UC Davis

The Environmental Control and Life Support System (ECLSS) of a spacecraft is an integral part of human exploration. Current efforts to advance state of the art ECLSS subsystems in the International Space Station (ISS) are geared towards modeling and simulating various conditions and mission types. NASA and private partners are looking to enable life on the Moon and Mars, and a key factor to enabling that is carbon dioxide (CO2) removal. Astronauts depend on the removal of metabolic CO2 to keep cabin atmosphere at breathable levels which typically utilize adsorbent-based systems. Thus, modeling and simulating carbon dioxide removal aligns with the effort to advance future ECLSS technology while saving cost and time as compared to the traditional design-build-test approach. In addition, modeling and simulation can generate copious amounts of data and benefits research and development into ECLSS diagnostics and prognostics that require masses of data. This thesis aims to provide models of a carbon dioxide removal system that mimic the physical system, test what-if scenarios, simulate faulty and degraded conditions, implement state estimation and describes the development and results of an adsorbent degradation-focused testbed with relevance to deep space habitat settings. Chapter 1 is an introduction to the thesis with focus on the NASA HOME Institute which has funded this work, an overview of ECLSS, a description of CO2 Removal technology, modeling and simulation objectives, and model options and selection for CO2 removal. Chapter 2 provides an extensive literature review of the current status of ECLSS roadmaps, lessons learned, maintenance and spares logistics as well as ECLSS data analysis processes relevant to diagnostics and prognostics applications for deep space habitats. Chapter 3 details the development of a one-bed carbon dioxide removal system using Aspen Adsorption, a ready-made platform with built-in mathematical computations and capabilities for fault injection, to generate a multitude of data signatures, nominal and off-nominal, and validate against experimental data. Next, Chapter 4 describes model development using MATLAB, a mathematical program with full customization and algorithm integration capabilities but challenging development of numerical computations and fault injections, to generate nominal data signatures with the off shoot of applying state estimation to increase the model’s overall ability to combine measurement data with theoretical models to estimate sensor data, whether available or not. Finally, Chapter 5 details the assembly and test of a supplementary carbon dioxide removal testbed focused on sorbent degradation which achieved proof of concept operation and ultimately generated test protocols and documentation for hardware and software improvements for the next generation testbed.