AI 中文总结
针对冯·卡门湍流旋转流低频状态切换问题，利用粒子图像测速实验数据和扭矩测量数据，开发数据驱动叶轮模型，通过B样条参数化、动量强迫耦合大涡模拟，并经多方法优化系数，实现长时间模拟，再现平均流及亚稳态切换动力学。
AI 中文摘要
冯·卡门湍流旋转流呈现出有趣的大规模亚稳态动力学，包括低频状态切换。研究状态切换需要在高雷诺数下进行长时间高保真模拟以捕捉叶轮产生的流动。叶片解析大涡模拟计算成本过高。本文利用粒子图像测速实验数据和冯·卡门流扭矩测量数据，开发了叶轮对流动作用的模型。通过B样条对叶轮区域速度进行参数化并通过动量强迫与大涡模拟耦合。利用优化离散损失框架结合雷诺平均纳维 - 斯托克斯方程、设备光学可及部分的粒子图像测速测量和叶轮扭矩测量来推断初始B样条系数，再用协方差矩阵自适应进化策略进行优化。利用数据驱动的叶轮模型对冯·卡门流进行长时间大涡模拟，结果显示能再现主体平均流并呈现亚稳态切换动力学，且亚稳态非轴对称，由围绕圆柱形容器轴缓慢旋转的交替四单元流型组成。该方法为研究叶轮驱动湍流中的大规模动力学提供了实用且计算高效的途径。
英文摘要
The von Kármán turbulent swirling flow exhibits intriguing large-scale metastable dynamics, including low-frequency state switching. The study of state switching demands long-duration high-fidelity simulations at high Reynolds numbers that capture the flow generated by the impellers. Blade-resolved Large Eddy Simulations (LES) are computationally prohibitive, limiting access to these slow dynamics. Here, we develop a model for the action of the impellers on the flow using experimental data from Particle Image Velocimetry (PIV) and torque measurements of the von Kármán flow. The impeller-region velocity is parametrized via B-splines and coupled to the LES through momentum forcing. An initial set of B-spline coefficients is inferred using the Optimizing a DIscrete Loss (ODIL) framework constrained by the Reynolds-Averaged Navier--Stokes (RANS) equations, PIV measurements in the optically accessible portion of the device, and impeller torque measurements. The coefficients are then refined by the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), which minimizes the discrepancy between the LES time-averaged velocity and torque and their experimental counterparts. Using the data-driven impeller model, we perform long-duration LES of the von Kármán flow. We find that the simulation reproduces the mean flow in the bulk and displays metastable state-switching dynamics. We further show that these metastable states are not axisymmetric and consist of an alternating four-cell flow pattern that slowly rotates around the axis of the cylindrical vessel. The proposed approach provides a practical and computationally efficient route to investigating large-scale dynamics in impeller-driven turbulent flows.
