On the use of geometry design variables in the MDO analysis of wing structures with aeroelastic constraints on stability and response

A Multidisciplinary-Design-Optimization (MOO) approach for the initial design of wing structures based on the integrated modeling of structures, aerodynamics, flight dynamics, and aeroelasticity, has been developed. The procedure is based on the use of a standard numerical optimizer which employs structural, aeroelastic, and aerodynamic analyzers (a finite-element analyzer for structures, panel method analyzer for aerodynamics, and a longitudinal trim analyzer for flight mechanics) in the MOO process. The space discretizations of the numerical models - namely, the structural and aerodynamic meshes - are dynamically updated as function of wing-structure geometry variables during the optimization process. The geometric-updating procedure has been validated using available analytical solutions for a structural optimization problem. The obtained MDO solutions are essentially based on a first-principle formulation but, in the meanwhile, they do not result to be computationally expansive. Moreover, in order to make the implemented MOO algorithm more computationally efficient and effective, an analytical method based on Matched-Filter Theory (MFT) has been used to evaluate the worst aeroelastic-response case due to a gust input having an assigned energy level with the purpose of obtaining a final design also optimizing the gust-response performance. The optimization capabilities of using geometry design variables vs the standard structural design variables in the MOO process have been highlighted. Moreover, the use of a composite objective function, combining structure and aerodynamic issues, has been shown further capabilities of the presented MOO procedure. Finally, some results are also presented for a benchmark regional aircraft wing to show the potentiality of the approach. (C) 2010 Elsevier Masson SAS. All rights reserved