Detached-eddy Simulation Formulation Research Articles

Standard and consistent Detached-Eddy Simulation for turbulent engine flow modeling: an application to the TCC-III engine

Multidimensional modeling of Cycle-to-Cycle Variability (CCV) has become a crucial support for the development and optimization of modern direct-injection turbocharged engines. In that sense, the only viable modeling options is represented by scale-resolving approaches such as Large Eddy Simulation (LES) or hybrid URANS/LES methods.Among other hybrid approaches, Detached-Eddy Simulation (DES) has the longest development story and is therefore commonly regarded as the most reliable choice for engineering-grade simulation. As such, in the last decade DESbased methods have found their way through the engine modeling community, showing a good potential in describing turbulence-related CCV in realistic engine configurations and at reasonable computational costs.In the present work we investigate the in-cylinder modeling capabilites of a standard two-equation DES formulation, compared to a more recent one which we call DESx. The DESx form differs from standard DES in the turbulent viscosity switch from URANS to LES-like behavior, which for DESx is fully consistent with Yoshizawa’s one-equation sub-grid scale model. The two formulations are part of a more general Zonal-DES (ZDES) methodology, developed and validated by the authors in a series of previous publications. Both variants are applied to the multi-cycle simulation of the TCC-III experimental engine setup, using sub-optimal grid refinement levels in order to stress the model limitations in URANS-like numerical resolution scenarios. Outcomes from this study show that, although both alternatives are able to ouperform URANS even in coarse grid arrangements, DESx emerges as sligthly superior and thus it can be recommended as the default option for in-cylinder flow simulation.

Detached-eddy simulation based on the [formula omitted] model

Detached-eddy simulation (DES) based on the v2-f Reynolds-averaged Navier–Stokes (RANS) model is developed and tested. The v2-f model incorporates anisotropy of near-wall turbulence, which is absent in other RANS models commonly used for DES. The v2-f RANS model is modified in order that in the DES formulation it reduces to a transport equation for the subgrid-scale kinetic energy in isotropic turbulence. First, three coefficients in the elliptic relaxation equation are modified and the modification is tested for RANS of channel flow. Next, the proposed v2-f DES model is formulated. The constant, CDES, required in the DES formulation is calibrated by simulating both decaying and statistically-steady isotropic turbulence. The formulation is tested for flow around a circular cylinder at a Reynolds number of 3900, in which case turbulence develops after separation. Simulations indicate that this model represents the turbulent wake nearly as accurately as the dynamic Smagorinsky model. For comparison, Spalart–Allmaras (SA) based DES is also included in the cylinder flow simulations. Finally, the proposed model is tested for a separated turbulent boundary layer on an airfoil. The delayed-DES (DDES) approach is found to be necessary. The v2-f DDES calculation gives good predictions for the aerodynamic forces and the pressure distribution measured in experiments. The proposed model is also compared with SA-DDES for the airfoil.

Assessment of DES Models for Separated Flow From a Hump in a Turbulent Boundary Layer

Separated flow past a hump in a turbulent boundary layer is studied numerically using detached-eddy simulation (DES), zonal detached-eddy simulation (ZDES), delayed detached-eddy simulation (DDES), and Reynolds-averaged Navier–Stokes (RANS) modeling. The geometry is smooth so the separation point is a function of the flow solution. Comparisons to experimental data show that RANS with the Spalart–Allmaras turbulence model predicts the mean-field statistics well. The ZDES and DDES methods perform better than the DES formulation and are comparable to RANS in most statistics. Analyses motivate that modeled-stress depletion near the separation point contributes to differences observed in the DES variants. The order of accuracy of the flow solver ACUSOLVE is also documented.

A Comparative Study of DES and URANS for Flow Prediction in a Two-Pass Internal Cooling Duct

The capabilities of the detached eddy simulation (DES) and the unsteady Reynolds averaged Navier-Stokes (URANS) versions of the 1988 k-ω model in predicting the turbulent flow field in a two-pass internal cooling duct with normal ribs is presented. The flow is dominated by the separation and reattachment of shear layers; unsteady vorticity induced secondary flows and strong streamline curvature. The techniques are evaluated in predicting the developing flow at the entrance to the duct and downstream of the 180deg bend, fully developed regime in the first pass, and in the 180deg bend. Results of mean flow quantities, secondary flows, and the average friction factor are compared to experiments and large-eddy simulations (LES). DES predicts a slower flow development than LES, whereas URANS predicts it much earlier than LES computations and experiments. However, it is observed that as fully developed conditions are established, the capability of the base model in predicting the flow is enhanced by the DES formulation. DES accurately predicts the flow both in the fully developed region as well as the 180deg bend of the duct. URANS fails to predict the secondary flows in the fully developed region of the duct and is clearly inferior to DES in the 180deg bend.

LES and DES Investigations of Turbulent Flow over a Sphere at Re = 10,000

Large Eddy Simulation (LES) using a dynamic Smagorinsky type subgridstress (SGS) model and Detached Eddy Simulation (DES) are applied toprediction and investigation of the flow around a sphere at a Reynoldsnumber of 104 in the subcritical regime. In this regime the boundarylayers at separation are laminar, and transition to turbulence occursfarther downstream in the separated shear layers via Kelvin–Helmholtz(K–H) instabilities. The dynamic eddy viscosity model of Germano et al.(Physics of Fluids3 (1991) 1760–1765) is used in the LES, while the current implementation of the DESemploys a formulation based on the Spalart–Allmaras (S–A) model. DES isa hybrid approach in which the closure is a modification to theproduction/destruction term of the original Reynolds-AveragedNavier–Stokes (RANS) model, reducing to RANS in the attached regions,and to LES away from the wall. In the present work where we simulate theflow over a sphere in the subcritical regime in which the boundarylayers at separation are laminar, DES can be viewed as LES with adifferent SGS model. Effects of the discretization scheme used toapproximate the convective terms are considered, along with sensitivityof predictions to changes in the additional model coefficient, CDES, in the DES formulation. DES and LES yield similar predictions of the wakestructure, large-scale vortex shedding and the Strouhal numberassociated with the low frequency mode in the wake. Predictions ofquantities such as the drag coefficient, wake frequencies, position oflaminar separation on the sphere, and the mean pressure andskin-friction distributions along the sphere are in good agreement withthe measurements of Achenbach (Journal of Fluid Mechanics54 (1972) 565–575). Predictions of the primaryReynolds shear stress, turbulent kinetic energy, eddy viscosity, andturbulent dissipation for the two models are also similar. In addition,both models successfully resolve the formation of the vortex tubes inthe detached shear layers along with the value of the Strouhal numberassociated with the high frequency instability mode, provided that thelevel of numerical dissipation introduced by the discretization schemeremains sufficiently low. Flow physics investigations are focused onunderstanding the wake structure in the subcritical regime.