(615c) Dynamic Simulation of Unsteady Heat and Mass Transfer of Coolant and Cryogenic Product Fluid in Portable Tanks with a Vapor-Cooled Shield during an Operation Cycle

Various process gases such as hydrogen, helium, argon, or oxygen are supplied in large quantities in liquified form by portable cryogenic tanks. These tanks consist of an inner tank with the product to be transported, an outer container, and insulation in-between to keep the system refrigerated. Sometimes, additional liquid coolant is used as a shield to reduce the heat leak into the inner product tank by vaporization. The liquified product exists either in a single phase, e.g., in a subcooled state, or as a two-phase mixture of liquid and vapor. Due to undesired heat input into the system, a part of the liquid phase evaporates, causing an increase in pressure. If the maximum allowed pressure is exceeded, the vapor phase is vented for safety reasons, which results in product loss. The maximum travel distance and time that the product can be transported without loss characterize the performance of the tank. Ideally, the performance should remain the same with as little maintenance as possible.

Predictive maintenance and performance analysis can be achieved with hybrid modeling of the portable tank using field data and physics-based simulation of the underlying processes. The corresponding physical model must be as close as possible to the actual operation of the tank so that it can be used as a benchmark for the data-driven model. However, most published models consider the gaseous phase an ideal gas and the liquid phase as incompressible. Further assumptions are often phase equilibrium everywhere in the tank and a steady-state condition.

This contribution describes an extensive model of a portable vapor-cooled cryogenic tank. A time-resolved thermodynamic model for unsteady heat and mass flow under cryogenic conditions considers non-equilibrium evaporation and condensation of both coolant and product fluids, heat leaks, and self-pressurization. Both two-phase and single-phase supercritical system states are considered. All fluids are treated as real fluids.

The model was implemented in Python by solving a coupled differential-algebraic equation system for the coolant tank, the vapor-cooled shield, the product fluid tank, and the multi-layer insulation. REFPROP reference database was used for the calculation of required fluid properties. Four operation states were implemented: coolant refill, standard operation (dormancy), venting, and absence of coolant. The results provide a physics-based overview of the tank operation cycle and coolant consumption and can be used as a benchmark for data-driven predictive models. This benchmark is required to ensure that the data-driven models, which are not linked to the actual thermodynamics of the system, produce a physically sensible and realistic solution.

The developed model describes the time-resolved behavior of a vapor-cooled cryogenic tank. The assumptions of unsteady flows and real fluids allow for a more precise representation of coupled thermodynamic processes in tanks. A two-dimensional temperature and density distribution will be implemented in the future for a detailed analysis of the two-phase state. Also, the exact internal geometry requires a closer examination to take into account specific pressure losses and heat conduction.