A new approach to the constrained design of aerodynamic shapes is suggested. The approach employs Genetic Algorithms (GAs) as an optimization tool in combination with a Reduced-Order Models (ROM) method based on linked local data bases obtained by full Navier-Stokes computations. The important features of the approach include: (1) a new strategy for efficient handling of non-linear constraints in the framework of GAs (2) scanning of the optimization search space by a combination of full Navier-Stokes computations with the ROM method (3) multilevel parallelization of the whole computational framework.
The method was applied to the problem of one-point transonic profile optimization with non-linear constraints. The results demonstrated that the approach combines high accuracy of optimization (based on full Navier-Stokes computations) and efficient handling of various non-linear constraints with high computational efficiency and robustness. A significant computational time-saving (in comparison with optimization tools fully based on Navier-Stokes computations) allowed the method to be used in a demanding engineering environment.