Architecture Modeling of In-Situ Oxygen Production and its Impacts on Lunar Campaigns

Abstract

In-situ lunar oxygen production has the potential to reduce the cargo mass launched from Earth necessary to sustain a lunar base. As research and development in lunar oxygen production continue, modeling tools are being used to help characterize the many possible system architectures and guide decisions for future plant designs. Using the previously built NASA In-Situ Resource Utilization (ISRU) System Model, an optimization tool was developed to facilitate exploration of the design space of the different system architectures represented in the model. For each architecture, an optimization of the continuous design space is performed using a gradient-based search. In instances when the gradient-based search cannot converge, the tool changes to simulated annealing, a heuristic method. Nine primary lunar oxygen production system architectures were optimized to minimize system mass for oxygen production levels from 500 kg/yr to 6000 kg/yr. Good designs minimized mass and maximized produced oxygen with system masses in the range of 100 kg to 700 kg. Preliminary results show that two particular architectures populate the Pareto-optimal front of best designs for most production levels, making them attractive for further investigation. An economy of scale of .837 was found using a power law regression, indicating that some economy of scale exists (values less than one have economy of scale) and that launching fewer, higher-capacity plants will be less massive overall than many small-capacity plants to achieve the same total production level. A simplified comparison of lunar-produced oxygen for crew breathing supply and ECLSS (environmental control and life support systems) technologies was performed with a space logistics planning tool, SpaceNet. For all but the most advanced ECLSS technologies, use of in-situ oxygen over a 10-year campaign resulted in more than 12,000 kg of consumables cargo launch mass savings.