A new scale-adaptive hybrid Reynolds-averaged Navier–Stokes and large eddy simulation model considering rotational effect for separated flow predictions

Hybrid Reynolds-averaged Navier–Stokes (RANS) and large eddy simulation (LES) methods, abbreviated as hybrid RANS-LES, have garnered increasing attention due to their balanced integration of RANS' computational economy and LES' resolution fidelity. In this study, a new scale-adaptive hybrid RANS-LES model considering the rotational effect for separated flow predictions, named as the scale-adaptive rotation correction model, abbreviated as SA-RCM, is developed by introducing a novel damping function into the turbulent viscosity of the shear stress transport k-omega (SST k–ω) model. This function is constructed through the ratio of unresolved-to-total turbulence kinetic energy derived from the turbulence energy spectrum, and incorporated rotation and scale-adaptive corrections for grid length scale inspired by delayed detached eddy simulation method. The SA-RCM model is validated by four benchmark cases: swirling flow through an abrupt axisymmetric expansion, periodic hill flow, rotating channel flow and flow in a centrifugal pump impeller. Comparative analysis with experimental data or direct numerical simulation results demonstrates superior prediction of average velocity profiles and Reynolds stresses of the SA-RCM model in high-curvature recirculation zones and rotational-dominated regions, while adaptively adjusting turbulent viscosity to avoid structural distortions inherent in very large eddy simulation (VLES) model. In the simulation of rotating machinery, the SA-RCM model shows better resolving ability than the VLES model, which can analyze the flow field information under the design condition and capture the number, position, and range of stall vortices under the complex small flow conditions more accurately.