Data-driven investigation and modeling of flow in swept compressor cascades

Swept blades are widely adopted in fan and compressor designs to balance high load, high throughflow, and wide stall margins. However, their complex effects on the incoming flow field often result in inefficient trial-and-error design processes. To address this challenge, this study introduces a data-driven approach to analyze cascade flow fields and reveal the spanwise migration effects induced by swept blade geometry. A novel concept, the “actual sweep angle,” is proposed to quantify these effects, particularly their impact on incidence angle variations. Leveraging this concept, a predictive model is developed, achieving a high coefficient of determination (R2 = 0.9821). The swept blade design methodology based on this model significantly reduces the design cycle and improves aerodynamic performance in high-load fan and compressor applications, providing an efficient and reliable data-driven solution for modern swept blade optimization.