Reece OTTO, Kyle DAMM, Rowan GOLLAN
DOI Number: N/A
Conference number: HiSST-2025-259
The design of hypersonic airbreathing vehicles involves tightly coupled aerodynamics and propulsion, creating complex design spaces that are most effectively navigated using numerical optimisation techniques. However, achieving high-fidelity, many-parameter optimisation in three dimensions remains difficult while also maintaining efficiency and robustness. To address this challenge, this paper presents an adjoint-based aerodynamic shape optimisation (ASO) framework for three-dimensional hypersonic configurations. The framework integrates high-fidelity CFD analysis, adjoint-based sensitivity evaluation, and a two-level free-form deformation (FFD) parameterisation strategy. The method is demonstrated through the optimisation of a hypersonic lifting-body configuration subject to an internal payload constraint. The configuration was simulated using Reynolds-Averaged Navier–Stokes (RANS) equations and parameterised with 121 design variables. The optimisation required just over four days of runtime on a workstation and achieved a 21.8% improvement in lift-to-drag ratio. Geometric and flow field analyses revealed a reduced frontal area, a slenderised forebody, and waverider-like shock attachment
that improved aerodynamic efficiency and weakened side vortices. These results highlight the capability of adjoint-based methods to deliver efficient, high-fidelity optimisation of complex three-dimensional hypersonic vehicles, representing a step toward practical many-variable design studies in this regime.