Nonlinear modelling for Multi-Disciplinary and Multi-Objective Optimization of a complete aircraft

The aerospace engineering typically deals with multidisciplinary complex systems, and narrow margins of the design parameters make necessary the introduction of multi-objective approaches in order to pick the best design. Genetic algorithms, in addition to gradient-based ones, allow to evaluate Pareto frontiers, in the objective space, i.e. the set of best trade-offs, thanks to the current satisfactory level of computer performance. In the present paper, an integrated Multidisciplinary Design Optimization (MDO) has been used to solve a Multi-Objective Optimization (MOO) problem for a notional regional aircraft type comprised of fuselage, tail and wing, where the optimization criteria include minimal structural weight, maximum L/D and maximum mission range, taking into account also of aeroelastic constraints. The level of fidelity of the disciplines used for the analyses is a relevant issue in the optimization process although it may contrast computational cost. In the present paper a non-linear analyses have been proposed within the MDO/MOO process to evaluate the aerodynamic performance (i.e. the L/D is computed as the ratio of lift and drag coefficients) for a more fidelity estimate of the induced drag by using a free-wake approach.