Abstract:Turbulence modeling within the RANS equations' framework is essential in engineering due to its high efficiency. Field inversion and machine learning (FIML) techniques have improved RANS models' predictive capabilities for separated flows. However, FIML-generated models often lack interpretability, limiting physical understanding and manual improvements based on prior knowledge. Additionally, these models may struggle with generalization in flow fields distinct from the training set. This study addresses these issues by employing symbolic regression (SR) to derive an analytical relationship between the correction factor of the baseline turbulence model and local flow variables, enhancing the baseline model's ability to predict separated flow across diverse test cases. The shear-stress-transport (SST) model undergoes field inversion on a curved backward-facing step (CBFS) case to obtain the corrective factor field beta, and SR is used to derive a symbolic map between local flow features and beata. The SR-derived analytical function is integrated into the original SST model, resulting in the SST-SR model. The SST-SR model's generalization capabilities are demonstrated by its successful predictions of separated flow on various test cases, including 2D-bump cases with varying heights, periodic hill case where separation is dominated by geometric features, and the three-dimensional Ahmed-body case. In these tests, the model accurately predicts flow fields, showing its effectiveness in cases completely different from the training set. The Ahmed-body case, in particular, highlights the model's ability to predict the three-dimensional massively separated flows. When applied to a turbulent boundary layer with Re_L=1.0E7, the SST-SR model predicts wall friction coefficient and log layer comparably to the original SST model, maintaining the attached boundary layer prediction performance.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the insufficient prediction ability of Reynolds - averaged Navier - Stokes (RANS) turbulence models under complex flow conditions, especially in separated flows. Specifically, traditional RANS turbulence models such as the Shear - Stress Transport (SST) model have relatively large errors when predicting separated flows, especially performing poorly in flow predictions under complex geometric conditions such as periodic hills, curved backward - facing steps (CBFS), NASA bumps, iced airfoils, and three - dimensional Ahmed bodies. These problems are mainly reflected in incorrect prediction of separation points and inaccurate prediction of three - dimensional vortex structures.
In order to improve the prediction performance of these models, the author proposes a data - driven method based on symbolic regression (SR) to derive the analytical relationship between the baseline turbulence model correction factor and local flow variables. This method not only enhances the model's ability to predict separated flows but also improves the model's universality and interpretability, solving the problems of lack of interpretability and poor generalization ability of models generated by traditional data - driven methods (such as field inversion and machine learning, FIML).
By performing field inversion in the curved backward - facing step (CBFS) case to obtain the correction factor field \(\beta\), and then using symbolic regression techniques to derive an analytical mapping relationship between local flow characteristics and \(\beta\), the resulting analytical function is finally integrated into the original SST model to form the SST - SR model. Research results show that the SST - SR model performs well in various test cases, can accurately predict separated flow fields, and in particular, shows its good generalization ability in test cases completely different from the training set.