Automated Design Optimization for the P2 and P8 Hypersonic Inlets

Abstract

An automated design methodology incorporating industry-standard Navier -Stokes codes and a gradient-based optimizer has been developed. This system is used to redesign the well-known NASA P2 and P8 hypersonic inlets. First, the Navier -Stokes simulations of the original P 2 and P8 inlet designs are validated using numerical convergence studies and comparison with wind-tunnel experimental data for the original inlets published by NASA in the early 1970s. Second, the P 2 and P8 inlets are redesigned with the objective of canceling the cowl shock (and, in the case of the P8 inlet, the additional cowlgenerated compression) at the centerbody by appropriate contouring of the centerbody boundary. The original inlets were intended to achieve these same objectives, but detailed experimental measurements indicated that a substantial re ected shock system was present. The choice of the objective function, which is used to drive the optimization, has a signi cant impact on the nal design. Several different formulations for the objective function have been employed, and improvements of 60 -90% in the objective function have been achieved. This automated design system represents one of the rst successful combinations of numerical optimization methods with Reynolds-averaged Navier -Stokes uid dynamics simulation for high-speed inlets, and demonstrates a new area in which high-performance computing may have considerable impact on problems of military and industrial signi cance. Nomenclature

Adjoint-based aerodynamic design optimisation in hypersonic flow

“…Along the bodyside of the inlet, our simulations are in good agreement with the experimental data in the boundary layer region, however, an over-prediction in the core-flow region is observed. The numerical results compare well to others who have applied the sharp cowl approximation, however, prior works did not notice an over-prediction of the normalised Pitot pressures in the core flow [235][236][237]241]. These authors used a patched-grid approach by dividing the inlet into self-contained regions, in comparison to simulating the inlet in a tip-to-tail fashion, as done in this current work.…”

“…Since the mid-1990s, several researchers have studied integrating high-fidelity Reynolds-Averaged Navier-Stokes (RANS) calculations into the hypersonic inlet design process. Gelsey et al [235] and Shukla et al [236,237] presented some of the first RANS-based design optimisation results for hypersonic inlets: throughout a series of papers, the authors presented redesigns of the P2 and P8 hypersonic inlets using NASA's NPARC and GASP codes coupled with the k-ε turbulence model. Hasegawa et al [238,239] also utilised the GASP code in an automated design optimisation of generic 2D inlets using the k-ω turbulence model.…”

“…We have selected to redesign the P2 hypersonic inlet to demonstrate the applicability of the discrete adjoint method to hypersonic inlet design. This problem has been previously used to demonstrate the use of black-box finite difference style CFD optimisation for hypersonic inlet design [235][236][237], and, consequently, the inlet design community has some familiarity with the geometry. The P2 inlet (illustrated in Figure. 8.6) was a two-dimensional, planar, hypersonic inlet designed for a generic 8.3.…”

Intelligent gradient-based search of incompletely defined design spaces

“…Our confidence in rule-based gradients was high enough that we decided to run the optimizations with rule-based gradients. We obtained good optimization results in these domains [results which in the case of hypersonic inlets have been published in the aerodynamics engineering literature (Shukla et al, 1997)], and we believe that the results would have been less successful if we had not used rule-based gradients, although we have not been able to experimentally test this belief. uous, smooth, and defined everywhere.…”

“…We have found that using rule-based gradients results in good optimization performance in several other design domains. These domains include the design of racing yachts (Ellman et al, 1993;Schwabacher et al, 1994Schwabacher et al, , 1996, the design of exhaust nozzles for supersonic jets (Gelsey et al, 1996a), the design of inlets for hypersonic jets (Gelsey et al, 1995;Shukla et al, 1996Shukla et al, , 1997, and the design of inlets for supersonic missiles (Zha et al, 1996(Zha et al, , 1997. All of these domains use simulators that are more expensive than our airframe simulator.…”