A physics-informed conditional diffusion method for airfoil flow field prediction
Abstract
Accurate and efficient prediction of airfoil flow fields remains computationally expensive with computational fluid dynamics(CFD) solvers. This paper presents a physics-informed conditional diffusion method that generates flow fields under different inflow conditions. Physical constraints derived from the continuity and momentum equations are incorporated into the objective function as regularization terms. Experimental results show that the proposed method consistently outperforms the baseline diffusion model, reducing ‖L2‖ error by up to 27% and increasing R2 score to above 0.98 across all variables, highlighting the effectiveness of integrating domain knowledge into generative modeling for engineering applications.