Large Eddy Simulation Regions Research Articles

Adaptive detached eddy simulation of turbulent combustion with the subgrid dissipation concept

Detached eddy simulation has become a widely used method in eddy simulations due to its balance between cost and accuracy. The recently developed subgrid dissipation concept (SDC) combustion model [Liu et al., “On the subgrid dissipation concept for large eddy simulation of turbulent combustion,” Combust. Flame 258, 113099 (2023)] is found to be more reasonable and accurate than the conventional eddy dissipation concept model in large eddy simulation (LES). In this paper, the SDC model is adapted to the ℓ2-ω adaptive detached eddy simulation framework, named DES-SDC. The required key quantities, including the fine structure mass fraction and dissipation rate, are appropriately blended across Reynolds-averaged Navier–Stokes and LES regions. The DES-SDC approach is validated using premixed bluff body stabilized flame, non-premixed swirl flame, and premixed swirl flame with complex geometry. It is much more tolerant to coarse mesh resolution than pure LES, yet it preserves the capability of resolving the key unsteady feature critical for the combustion process, as it is designed to be. The DES-SDC approach is relatively insensitive to the grid resolution. The present research provides a promising approach for accurately simulating practical unsteady turbulent combustion problems at an affordable computational cost.

Advanced detached-eddy simulation of the MD 30P-30N three-element airfoil

An experimental version in the Detached-Eddy Simulation (DES) family (called Advanced DES or ADES) is introduced and tested on a geometry that is fairly complex but two-dimensional. The essential change in ADES is that the user is given control of the regions treated with full turbulence modelling (RANS) and those treated with Large-Eddy Simulation (LES). This zonal character makes the approach more powerful, but less practical, so that in its current state it is not ready for industrial CFD. The grid requirements of the two regions are very different, and multi-block grid structure is natural. Another key feature is a Volumetric Synthetic Turbulence Generator (VSTG), installed to feed the LES region with viable resolved turbulence, so that the resolved Reynolds stresses rapidly substitute for the modelled Reynolds stresses present in the RANS region. The VSTG operates in a volume, rather than on a surface and can be active in attached boundary layers, at a trailing edge, or after separation. The well-known McDonnell-Douglas 30P-30N airfoil is simulated with periodic lateral boundary conditions. The VSTG is successful, and the desired nature of simulation is obtained in each region. ADES involves zonal decisions, but appears robust. An inertial range is clearly indicated in frequency spectra. A grid-refinement study is included, as well as variations in lateral domain size and STG positions; this led to a matrix of 11 simulations. Cases are shown at four angles of attack and with three RANS models in addition to ADES. Pressure and friction distributions and velocity and shear stress profiles are compared in detail. The prospects for an evolution of ADES into a practical routine approach in the long term are discussed.

A dynamic version of the improved delayed detached-eddy simulation based on the differential Reynolds-stress model

A dynamic version of the improved delayed detached-eddy simulation (IDDES) based on the differential Reynolds-stress model (RSM), referred to as the RSM-DynIDDES, is developed by applying the dynamic Smagorinsky subgrid model to the large eddy simulation (LES) branch of the IDDES. The RSM-DynIDDES simulates the periodic hills flow after a basic numerical validation for the decaying isotropic turbulence simulation. Well-predicted velocity profiles and R eynolds stress distributions are obtained by the RSM-DynIDDES in the periodic hills flow. The simulation results indicate that the RSM-DynIDDES can capture more small-scale vortex structures in the LES region away from the wall than the original RSM-based IDDES (RSM-IDDES). The RSM-DynIDDES is also employed in simulating the transonic buffeting of a launch vehicle with a payload fairing. The numerical results have been compared with that of the RSM-IDDES. It is found that the RSM-DynIDDES can improve turbulence resolution in the off-wall region while retaining the advantages of the original RSM-IDDES in simulating the instability process of the free shear layer.

