An Efficient Multi-objective Aerodynamic Shape Optimization Based on Improved NSGA-II

In this paper, we propose an efficient multi-objective aerodynamic shape optimization combining the POD geometric parameterization method and the improved NSGA-II algorithm. Compared to other traditional airfoil parameterization methods, POD has been found to give the most efficient coverage of the airfoil design space using relatively fewer parameterization variables. Therefore, the application of the POD method in aerodynamic design optimization can effectively reduce the number of design variables and have the potential to improve optimization efficiency. Besides, when NSGA-II is adopted in aerodynamic optimization, the maximum generation is often used as the iteration stopping criterion, which can easily lead to the waste of computational resources. Here, the authors suggest a convergence criterion for Pareto solution sets, which determines whether the algorithm stops based on the distance between results of adjacent generations. The effectiveness of this method has been verified through test problems experiments. Finally, a transonic RAE2822 airfoil case with two objectives and three constraints is presented. Simulation results show that our proposed method significantly improves the efficiency of aerodynamic shape optimization under condition of the approximately unchanged optimal extent.