Abstract A modified STRUCT (MST) turbulence model for efficient engineering computations of turbulent flows in hydro-energy machinery is proposed in this paper. The MST model switches between URANS and LES-like modes using a new damping function to adjust the turbulent viscosity. Compared with the original STRUCT method, the modifications are as follows: (1) the BSL k-ω model with the Spalart-Shur correction is chosen as the new baseline to improve the sensitivity to rotation and curvature; (2) a new adaptive time-scale ratio is proposed to avoid the arbitrariness of geometric averaging operation in the original method; (3) the normalized helicity is introduced into the new damping function to detect the energy backscatter phenomenon. Five classical high Reynolds number flow cases are tested. The results show that the turbulent viscosity of the MST model is reasonably reduced in the massively separated regions and LES-like mode is activated, which captures more turbulent vortices and fluctuations on the URANS grids. With high efficiency and robustness, the MST model inherits the advantages of the original STRUCT method and improves the prediction accuracy of the turbulence with rotation and curvature, which enables efficient engineering computations of turbulent flows in hydro-energy machinery.

中文翻译：

---

一种改进的 STRUCT 模型，用于水能机械中湍流的有效工程计算

摘要 本文提出了一种改进的STRUCT (MST) 湍流模型，用于对水能机械中的湍流进行有效的工程计算。MST 模型使用新的阻尼函数在 URANS 和类 LES 模式之间切换，以调整湍流粘度。与原来的STRUCT方法相比，修改如下：（1）选择带有Spalart-Shur校正的BSL k-ω模型作为新的基线，以提高对旋转和曲率的敏感性；(2) 提出了一种新的自适应时标比例，避免了原有方法几何平均操作的随意性；(3) 在新的阻尼函数中引入归一化螺旋度来检测能量反向散射现象。测试了五个经典的高雷诺数流动案例。结果表明，MST 模型的湍流粘度在大量分离的区域中得到了合理的降低，并激活了类 LES 模式，从而在 URANS 网格上捕获了更多的湍流涡和波动。MST模型具有高效性和鲁棒性，继承了原有STRUCT方法的优点，提高了带有旋转和曲率的湍流的预测精度，使水能机械湍流的工程计算更加高效。