Improvement of double-buffer problem in LES–RANS interface region by introducing an anisotropy-resolving subgrid-scale model

The “double-buffer problem” has been regarded as a crucial concern for the strategy behind the hybrid large eddy simulation (LES)/Reynolds-averaged Navier–Stokes (RANS) model (or HLR model, for short). Such models are likely to show unphysical mean-velocity distributions in the LES–RANS interface region, where “super-streak structures” also appear that look like low-speed streaks generated in the near-wall region of wall turbulence. To overcome this difficulty, the stochastic backscatter model, in which the vortex structures in the interface region are divided into smaller scales, holds promise due to the effect of random source term imposed in the momentum equation. Although this method is effective, several parameters must be prescribed and their specification process is arbitrary and ambiguous. An alternative advanced HLR model has been proposed, in which an anisotropy-resolving subgrid-scale (SGS) model was adopted in the LES region as well as a one-equation nonlinear eddy viscosity model in the RANS region. Previous investigations indicated that this HLR model did not exhibit or, at least, largely reduced the “double-buffer problem” in the mean-velocity distribution, with no special treatment being applied. The main purpose of the present study is to reveal why this HLR model improves the predictive performance in the LES–RANS interface region. Specifically, we focus on the role of the extra anisotropic term introduced in the SGS model, finding that it plays an important role in enhancing vortex structures in the interface region, leading to a considerable improvement in model performance.

Implementation and Evaluation of an Embedded LES-RANS Solver

In the current work, we present the development and application of an embedded large-eddy simulation (LES) - Reynolds-averaged Navier Stokes (RANS) solver. The novelty of the present work lies in fully embedding the LES region inside a global RANS region through an explicit coupling at the arbitrary mesh interfaces, exchanging flow and turbulence quantities. In particular, a digital filter method (DFM) extracting mean flow, turbulent kinetic energy and Reynolds stress profiles from the RANS region is used to provide meaningful turbulent fluctuations to the LES region. The framework is developed in the open-source computational fluid dynamics software OpenFOAM. The embedding approach is developed and validated by simulating a spatially developing turbulent channel flow. Thereafter, flow over a surface mounted spanwise-periodic vertical fence is simulated to demonstrate the importance of the DFM and the effect of the location of the RANS-LES interface. Mean and second-order statistics are compared with direct numerical simulation (DNS) data from the literature. Results indicate that feeding synthetic turbulence at the LES interface is essential to achieve good agreement for the mean flow quantities. However, in order to obtain a good match for the Reynolds stresses, the LES interface needs to be placed sufficiently far upstream, which in the present case was six spoiler heights before the fence. Further, a realistic spoiler configuration with finite-width in the spanwise direction and inclined at 30 degrees was simulated using the embedding approach. As opposed to the vertical fence case this is a genuinely (statistically) three-dimensional case and a very good match with mean and second-order statistics was obtained with the experimental data. Finally, in order to test the present solver for high sub-sonic speed flows the flow over an open cavity was simulated. A good match with reference data is obtained for mean and turbulence profile comparisons. Tones in the pressure spectra were predicted reasonably well and an overall sound pressure level with a maximum deviation of 2.6 dB was obtained with the present solver when compared with the experimental data.

Investigation of the sensitivity of turbulent closures and coupling of hybrid RANS‐LES models for predicting flow fields with separation and reattachment

SummaryHybrid models have found widespread applications for simulation of wall‐bounded flows at high Reynolds numbers. Typically, these models employ Reynolds‐averaged Navier–Stokes (RANS) and large eddy simulation (LES) in the near‐body and off‐body regions, respectively. A number of coupling strategies between the RANS and LES regions have been proposed, tested, and applied in the literature with varying degree of success. Linear eddy‐viscosity models (LEVM) are often used for the closure of turbulent stress tensor in RANS and LES regions. LEVM incorrectly predicts the anisotropy of Reynolds normal stress at the RANS‐LES interface region. To overcome this issue, use of non‐linear eddy‐viscosity models (NLEVM) have started receiving attention. In this study, a generic non‐linear blended modeling framework for performing hybrid simulations is proposed. Flow over the periodic hills is used as the test case for model evaluation. This case is chosen due to complex flow physics with simplified geometry. Analysis of the simulations suggests that the non‐linear hybrid models show a better performance than linear hybrid models. It is also observed that the non‐linear closures are less sensitive to the RANS‐LES coupling and grid resolution. Copyright © 2016 John Wiley & Sons, Ltd.

