Approach. Tightly integrated simultaneous analysis and design (SAND) methods based on adjoint techniques can drastically reduce the computation cost relative to that for a conventional method using familiar, validated, high fidelity analysis and optimization codes, i.e. legacy codes. Simultaneous Aerodynamic and Structural Design Optimization (SASDO) is a SAND procedure that, using legacy codes, incorporates design improvement within expensive, iterative function and sensitivity analyses so as to achieve fully converged flow solutions only near optimal design, thereby gaining computational efficiency. The function and sensitivity analyses are iterative due to both the nonlinear CFD analysis and the loose coupling with the linear Finite Element Method (FEM) structural analysis. This SASDO procedure therefore exhibits simultaneity not only with respect to analysis and design, but also with respect to the aero-structural coupling. Our previous single discipline (aerodynamics) feasibility demonstrations used consistent, discrete, analytical sensitivity derivatives for quasi 1-D Euler; 2-D turbulent Navier-Stokes, and 3-D Euler flows. This initial SASDO procedure uses automatically differentiated versions of CFL3D and FEM for the flow and structure function and sensitivity analyses. Changes in eight design variables are sought to maximize the lift-to-drag ratio of a simple, trapezoidal planform wing, subject to solution-dependent constraints on the difference between lift and weight, the pitching moment coefficient and the compliance, a measure of the work required to deflect the wing. The eight design variables are tip chord, tip setback, semispan, root camber, and structural element sizes for four of the six regions of the wing as shown in the figure.

Accomplishment Description. These flexible wing optimization results demonstrate the feasibility of SASDO under dual simultaneity. As shown on the figure, SASDO finds essentially the same local minimum as a conventional technique. For the original design all constraints were satisfied, although the payload constraint was nearly active. For the optimized design, the constraints on pitching moment and compliance were active as was the side constraint on the tip setback. The computational cost was reduced by the SASDO method, but not by the amount anticipated from previous feasibility demonstrations due to the disproportionate CFD sensitivity analysis cost.

Figure: Simultaneous Aerodynamic and Structural Design Optimization Results

Related Presentations:

C. R. Gumbert, G. J.-W. Hou, and P. A. Newman: "Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) of a 3-D Flexible Wing", 39th AIAA Aerospace Sciences Meeting and Exhibit, Reno, NV, Jan 8-11, 2001.

Gumbert, C. R., Hou, G. J.-W., and Newman, P. A.: "Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) of a 3-D Rigid Wing", presented at the 14^th AIAA Computational Fluid Dynamics Conference, Norfolk, June 1999.

Gumbert, C. R., Hou, G. J. -W., and Newman, P. A.: "Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) of a 3-D Rigid Wing", Proceedings, 14^th AIAA Computational Fluid Dynamics Conference, Norfolk, June 1999;also AIAA Paper 99-3296.

Gumbert, C. R., Hou, G. J. -W., and Newman, P. A.: "Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) of a 3-D Flexible Wing", AIAA Paper 2001-1107, January 2001.

Gumbert, C. R., Hou, G. J. -W., and Newman, P. A.: "Simultaneous Aerodynamic and Structural Design Optimization (SASDO) of a 3-D Wing", Proceedings, 15^th AIAA Computational Fluid Dynamics Conference, Anaheim, June 2001;also AIAA Paper 2001-2527.