Numerical Investigation of Surface Roughness Effects on Drag Crisis Prediction Using the SST k-ω Turbulence Model

¹ Laboratory of Physics Materials and Subatomic, Ibn Tofail University, Kenitar, Morocco.
² Llaboratory LSIA, National School of Applied Sciences, BP. 03, Ajdir Al-Hoceima, Morocco.
³ "SATSIER" Team, Faculty of Sciences and Techniques of Al Hoceima, Abdelmalek Essaâdi University, Morocco.
⁴ Euromed Research Center, Euromed University of Fes, Meknes Road (Rond-point Bensouda), 30 000, Fes - Morocco.

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

The accurate prediction of the drag crisis around a circular cylinder remains one of the major challenges in Computational Fluid Dynamics (CFD), particularly when conventional Reynolds-Averaged Navier-Stokes (RANS) turbulence models are employed. Their limited capability to reproduce the natural laminar-to-turbulent transition of the boundary layer often leads to significant discrepancies in drag prediction within the critical Reynolds-number regime. In the present study, surface roughness is introduced as a physical transition mechanism to enhance the predictive performance of the SST k–ω turbulence model. Two-dimensional CFD simulations are performed using ANSYS Fluent for a circular cylinder with a diameter of 0.1 m over a wide Reynolds-number range (6.31 × 10⁴ ≤ Re ≤ 5.06 × 10⁵), covering the subcritical, critical, and supercritical flow regimes. A systematic parametric analysis is conducted to evaluate the influence of different roughness heights on the drag coefficient, pressure distribution, Strouhal number, and wake topology, and the results are compared with those obtained for a smooth cylinder. The numerical results demonstrate that increasing surface roughness promotes an earlier boundary-layer transition, shifts the critical Reynolds number, and significantly improves the prediction of the drag coefficient. The proposed approach partially compensates for the limitations of conventional RANS modelling while preserving its computational efficiency. Consequently, it provides a practical and computationally affordable methodology for predicting turbulent flows around circular cylinders in engineering applications such as offshore structures, marine components, heat exchangers, nuclear steam generators, and other thermal and energy systems.