Zonal PANS: evaluation of different treatments of the RANS–LES interface

ABSTRACTThe partially Reynolds-averaged Navier–Stokes (PANS) model can be used to simulate turbulent flows either as RANS, large eddy simulation (LES) or DNS. Its main parameter is fk whose physical meaning is the ratio of the modelled to the total turbulent kinetic energy. In RANS fk = 1, in DNS fk = 0 and in LES fk takes values between 0 and 1. Three different ways of prescribing fk are evaluated for decaying grid turbulence and fully developed channel flow: fk = 0.4, fk = k3/2tot/ϵ and, from its definition, fk = k/ktot where ktot is the sum of the modelled, k, and resolved, kres, turbulent kinetic energy. It is found that the fk = 0.4 gives the best results. In Girimaji and Wallin, a method was proposed to include the effect of the gradient of fk. This approach is used at RANS– LES interface in the present study. Four different interface models are evaluated in fully developed channel flow and embedded LES of channel flow: in both cases, PANS is used as a zonal model with fk = 1 in the unsteady RANS (URANS) region and fk = 0.4 in the LES region. In fully developed channel flow, the RANS– LES interface is parallel to the wall (horizontal) and in embedded LES, it is parallel to the inlet (vertical). The importance of the location of the horizontal interface in fully developed channel flow is also investigated. It is found that the location – and the choice of the treatment at the interface – may be critical at low Reynolds number or if the interface is placed too close to the wall. The reason is that the modelled turbulent shear stress at the interface is large and hence the relative strength of the resolved turbulence is small. In RANS, the turbulent viscosity – and consequently also the modelled Reynolds shear stress – is only weakly dependent on Reynolds number. It is found in the present work that it also applies in the URANS region.

3-D hybrid LES-RANS model for simulation of open-channel T-diversion flows

The study of flow diversions in open channels plays an important practical role in the design and management of open-channel networks for irrigation or drainage. To accurately predict the mean flow and turbulence characteristics of open-channel dividing flows, a hybrid LES-RANS model, which combines the large eddy simulation (LES) model with the Reynolds-averaged Navier-Stokes (RANS) model, is proposed in the present study. The unsteady RANS model was used to simulate the upstream and downstream regions of a main channel, as well as the downstream region of a branch channel. The LES model was used to simulate the channel diversion region, where turbulent flow characteristics are complicated. Isotropic velocity fluctuations were added at the inflow interface of the LES region to trigger the generation of resolved turbulence. A method based on the virtual body force is proposed to impose Reynolds-averaged velocity fields near the outlet of the LES region in order to take downstream flow effects computed by the RANS model into account and dissipate the excessive turbulent fluctuations. This hybrid approach saves computational effort and makes it easier to properly specify inlet and outlet boundary conditions. Comparison between computational results and experimental data indicates that this relatively new modeling approach can accurately predict open-channel T -diversion flows.

Prediction of Transonic Buffet by Delayed Detached-Eddy Simulation

A delayed detached-eddy simulation of the transonic buffet over a supercritical airfoil is performed. The turbulence modeling approach is based on a one-equation closure, and the results are compared to an unsteady Reynolds-averaged Navier–Stokes simulation using the same baseline model as well as experimental data. The delayed detached-eddy simulation successfully predicts the self-sustained unsteady shock-wave/boundary-layer interaction associated with buffet. When separation occurs, the flow exhibits alternate vortex shedding and a spanwise undulation. The method also captures secondary fluctuations in the boundary layer that are not predicted by unsteady Reynolds-averaged Navier–Stokes simulation. A map of flow separation emphasizes the differences between the delayed detached-eddy simulation and unsteady Reynolds-averaged Navier–Stokes flow topologies. Statistical pressure distributions and velocity profiles help assess the performance of each model. They indicate that the delayed detached-eddy simulation tends to overestimate the flow unsteadiness near the trailing edge. Instantaneous distributions of Reynolds-averaged Navier–Stokes and large-eddy simulation regions during buffet show that the delayed detached-eddy simulation enforces Reynolds-averaged Navier–Stokes mode near the airfoil even when the boundary layer gets very thick.

Partially integrated transport modeling method for turbulence simulation with variable filters

The basis of the partially integrated transport modeling method was introduced in papers of Schiestel and Dejoan [“Towards a new partially integrated transport model for coarse grid and unsteady turbulent flow simulations,” Theor. Comput. Fluid Dyn. 18, 443 (2005)] and Chaouat and Schiestel [“A new partially integrated transport model for subgrid-scale stresses and dissipation rate for turbulent developing flows,” Phys. Fluids 17, 065106 (2005)]. This method provides a continuous approach for hybrid Reynolds averaged Navier-Stokes (RANS)-large eddy simulation (LES) simulations with seamless coupling between RANS and LES regions. The method, like in usual LES techniques, makes use of space filtering in the turbulent field. In the foundation papers cited above and in the main applications considered so far, the filter width has been supposed constant or at least slowly varying. In the present paper, we examine the effect of variable filter width in the model equations and how to account for this effect in practical numerical simulations. With the aim to illustrate the theoretical development of the effect of varying filter width in time and space on the governing equations of mass, momentum, and turbulence model, and to show the usefulness of the proposed approach, we perform then numerical simulations of isotropic decaying turbulence.

