1 July 2025 in Physics of Fluids
A multi-point aerodynamic design method using a conditional diffusion model addresses aerodynamic challenges across speed regimes for flying wing configurations. The model is based on classifier-free guidance and adopts a three-dimensional (3D) airfoil profile design method for a flying wing layout to perform fast design for five wing profiles of a flying wing layout. The differences between the predicted profiles and aerodynamic performance of the method in the high and low speed integrated design states and the intended target are analyzed, and the aerodynamic performance of the flying wing layout before and after the prediction is compared. The results show that the errors of each of the profile parameters predicted by the method and the real profile parameters are within 3%. Furthermore, the aerodynamic performance of the airfoils derived from the inverse design closely matches the real aerodynamic data at each profile station, meeting the requirements of the inverse design methodology. In addition, the lift coefficient of each profile predicted by the given better aerodynamic performance is improved by about 7.78% on average at high speed, and the drag coefficient is reduced by about 9.23% on average at low speed, which is in line with the predetermined performance enhancement target. Replacing original airfoils with optimized designs improves the flying wing's lift-to-drag ratios in both flight conditions, confirming the method's potential for inverse aerodynamic design in multi-operational flying wing configurations.
1 July 2024 in Physics of Fluids
Aerodynamic/stealth design is becoming an important factor in the advanced airfoil design. In this work, a supervised machine learning method is proposed for aerodynamic and stealth integrated airfoil design. The conditional generative adversarial network (CGAN) is constructed for the multidisciplinary design of airfoil. Then, the generator and discriminator simply using deep neural network have good robustness and stability in training. The CGAN model also shows good generalization capability in the test set, with less than 1% error in fitting to the airfoil profile data, and the generated airfoils are within 10% error compared to the test airfoil aerodynamic stealth characteristics. In addition, the optimization results based on the CGAN model demonstrate that aerodynamic performance improvement would increase the airfoil camber and stealth performance improvement would sharpen the airfoil leading edge.