Enhanced Data Exploration through Reduced-Order Models

Evaluating spacecraft thermal control subsystem (TCS) performance is commonly achieved through high-fidelity computer models. They are especially useful during design stages, although they too have inherent costs. Development of a nominal spacecraft thermal model can take days to months to develop with run times on the order of hours or even days. Comparing and evaluating multiple TCS approaches, especially important in early design stages, amplifies these timelines. Considering the myriad TCS design approaches available, computational expense can become unwieldy. Consequently, there is a need for the development of reduced order models (ROMs) that can capture the effect of a high-resolution computer experiments without incurring significant computational expense. ROMs can then be used to evaluate different TCS approaches and provides a relatively quick means of evaluating design trade-offs.

An approach for developing ROMs was developed that provide a useful surrogate to more computationally expensive computer simulations. The approach relies on Latin Hypercube sampling in combination with Gaussian-Process data fitting techniques. For a NASA Crew Exploration Vehicle thermal model, a ROM was developed. Testing of the ROM showed that it did a good job replicating temperature, hydraulic power, and pressure responses for three different working fluids. These results will be presented along with a discussion of the advantages/disadvantages of the approach. Finally, an example of how ROMs can be used in practice will be provided.