Automated Design Optimization for Hypersonic Plasma-Aerodynamics

Clear Science Corp. and the University of Minnesota propose to evaluate and develop design optimization methods that exploit flow ionization associated with hypersonic, trans-atmospheric flight vehicles. Using magnetic fields from on-board devices, power may be extracted from ionized gas generated in the high-energy flow and electromagnetic forces can control heating and aerodynamic loading on the vehicle surface by altering the structure of the shock and boundary layers. Two challenges exist in meeting the objective of design optimization. First, all of the relevant physical properties of the flow must be appropriately modeled. These include compressibility, viscous effects, thermo-chemical non-equilibrium, Hall and ion-slip effects, and an accurate representation of the coupling between the plasma and the applied magnetic field. The resulting system of equations is complex, and the second challenge consists of deriving a compatible optimization algorithm that is efficient, versatile, and scalable. During Phase I, we will evaluate adjoint (backward) formulations, sensitivity (forward) methods, and algorithms that combine optimization with reduced-order modeling. Evaluations will be based on tests involving two-dimensional blunt-body flow in the low magnetic Reynolds number regime. Figures of merit will include scalability to three-dimensional flows and to problems with two-way coupling between the plasma and the magnetic field. The down-selected optimization algorithm will be developed in Phase I, and a Phase II work plan will be formulated for demonstration testing.