Efficient Aero-Structural Wing Optimization Using Compact Aerofoil Decomposition

Aerodynamic shape optimization of an aero-structural transonic wing is presented using a compact set of design variables. Modal decomposition (via singular value decomposition) of a training library of aerofoils to extract a compact set of aerofoil design variables has previously been shown to be highly effective. These have been used for high-fidelity global optimization of aerofoils. Hence, this work applies these design variables to optimization of an aero-structural wing. A representative transonic aircraft wing with defined structural modes is optimized as a rigid and aeroelastic shape. For fixed planform optimization, using as few as nine variables at ten spanwise sections is enough permit shock-free solutions when using a gradient-based optimizer for both the rigid and aeroelastic shapes. A drag reduction of 21.7% of the rigid wing is obtained. However, the aero structural solution of the optimised rigid wing shows a shocked solution, demonstrating the need to consider the optimization of the aeroelastic shape. As such, optimization of the aeroelastic wing is considered, where a drag reduction of and 29.7% is obtained against the aeroelastic baseline wing.