Neural network metamodel-based MDO for wing design considering aeroelastic constraints

MDO is an important tool for designers and its use is spreading out as new implementations establishes. The focus of this methodology is to bring together disciplines involved with the design to work all their variables concomitantly, at an optimization environment to obtain better solutions. It is possible to use MDO in any stage of the design process, that is in the conceptual, preliminary or detailed design, as long as the numerical models are tted to needs of each of this stages. This work describes the development of a MDO code for the conceptual design of aeroelastic aircraft wings. As a tool for the designer at the conceptual stage, the numerical models must be fairly accurate and fast. The aim of this study is to analyse the use of a neural network based metamodel for the utter prediction of aircraft wings in the MDO code, instead of a conventional model itself, what may aect signicantly the computational cost of the optimization. Well trained neural networks are able to provide accurate results within short time, making them very useful for the type of the proposed methodology. The neural metamodel is prepared using aeroelastic code based on nite element model coupled with linear strip aerodynamics. The MDO process is achieved using genetic algorithm. Two case studies are presented to evaluate the performance of the MDO code, revealing that the metamodel approach does improve the overall optimization process.