The Impact of Flow-Fields on Uncertain Reduced Order Models

Blunt-body entry vehicles dissipate atmospheric heating and leverage aerodynamic drag to safely decelerate payloads during Entry, Descent, and Landing (EDL). Their blunt shape creates turbulent, recirculating wakes, leading to uncertain flight paths and challenging uncertainty mitigation strategies. Traditionally, conservative scalars and multipliers are applied to aerodynamic coefficient predictions which results in over-engineered designs with reduced payload capacity and less accurate landings. While advances in Computational Fluid Dynamics (CFD) enable high-fidelity analysis, the cost of extensive simulations remains prohibitive. Thus, surrogate models—such as parametric Reduced Order Models (ROMs)—are being trained with pressure distribution from the vehicle surface. This paper investigates the impact of including additional information from the flow-field to the ROM. Individual parametric ROMs are created from boundary and flow-field data using Proper Orthogonal Decomposition (POD) and sparse Kriging. In this preliminary analysis, it was concluded that boundary data is a sufficient source of aerodynamic information for ROM uncertainty predictions.