Multi-Objective system optimization of a Mars atmospheric ISRU plant for oxygen production

The Mars Oxygen ISRU Experiment (MOXIE) is an instrument traveling to Mars onboard NASA's Perseverance rover. It will demonstrate, for the first time ever, in-situ resource utilization (ISRU) on the surface of another celestial body. MOXIE will utilize the carbon dioxide atmosphere of Mars to create oxygen as a demonstration of a planned larger mission. The instrument itself will produce oxygen at roughly 0.5 percent of the scale that would be necessary to support a human mission to Mars. A scaled-up version of MOXIE would be sent to Mars twenty-six months ahead of the first human mission and would aim to produce approximately 3 kg/hr of oxygen while in operation. This production rate would fully fuel the oxidizer portion of a Mars Ascent Vehicle prior to the first crew landing on Mars, which would enable that crew to return to Earth. This is a key capability to reduce mission risk by providing a safe return option on Mars prior to the crew arriving. Additionally, the system could provide oxygen for life support systems and habitation pressure. The intent of this paper is to describe a model that has been created to optimize the design of this scaled ISRU plant. It takes lessons learned from the MOXIE project and combines them with parameters and constraints of a planned human mission to systematically identify optimal design solutions. The extensibility of MOXIE is formulated through a multiobjective optimization problem for early-stage conceptual design. The objective functions minimize the power and mass required to build this ISRU system by changing operating conditions and system architecture while satisfying a set of constraints. The subsystems modeled for this problem include carbon dioxide acquisition and compression (CAC) to compress the Mars atmosphere, solid oxide electrolysis (SOE) to produce oxygen from carbon dioxide, and liquefaction to prepare the oxygen for storage. Additionally, the power, electronics, and heat exchange systems are simulated to capture gas transfer and control mechanisms. The model is built in MATLAB and uses Simulink as a framework. Results from this multiobjective optimization study and an analysis on the scalability of the MOXIE instrument show that an ISRU system that produces 22,717 kg of oxygen over 14 months would have a mass of 7,512 kg and a power requirement of 19,526 W. These results provide NASA and other agencies with an optimized design of a scaled ISRU system and its potential to reduce cost and risk as they prepare for a human mission to Mars.