Implementation and demonstration of RANS-based longitudinal stability constraint in a multidisciplinary design optimization environment

Abstract

Longitudinal static stability is the stability of an aircraft about its lateral axis. A longitudinally unstable aircraft tends to dive or climb progressively when subjected to minor changes in the forces that act on it. Currently, the gradient-based MDO chain at the DLR Institute of Aerodynamics and Flow Technology considers this stability constraint via lower fidelity methods. The target of this study is to add this constraint to an aero-structure optimization chain based on RANS computations. This means that the detection of the feasibility of the stability constraint will be predicted via RANS-based solvers, and the gradients of the constraints are evaluated, while employing the existing high fidelity coupled aeroelastic adjoint approach. Afterwards, the engagement of this constraint in MDO is demonstrated and the benefit of engaging it, is evaluated. The results showed, that the optimized design showed a improvement by 2 drag count , in comparison to the baseline design. However, the static margin valued increased by 2.37%, relative to baseline. The gain in performance, is due to shocks being better smeared along the wing span, in comparison to baseline design. However, It is also understood, how the static stability constraint cannot be considered solely by itself, rather it needs to be modified, to consider for controllability as well.