Zonal Detached-Eddy Simulation of an Airfoil in Poststall Condition

This paper presents a zonal detached-eddy simulation of a leading-edge stall airfoil in poststall conditions. Different detached-eddy-simulation-type methodologies are first evaluated and discussed and the paper presents in detail the results obtained with the mode 1 of the zonal detached-eddy simulation that is based on a prescribed switch between the Reynolds-averaged Navier–Stokes and the large-eddy simulation regions. The results obtained with the selected approach are presented and compared to detailed experimental results including unsteady pressure measurements and unsteady velocity flowfield measurements acquired using time-resolved particle image velocimetry and laser Doppler velocimetry. Important features of the flow are correctly captured by the computation and good agreements are observed with experiments on pressure distribution, separation location and aerodynamic forces. The unsteady properties of the calculation are also compared with unsteady measurements showing that the frequency indicating that the vortex shedding is correctly predicted in the computation. Detailed analysis of the mixing-layer and the wake properties highlights some deficiency of the computations that are fully analyzed. Especially, the grid resolution in the early stages of the mixing layer appears to be of primary importance for the present application. The detailed comparisons of the zonal detached-eddy simulation result with the experimental data allow general requirements to be drawn for further improvements.

Derivation, implementation, and initial testing of a compressible wall-layer model

Traditional wall functions have been used successfully for decades to decrease the computational cost for obtaining solutions to incompressible flows with equilibrium turbulence. However, these traditional, analytic wall functions are poorly suited for more complex flows. The present work describes an alternative approach named the ‘diffusion model’. The diffusion model is a subgrid model developed by Blottner and Bond that solves a system of ODEs in the near-wall region instead of assuming an analytic profile. The diffusion model has previously been shown to reproduce profiles of various turbulence models through the log layer with fixed outer boundary conditions. This paper documents the implementation and the testing of the diffusion model fully coupled with a 3-D Reynolds-averaged Navier–Stokes (RANS) algorithm, using the Spalart–Allmaras model. The results indicate that the coupled algorithm is valid when the interface between diffusion model and the 3-D RANS algorithm is below the outer edge of the log layer. Although the current work focuses on using steady 3-D RANS outside of the log layer, modeling assumptions introduced in the current work could be used to derive a diffusion model for coupling with a large eddy simulation region. Published in 2010 by John Wiley & Sons, Ltd.

Compressibility effects on the vortical flow over a 65° sweep delta wing

The aim of this study is to investigate numerically the compressibility effects on the vortical flow developing over the VFE-2 delta wing. This wing is equipped with a sharp leading-edge and a sweep angle equal to 65°. The angle of attack is set equal to 25.5°, and two different freestream Mach numbers are considered: M∞=0.4 and M∞=0.8. The simulations are based on a turbulent modeling coupling Delayed Detached Eddy Simulation and Zonal Detached Eddy Simulation approaches, allowing a faster decay of the eddy-viscosity in Large Eddy Simulation regions. Such a method allows a good agreement with available experimental data. These computations highlight a modification of the flow behavior with an increase in the freestream Mach number. Indeed, the leading-edge vortex moves toward the wing and its core is dilated. Moreover, it is demonstrated in this study that the leading-edge vortex interacts with shock wave at midchord. This interaction induces a decay of the Rossby number in the leading-edge region and then an earlier breakdown in the transonic regime than in the subsonic one. Furthermore, several cross-flow shock waves deeply alter the aerodynamic field, in particular, a shock wave sitting beneath the leading-edge vortex. It is shown that this cross-flow shock wave moves in the lateral direction and that the frequency of this shock unsteadiness corresponds to the one classically encountered in shock-induced separation problems. As a result, the probability density function of the lateral location of the vortex core is altered in comparison with the subsonic case.

