Validation of Computational Fluid Dynamics Model of Self-pressurization in a Flight-scale Liquid Hydrogen Tank

A computational fluid dynamics (CFD) methodology, developed and demonstrated to satisfactorily model multiple cryogenic fluid management (CFM) operations, is applied to predict self-pressurization of a flight-scale liquid hydrogen tank in four ground tests. Axisymmetric two-phase CFD analyses are performed using the Loci-Stream-VOF CFD solver with the frozen gas-liquid interface method. Conjugate heat transfer modeling accounts for thermal transport between the tank walls and propellant fluids. Turbulent natural convection flow is modeled with three turbulence closures of differing fidelity --- Menter's shear stress transport (SST), improved delayed detached eddy simulation (IDDES), and wall-adapted local-eddy-viscosity (WALE) --- which show similar predictive capabilities. The asymmetric interface turbulence transport model ensures accurate phase change and thermal gradient predictions at the two-phase interface. Phase change is modeled with gas-kinetic-theory-based Hertz-Knudsen-Schrage formulation. The 50% fill-level test is validated for the CFM Modeling Portfolio with relative pressure-change error of -2.2% and satisfactory temperature predictions. Predictions are independent of spatial and temporal resolution, and of the accommodation factor parameter of the phase change model. The predictions for 25% and 90% fill-level tests were unsatisfactory, prompting a CFD sensitivity analysis of pressurization rate on spatial distributions of heat leak rates reveals that the tank wall redistributes most heat leakage to the liquid. However, the heat leak rate through the tank wall to the vapor phase governs the pressurization rate. The sensitivity analysis also produced spatial distributions of heat leak that accurately predict tank pressure and temperature for 25% and 90% fill-level tests; however, the findings warrant further investigations on spatial distribution. This study provides insights into self-pressurization mechanisms and along with the sensitivity analysis technique provide a way to improve self-pressurization predictions for NASA's spaceflight missions.