High-fidelity surface flow data-driven aerodynamic solution strategy for non-smooth configurations: Study of compressor cascade with micro riblet surface

In this paper, a new aerodynamic solution strategy for non-smooth configurations is proposed based on the wall modification model by machine learning to perform numerical simulations, rather than directly describing global flow field with massive grids. The effect of in presence pressure gradients investigated utilizing method. Flow features surface are provided high-fidelity data acquired through lattice Boltzmann method simulation. constructed Fruit fly Optimization Algorithm-Generalized Regression Neural Network (FOA-GRNN) reproduce behavior microflow near surface. Typical features, e.g., velocity corrections induced texture as output FOA-GRNN model, imposed configuration boundaries, improving computational efficiency and resolution. novel validated comparing results experiment. addition, performance analysis compressor cascade micro riblet above conducted. indicate that structure decreases skin friction turbulent intensity channel compared smooth cascade, thus significantly reducing total loss. paper shows positive prospect data-driven evaluating performances provides reliable subsequent design micro-nano surfaces.