TL;DR
This paper introduces a novel curvature correction factor ($f_c$) to the widely-used k-ω SST turbulence model. This modification significantly improves the accuracy of simulating turbulent flows around bluff bodies, particularly in capturing vortex shedding and reducing drag prediction errors, by better accounting for streamline curvature and fluid rotation.
What Problem Does This Paper Address?
Simulating fluid flow around bluff bodies, like buildings or bridge piers, is crucial for engineering design. While Reynolds-Averaged Navier-Stokes (RANS) models like k-ω SST are computationally efficient, they struggle to accurately capture the complex, unsteady turbulent fluctuations and vortex dynamics in the wake of these bodies. This limitation leads to significant errors, especially in predicting turbulent kinetic energy and drag coefficients. The paper aims to bridge this gap by enhancing the k-ω SST model to better handle the effects of streamline curvature and fluid rotation inherent in bluff body flows.
Methods
The core of this research is the introduction of a curvature correction factor ($f_c$) into the production term ($P_k$) of the turbulent kinetic energy equation within the k-ω SST turbulence model. This factor is derived from the local fluid rotational rate and is designed to improve the model's sensitivity to streamline curvature and unsteady vortex dynamics. The modified k-ω SST model, termed the "curvature-corrected turbulence model," was then applied to simulate two distinct flow scenarios: flow around a 2D airfoil and flow around a 2D square column. Numerical simulations were performed using the STAR-CCM+ software.
Datasets
The study relies on existing experimental data for validation:
- 2D Airfoil Flow: Experimental data from Gregory [30] for the NACA0012 airfoil at various angles of attack and a Reynolds number of $2.88 \times 10^6$.
- 2D Square Column Flow: Experimental data from Lee [18] for a square column at a Reynolds number of $1.76 \times 10^5$.
No new experimental datasets were generated for this study; rather, the focus was on validating the proposed model against established benchmarks.
Key Insights and Ideas
- Curvature Matters for Bluff Bodies: The study strongly suggests that while standard RANS models are adequate for simpler flows like airfoils with mild curvature, they fall short when dealing with the intense rotational motion and significant vortex shedding characteristic of bluff bodies.
- $f_c$ Captures Rotational Effects: The proposed curvature correction factor ($f_c$) effectively accounts for the local fluid rotational rate, leading to a more accurate representation of the turbulent kinetic energy production in complex flow regions.
- Improved Drag Prediction: The most significant impact of the correction is observed in bluff body flows, where it drastically reduces the error in predicting the time-averaged drag coefficient.
- Vortex Dynamics are Key: The improved accuracy in bluff body simulations is directly linked to the model's enhanced ability to capture the formation, shedding, and evolution of vortices in the wake.
Key Results
- Airfoil Flow: For 2D airfoils, the introduction of $f_c$ showed only a limited improvement in simulation accuracy. The error in the average lift coefficient ($C_L$) remained largely similar, with the standard k-ω SST model already providing good results due to the relatively mild flow variations.
- Square Column Flow (Zero Angle of Attack): The curvature-corrected model significantly improved accuracy. The error in the time-averaged drag coefficient ($C_D$) was reduced from 24% (standard k-ω SST) to 8.3%.
- Square Column Flow (Non-zero Angles of Attack): Similar trends were observed for non-zero angles of attack, indicating the robustness of the correction for bluff body scenarios. The corrected model consistently showed better agreement with experimental data for drag coefficients and pressure distributions.
- Qualitative Improvements: Visualizations of streamline and vorticity distributions clearly show the curvature-corrected model better capturing vortex structures and their dynamics compared to the standard model.
Limitations and Future Work
- Limitations: The paper acknowledges that the improvement for airfoil flows is limited, suggesting that the correction is most impactful in scenarios with strong vorticity and curvature. The study is limited to 2D cases.
- Future Work: The authors plan to extend this research to three-dimensional bluff bodies. This will involve both numerical simulations and experimental investigations to assess the applicability and effectiveness of the curvature-corrected model in more complex, real-world scenarios.
Summary updated