Separation prediction and roughness effect simulation of iced airfoils based on a modified k–ω–γ transition model

• A modified k –ω–γ transition model with separation correction and roughness simulation is developed for icing simulations. • Accurately predicts flow separation and aerodynamic forces over iced airfoils . • Achieves high accuracy in predicting maximum lift with errors below 5 %. • Demonstrates capability to model early transitions induced by surface roughness . • Outperforms the integral boundary-layer approach in predicting heat transfer in separated flows . Ice accretion significantly complicates flow separation over airfoils , posing substantial challenges for turbulence models. In this work, a modified k − ω − γ transition model, which incorporates a separation correction and a roughness effect simulation, is developed for accurately predicting complex flows over iced airfoils . The model's ability to predict large separations is validated in several cases, including a periodic hill and airfoils with glaze ice and rime ice. The results indicate that the modified model can accurately predict the flow separation and aerodynamic force characteristics. At the stall angle of attack, the relative error between the modified k − ω − γ model and the experimental maximum lift coefficient is less than 5.0 %, in contrast, the error of the SST model results ranges from 22.0 % to 35.8 %. Additionally, a rough flat plate and several rough airfoils without or with ice are adopted to evaluate the model's ability to simulate roughness effects. The results show that the model can capture the early transition induced by different roughness heights. This approach also has advantages over the integral boundary-layer approach in calculating the heat transfer coefficient in the presence of separated flow.