Aerodynamic optimization based on variable-fidelity models

Based on the reliable region methodology, combined with the CFD simulations from both of the N-S and Euler equations, an aerodynamic optimization scheme with gradient-based approximation management framework (AMF) is proposed, where the variable-fidelity models collecting full advantages of the low-fidelity models are organized and managed. With this framework most of the workload is concentrated on the low-fidelity models in optimization iterative procedures. Furthermore, the AMF adopts high-fidelity model to monitor the progress to lead to a globally converge for a solution of high-fidelity model. An airfoil is designed to show that the method enables to achieve the same optimal results as that optimized by N-S equations with a shorter period, and the results are much better than those given by the optimization with Euler equations.