Multiobjective shape optimization of a hypersonic lifting body using a correlation-based transition model

Abstract

Although substantial numbers of aerodynamic shape optimization works have been carried out in the past few decades, the effects of boundary layer transition are not considered in the overwhelmingly majority of those previous studies. For more accurate prediction of the flow field, and for exploration of relations between aerodynamic heating and geometrical features, the commonly used local correlation-based transition model [Formula: see text] is implemented into a computational fluid dynamics (CFD) solver. A CFD-based multiobjective shape optimization method is then described and applied to a hypersonic lifting body in this paper. For parametric modeling and grid deformation, the flexible free form deformation and robust transfinite interpolation techniques are used respectively. The response surface model is adopted as the approximation model for the reduction of computational cost, and the multiobjective genetic algorithm is employed for the exploration of optimal solutions. Optimization processes have been conducted by the optimization toolkit DAKOTA and two Pareto fronts are obtained. Besides, typical solutions on the Pareto fronts are compared and analyzed. The results obtained from the case considered in this study indicate that the lift-to-drag ratio and transition onset position are closely related with the volume of the vehicle, and trade-off can be made for aerodynamic and aerothermodynamic performance from the optimal solutions. Moreover, the design problem of lifting body demonstrates the robustness and practicality of the new optimization procedures, which incorporates the transition phenomenon for hypersonic flow.