With the rapid advancement of high-performance computing and artificial intelligence technologies, data-driven AI models have been extensively researched in the field of civil aircraft aerodynamic design, demonstrating significant potential in design space compression, key feature extraction, flow field prediction, and intelligent optimization design. However, the application of purely data-driven models in engineering design still faces many challenges, including the scarcity and high acquisition cost of domain-specific data, as well as deficiencies in model reliability, generality, interpretability, and usability. Integrating physical knowledge and aerodynamic design experience into model development has become a key approach to addressing these challenges, providing an important direction for advancing technology in this field. This paper, from the perspective of civil aircraft engineering design and supported by relevant practices in intelligent aerodynamic design, reviews recent theories and progress in data- and knowledge-driven AI models in the areas of knowledge embedding, knowledge correction, and knowledge discovery. It further explores the current state of research and application potential of data- and knowledge-driven methods in civil aircraft aerodynamic design, while offering insights into the future of new paradigms in intelligent aerodynamic design.

Inverse design has long been an efficient and powerful design tool in the aircraft industry. In this paper, a novel inverse design method for supercritical airfoils is proposed based on generative models in deep learning. A Conditional Variational AutoEncoder (CVAE) and an integrated generative network CVAE-GAN that combines the CVAE with the Wasserstein Generative Adversarial Networks (WGAN), are conducted as generative models. They are used to generate target wall Mach distributions for the inverse design that matches specified features, such as locations of suction peak, shock and aft loading. Qualitative and quantitative results show that both adopted generative models can generate diverse and realistic wall Mach number distributions satisfying the given features. The CVAE-GAN model outperforms the CVAE model and achieves better reconstruction accuracies for all the samples in the dataset. Furthermore, a deep neural network for nonlinear mapping is adopted to obtain the airfoil shape corresponding to the target wall Mach number distribution. The performances of the designed deep neural network are fully demonstrated and a smoothness measurement is proposed to quantify small oscillations in the airfoil surface, proving the authenticity and accuracy of the generated airfoil shapes.