a real need to complement these expensive experiments and capture multi-physics effects using numerical modeling and simulation tools. This dissertation introduces a steady-state OpenFOAM multiphysics package tailored for NTP core simulation. Denoted as NTPSteadyFOAM, this package consists of three sub-modules that allow to model NTP fuel-to-coolant heat transfer, element-to-element heat transfer, and power generation through neutron diffusion. In this package, the single element fuel-to-coolant heat transfer is modeled using a 1.5-D sub-channel approach in conjunction with multiple empirical convective heat transfer models. The element-to-element heat transfer is governed by the Laplacian thermal conduction model with customized boundary interface treatments to capture heat flux across reactor sub-elements. The neutronic simulation relies on a traditional multi-group diffusion model using the P_1 approximation. Moreover, the multiphysics coupling approach across modules is built upon an existing reactor solver, GeN-Foam, designed to tighten the coupling between thermal-hydraulics (T/H), neutronic, and other important feedback. The NTPSteadyFOAM package is heavily verified against higher-fidelity models (e.g., Conjugate Heat Transfer CFD solver for T/H and Monte Carlo simulation for neutronic models) that compare single sub-element T/H analysis as well as different patterns of inter-elemental heat transfer. Furthermore, full core neutronic-T/H coupling capability is demonstrated using a miniature sample NTP core. As result, this dissertation presents the fidelity and the multiphysics capabilities of the coarse-mesh-based multiphysics package for NTP core simulations. The overarching goal of this work is to establish a multiphysics package designed to obtain a full core solution with a timely-efficient computational performance.Ph.D