Transition model assessment and modification in low-pressure turbine cascade

In order to assess and improve the prediction accuracy of existing transition models, numerical investigations on low-pressure turbine cascade T106-EIZ are conducted with five models: laminar, turbulence, shear stress transport (SST) low Reynolds model, k-kl-ω model and γ-Re_θ model, by using computational fluid dynamics (CFD) software FULENT 12.1. The laminar and turbulence models are used as the baseline models for testing the last three transition models. Their ability to accurately simulate transition and related parameters is also evaluated by comparison with the experimental results. Subsequently, the comparison results and mechanisms of models are analyzed. The results show that no model can accurately predict separated flow transition and wake-induced transition, while the γ-Re_θ model is slightly better than the others. A new method of modifying the γ-Re_θ model is proposed by correcting the constant c_a1 and c_a2 in the transport equation of intermittency factor, and the results show that this method is quite effective in improving prediction accuracy.