Abstract:We improve upon two key aspects of the Menter shear stress transport (SST) turbulence model: (1) We propose a more robust adverse pressure gradient sensor based on the strength of the pressure gradient in the direction of the local mean flow; (2) We propose two alternative eddy viscosity models to be used in the adverse pressure gradient regions identified by our sensor. Direct numerical simulations of the Boeing Gaussian bump are used to identify the terms in the baseline SST model that need correction, and a posteriori Reynolds-averaged Navier-Stokes calculations are used to calibrate coefficient values, leading to a model that is both physics driven and data informed. The two sensor-equipped models are applied to two thick airfoils representative of modern wind turbine applications, the FFA-W3-301 and the DU00-W-212, with maximum thicknesses of 30% and 20% of their chord lengths, respectively. While the baseline SST model predicts stall (onset of separation) $3^\circ$ to $5^\circ$ late for all cases considered, the proposed models predict stall within the margins of experimental uncertainty, which greatly improves the prediction of the maximum lift generated. For the FFA airfoil, the models also improve the prediction of the linear region of the lift curve likely due to their improved prediction of a pressure-side separation at low angles of attack. The models are shown to generalize well across the two airfoil geometries (despite their difference in thickness) and across almost a factor of 10 in variations in chord-based Reynolds numbers from $1.6\times10^6$ to $1.5\times10^7$.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: **Improve the accuracy of the Menter Shear - Stress Transport (SST) turbulence model in predicting separation points, especially the problem of being unable to reliably predict separation in smooth - body flows**. Specifically, the objectives of the paper can be summarized as follows: 1. **Improve separation prediction**: The current SST model performs poorly in predicting separation points, especially under thick airfoil and high - Reynolds - number conditions. The paper proposes a new adverse - pressure - gradient sensor and two improved eddy - viscosity models to enhance the ability to predict separation points. 2. **Maintain the accuracy of the zero - pressure - gradient boundary layer (ZPGBL)**: Many existing improvement methods, although improving separation prediction, will undermine the accurate prediction of the zero - pressure - gradient boundary layer. The method proposed in the paper aims to avoid this problem. 3. **Generality of model coefficients**: The improved model should be able to maintain consistency across different geometries and Reynolds - number ranges without the need to adjust coefficients for each specific situation. ### Specific problem description The paper points out that although the Menter SST model is still one of the most advanced open - source models in many research and industrial applications, it has significant deficiencies in predicting separation points on smooth bodies. For example: - In the Boeing Gaussian bump case, the SST model fails to correctly capture the separation phenomenon. - For the FFA - W3 - 301 and DU00 - W - 212 airfoils, the SST model predicts the stall angle 3° to 5° later than the experimental data. These problems prompt the author to develop a new adverse - pressure - gradient sensor and improved eddy - viscosity models to more accurately predict separation points while maintaining good predictive ability for other flow characteristics. ### Mathematical formulas and improvement methods 1. **New adverse - pressure - gradient sensor**: The author proposes a new sensor based on the pressure - gradient intensity in the local average flow direction: \[ I_p=\left(\frac{\hat{u}_i - U_{i,w}\frac{\partial p}{\partial x_i}F_2}{s_T\rho_{\infty}U_{\infty}^2/L}> 1\right) \] where: - $\hat{u}_i$ is the unit velocity vector; - $U_{i,w}$ is the velocity closest to the wall; - $\frac{\partial p}{\partial x_i}$ is the pressure gradient; - $F_2$ is the boundary - layer sensor; - $s_T$ is a model constant used to define a sufficiently strong adverse - pressure - gradient. 2. **Improved eddy - viscosity models**: Two improved eddy - viscosity models are proposed: - $a_1, APG$ model: \[ \mu_t = \begin{cases} \frac{\rho k}{\omega}, & \text{if }a_1\omega > S F_2, \\ a_1'\frac{\rho k}{S F_2}, & \text{otherwise}, \end{cases} \] where $a_1'$ is modified to $a_1, APG$ in the adverse - pressure - gradient region. - $\beta^*, APG$ model: \[ \frac{\partial(\rho k)}{\partial t}+\frac{\partial(\rho u_j k)}{\partial x_j}=P-\beta^{*'}\rho\omega k+\frac{\partial}{\partial x_j}\left[(\mu+\sigma_k\mu_t)\frac{\partial k}{\partial x_j}\right]

Improved pressure-gradient sensor for the prediction of separation onset in RANS models

Les and Cdns Investigation of T106c Low Pressure Turbine

Large Eddy Simulation and CDNS Investigation of T106C Low-Pressure Turbine

Computation of Hypersonic Transitional Flows Over Cones Using Gas-Kinetic Scheme Coupled with the Turbulence Model

Prediction of Separation Induced Transition on Thick Airfoil Using Nonlinear URANS Based Turbulence Model

Improved SST turbulence model for supersonic flows with APG/separation

Aerodynamic Prediction of the High-Lift Common Research Model with Modified Turbulence Model

Machine Learning-augmented Predictive Modeling of Turbulent Separated Flows over Airfoils

Transition Modeling for Low Pressure Turbines Using Computational Fluid Dynamics Driven Machine Learning

Analysis of Turbulent Separated Flows for the NREL Airfoil Using Anisotropic Two-Equation Models at Higher Angles of Attack

Development of a Generalizable Data-driven Turbulence Model: Conditioned Field Inversion and Symbolic Regression

Improved Self-Adaptive Turbulence Eddy Simulation for Complex Flows and Stall Prediction Using High-Order Schemes

Study of Delayed-Detached Eddy Simulation with Weakly Nonlinear Turbulence Model

Data-driven turbulence modeling in separated flows considering physical mechanism analysis

Progressive augmentation of Reynolds stress tensor models for secondary flow prediction by computational fluid dynamics driven surrogate optimisation

Assessment of advanced RANS turbulence models for prediction of complex flows in compressors

Aerodynamic Prediction of High-Lift Configuration Using K−v2¯−ω Turbulence Model

Robust experimental data assimilation for the Spalart-Allmaras turbulence model

Prediction of Airfoil Stall Based on a Modified K−v2¯−ω Turbulence Model

Aerodynamic Prediction of a CRM High-lift Configuration using a modified three equation turbulence mode

Study of the wall pressure variations on the stall inception of a thick cambered profile at high Reynolds number