Reynolds Number Effect on Transitional Delayed Detached-Eddy Simulation

Abstract

The accurate prediction of a wide range of operating conditions is desired by numerical methods. In particular, off-design conditions are a major challenge for steady Reynolds-averaged Navier-Stokes (RANS) methods. In this study, we evaluate the delayed detached-eddy simulation (DDES) method for these flow conditions. Therefore, we consider a transitional linear compressor cascade, which has been measured experimentally at DLR for different Reynolds numbers and inflow angles. For the lowest Reynolds number, numerical large-eddy simulation (LES) reference data is available additionally. We begin with a basis validation for Re = 150 000 at the aerodynamic design point before evaluating off-design conditions, characterized by positive and negative incidence angles, for this Reynolds number. This is followed by a Reynolds number variation up to Re = 900 000 for two off-design inflow angles. The potential for the transitional DDES is visible for the lowest Reynolds number where reference data is met well with a noticeable reduction of computational resources. The Reynolds number variation yield two-fold results. While the negative incidence confirmed the beneficial behavior of DDES, the positive incidence reveals deficits of the DDES for this specific setup. We give explanations for the improved, but also the limited behavior and discuss the mesh requirements, directly linked the computational requirements.