Multiobjective aerodynamic shape optimization of NACA0012 airfoil based mesh morphing

Abstract

The actual use of computational fluid dynamics (CFD) by aerospace companies is the trade-off result between the perceived costs and benefits. Computational costs are restricted to swamp the design process even if the benefits are widely recognized. The need for fast turnaround, counting the setup time, is also crucial. CFD integrates mathematical relations and algorithms to analyze and solve fluid flow problems. CFD analysis of an airfoil produces results such as the lift and drag forces that determine the performance of an airfoil. Thus, optimizing these aerodynamic performances has proved extremely valuable in practice. The aim of this paper is to model a transonic, compressible and turbulent flow over a NACA 0012 airfoil, using a density based implicit solver, for which a comparison and a validation will be made throught the published experimental data. The numerical results show that the predicted aerodynamic coefficients are in a satisfying agreement with experimental data. Then an aerodynamic shape optimization algorithm, based on a multiobjective algorithm that is an extension of the Backtracking Search Algorithm which was initially developed for single-objective optimization problems only, was used in order to obtain an improved performance control of the aerodynamic coefficients of the optimized airfoil.