Model Development

Development of well calibrated and validated models of complex multi-physics systems presents a number of challenges. First one needs calibrated and validated models for the various physical phenomena. In the reentry vehicle problem being considered at PECOS, these include models for ablation, chemistry, hypersonic flow, radiation and turbulence.  Second, one needs models for the interaction of the various physical phenomena in the system of interest. These coupling models themselves will generally include uncertain parameters that must be calibrated, and they must of course be validated. Third, the computational cost of these coupled multi-physics models can be prohibitive, especially when they are being used in the context of uncertainty quantification in which the model will generally by solved many times (thousands) to characterize uncertainties in the predicted quantities. As a consequence, one generally wants to use the simplest possible models that can be used to predict the quantities of interest.

In the PECOS Center, for each modeling domain, we begin by considering the simplest models that are in common use for problems similar to those we are pursuing (reentry vehicles). These models are then subjected to a rigorous validation process, and more sophisticated models are employed only when simpler models fail validation.

New mathematical models of physical phenomena are being developed in modeling domains in which we anticipate that currently available models will be inadequate for predictions of our quantities of interest in reentry vehicle simulations. Thus, at PECOS, new physics models are being pursued in ablation, thermal non-equlibrium and turbulence. Furthermore, simulations of reentry vehicles have not previously fully coupled the models for the various physical phenomena. While the physics of these coupling interactions is generally known, we have had to develop an integrated coupling model formulation to support our fully coupled multi-physics model.