Ganbo Deng

Three-Dimensional Flow Computation with Reynolds Stress and Algebraic Stress Models

A quadratic explicit algebraic stress model (EASM) that takes into account the variation of production-to-dissipation rate ratio is compared with an implicit algebraic stress model (ASM) and with their parent Reynolds stress model (RSM) in this paper. A new implementation of ASM model where the turbulent eddy viscosity provided by the explicit solution is employed is found to be robust. Computations for ship flows at model and full scale are performed to assess the accuracy of different models. Explicit and implicit algebraic stress models give similar prediction for the flow investigated. The RSM model provides better prediction in the region dominated by convex curvature. However, no much improvement is observed near the concave surface.

A new fully coupled method for computing turbulent flows

Abstract This work discusses the computation of the steady, turbulent, two-dimensional incompressible viscous flow on structured cell-centered collocated grids. A rather new computational approach – the so-called fully coupled procedure with defect correction technique – is presented as an alternative both to classical decoupled approaches (SIMPLE [Int. J. Heat Mass Transfer 15 (1972) 1787–1806], PISO [J. Comput. Phys. 62(1) (1986) 40–65] and variants) and to weakly coupled solution methods of Vanka type [Fifth symposium on Turbulent Shear flows, 1985. p. 20:27–32, J. Comput. Phys. 65 (1986) 138–158]. Its main features will be highlighted and detailed. This strategy is evaluated on two demanding test cases (the simulation of the separated flow past an AS240-B airfoil at high incidence (19°, Re=2×10 6 ) and the simulation of the wake flow behind a two-dimensional hill ( Re =60 000 )), for which documented experimental data are available. Both robustness and computational efficiency of the new approach are shown.

RANS prediction of the KVLCC2 tanker in head waves

The present study is devoted to the computation of the KVLCC2 tanker in head wave with free heave and pitch motion. A RANS solver using finite-volume discretization and free-surface capturing approach is employed for the computation. Free ship motion is captured with a mesh deformation approach. Three different wave lengths (0.6Lpp, 1.1Lpp and 1.6Lpp) are computed. We focus on numerical uncertainty estimation in this paper. For each test case, three different meshes and at least three different time steps have been used to access both time and spatial discretization error. Additional computations with different setups aimed at identifying different numerical discretization errors will also be performed. It is demonstrated that special attention needs to be paid to time discretization. To keep the same time accuracy, time step needs to be reduced on fine mesh for such kind of unsteady free-surface computation involving important pitch or roll motion.

2-D COMPUTATIONS OF UNSTEADY FLOW PAST A SQUARE CYLINDER WITH THE BALDWIN-LOMAX MODEL

The unsteady turbulent flow past a square cylinder is calculated with the Baldwin-Lomax Model. The influence of the numerical schemes is studied. Comparisons with experimental data and with previous calculations using a Reynolds stress model and 3-D large eddy simulations are presented. It is shown that a good prediction is obtained with a Baldwin-Lomax model by using the recently proposed CPI discretization scheme.

Turbulent flow prediction around appended hulls

Two appended hull configurations have been simulated using all hexahedral unstructured grids. Numerical uncertainty is assessed with grid refinement. Influence of turbulence model and wall function approach have been investigated. Scale effect has been studied for one configuration. Numerical results are validated both with measurement data and with computational results using structured grid when possible.

Hybrid LES–RANS-Coupling for Complex Flows with Separation

The investigations carried out within the French-German DFG-CNRS Research Initiative on ‘LES of Complex Flows (FOR 507)’ aimed at the development and application of hybrid LES-RANS methods for turbulent flows with separation. The objectives were twofold.On the one hand, the performance of a well-established hybrid approach, namely the detached-eddy simulation, was studied based on really challenging test cases such as the flow over the 3D hill and around the Willy car model. On the other hand, a new hybrid LES-RANS methodologywas set up which aims at overcoming certain drawbacks associated with DES. It relies on the coupling of a near-wall RANS model with LES for the outer flow which allows an appropriate representation of large-scale flow phenomena. In order to take the anisotropies in the near-wall region rigorously into account, an explicit algebraic Reynolds stress model was chosen and its performance in comparison with linear eddy-viscosity models was analyzed for different test cases such as the flow over 2D hills and in a 3D diffuser. A further advantage of the new hybrid method is that no interface predefinition is required. Instead the interface is dynamically determined on-the-fly based on instantaneous physical quantities which guarantees that local changes in the flow are accounted for.

Direct Numerical and Statistical Simulation of Turbulent Boundary Layer Flows with Pressure Gradient

A zonal grid approach has been developed as a means to reduce the large number of grid points required for direct numerical simulation (DNS) of boundary layers. The case of a boundary layer flow with zero pressure gradient shows that computational resources can be saved by zonal DNS compared to a full fine grid DNS and that the accuracy of the results is increased as compared to a completely coarse grid. The zonal grid approach is also used for DNS of boundary layer flow with adverse pressure gradient. DNS results are compared with experimental data. In order to evaluate the performance of existing turbulence models, RANS simulations have been performed for the same boundary layer under adverse pressure gradient. Several models, ranging from two-equation models to Reynolds stress transport models have been tested and the results are compared with DNS and experimental data.

Can adaptive grid refinement produce grid-independent solutions for incompressible flows?

Abstract This paper studies if adaptive grid refinement combined with finite-volume simulation of the incompressible RANS equations can be used to obtain grid-independent solutions of realistic flow problems. It is shown that grid adaptation based on metric tensors can generate series of meshes for grid convergence studies in a straightforward way. For a two-dimensional airfoil and the flow around a tanker ship, the grid convergence of the observed forces is sufficiently smooth for numerical uncertainty estimation. Grid refinement captures the details of the local flow in the wake, which is shown to be grid converged on reasonably-sized meshes. Thus, grid convergence studies using automatic refinement are suitable for high-Reynolds incompressible flows.

Creating Free-Surface Flow Grids with Automatic Grid Refinement

The objective of this work is to create grids for free-surface water flow simulation entirely with automatic grid refinement. It is shown why it is necessary to refine the mesh iteratively as the solution converges and why refinement and derefinement of hexahedral cells must be treated anisotropically.The proposed refinement criterion is a combination of the pressure Hessian with refinement at the free surface, in order to capture the flow which drives the surface motion and the position of the surface itself. Smoothing is needed in the computation of the Hessian in order to remove oscillations in the pressure, the pressure Hessian is extrapolated through the free surface to remove its discontinuity there.Two test cases confirm that effective fine meshes for wave computation can be created with the proposed automatic refinement procedure.

Direct versus statistical simulation of accelerated/retarded and separating/reattaching turbulent boundary layers

Direct and statistical simulations have been performed in order to demonstrate their strengths and weaknesses in complex turbulent flow situations. The first flow to be predicted is Watmuff’s [20] favourable/adverse pressure gradient boundary layer at a Reynolds number of 670, based on inlet momentum thickness and freestream velocity. The second case is a separating/reattaching turbulent boundary layer which has been investigated experimentally by Kalter and Fernholz [4] at Re θ = 1500. Results for global and local statistical quantities are compared. They underline the need to use sophisticated turbulence models for reliable prediction of complex flows.