Hierarchical design and flow mechanisms of a novel bionic high-lift low-pressure turbine blade

High-lift low-pressure turbine (LPT) blades experience considerable aerodynamic losses caused by suction-side laminar separation and endwall secondary flows. This study proposes a spanwise non-uniform bionic blade design method for high-lift low-pressure turbine annular cascades. The proposed design enables region-specific flow control by considering the different dominant loss mechanisms along the blade span. Based on the proposed parameterization approach, a hierarchical optimization framework integrating subspace refinement and Bayesian optimization is developed. This framework improves the efficiency of exploring the high-dimensional design space. Large eddy simulations (LES) and low-speed annular cascade experiments are conducted to evaluate the aerodynamic performance and reveal the underlying flow mechanisms. The results demonstrate that the proposed method improves the integrated aerodynamic performance by optimizing the flow characteristics in different spanwise regions. In the mid-span region, the bionic blade promotes earlier transition, suppresses laminar separation, and reduces the profile loss by 5.6%. In the hub region, the bionic blade modifies the endwall secondary-flow structure by enhancing the entrainment of low-momentum fluid and reducing the accumulation of low-momentum fluid in the corner region. Consequently, the spanwise penetration of the secondary-flow loss region and the associated aerodynamic losses are both reduced. The proposed bionic blade configuration enables coordinated flow control across different spanwise regions and provides a new approach for the design of bionic blades in high-lift low-pressure turbines.