Hybrid LES–RANS: back scatter from a scale-similarity model used as forcing

A dissipative scale-similarity subgrid model was recently proposed in which only the dissipative part of the subgrid stresses was added to the momentum equations. This was achieved by adding the gradient of a subgrid stress only when its sign agreed with that of the corresponding viscous term. In the present work, this idea is used the other way around as forcing in hybrid large eddy simulation-Reynolds-averaged Navier-Stokes: only the part of a subgrid stress term that corresponds to back scatter is added to the momentum equations. The forcing triggers resolved turbulence in the transition region between the unsteady Reynolds-averaged Navier-Stokes and large eddy simulation regions. The new approach is evaluated for fully developed channel flow at Re(tau)=4000. It is found that the forcing indeed does increase the resolved turbulence in the transition region. The magnitude of the production (i.e. back scatter) due to forcing in the equation for resolved kinetic energy is of the order of that due to the usual strain-rate production term. The present approach of using back scatter from a scale-similarity model can also probably be useful for triggering transition.

Hybrid RANS/LES method for wind flow over complex terrain

AbstractThe use of Large Eddy Simulation (LES) to predict wall‐bounded flows has presently been limited to low Reynolds number flows. Since the number of computational grid points required to resolve the near‐wall turbulent structures increase rapidly with Reynolds number, LES has been unattainable for flows at high Reynolds numbers. To reduce the computational cost of traditional LES, a hybrid method is proposed in which the near‐wall eddies are modelled in a Reynolds‐averaged sense. Close to walls, the flow is treated with the Reynolds‐averaged Navier–Stokes (RANS) equations (unsteady RANS), and this layer acts as wall model for the outer flow handled by LES. The well‐known high Reynolds number two‐equation k − ϵ turbulence model is used in the RANS layer and the model automatically switches to a two‐equation k − ϵ subgrid scale stress model in the LES region. The approach can be used for flow over rough walls. Previous attempts of combining RANS and LES has resulted in unphysical transition regions between the two layers, but the present work improves this region by using a stochastic backscatter model.To demonstrate the ability of the proposed hybrid method, simulations are presented for wind flow over the Askervein hill. Comparisons show that both RANS and the new hybrid model are able to capture the simple flow windward of the hill. In the complex wake region, however, the RANS and the hybrid model give different results. The RANS model predicts the mean velocity well but underestimates the turbulent kinetic energy, whereas the new method captures the high turbulence levels well but underestimates the mean velocity. The presented results are for a relative mild configuration of complex terrain, but the proposed method can also be used for highly complex terrain where the benefits of the new method is expected to be larger. Copyright © 2009 John Wiley & Sons, Ltd.

Improvement of Delayed-Detached Eddy Simulation Applied to Separated Flow over Missile Fin

This paper presents computational simulations of the flow over a 50-deg sweep missile fin for an angle of attack equal to 25 deg. For such an angle of attack, the flow is expected to fully separate. Nevertheless, Reynolds-averaged Navier-Stokes computations still predict the presence of a leading-edge vortex. Then, hybrid Reynolds-averaged Navier-Stokes/large eddy simulation methods are assessed. In particular, this study focuses on the delayed-detached eddy simulation and on a proposed extension of this method (EDDES), in which the objective is to accelerate the destruction of the eddy viscosity in large eddy simulation regions. These two methods are assessed in the case of a boundary-layer flow, and it is shown that the extended delayed-detached eddy simulation behaves as the delayed-detached eddy simulation method. Nevertheless, in the case of a fully separated flow downstream from a backward facing step, the resolved fluctuations obtained with the extended delayed-detached eddy simulation method are in better agreement with the experimental data than those of the delayed-detached eddy simulation computation. Finally, these two methods improve the description of the flow over the missile fin, predicting a fully separated flow. However, the extended delayed-detached eddy simulation ensures a faster development of instabilities than the delayed-detached eddy simulation and the agreement with the pressure distribution obtained with pressure sensitive paint is much better with the proposed modification of delayed-detached eddy simulation.

Log-layer mismatch and commutation error in hybrid RANS/LES simulation of channel flow

Hybrid approach combining large eddy simulation (LES) with the Reynolds-averaged Navier–Stokes equation (RANS) is expected to accurately simulate wall-bounded turbulent flows at high Reynolds numbers. As an important issue in developing hybrid methods, it is known that the log layers in the RANS and LES regions are not lined up in hybrid RANS/LES simulations of channel flow. Although several methods including additional filtering near the RANS/LES interface have been proposed to eliminate the log-layer mismatch, there is no obvious physical justification for the methods and some ad hoc tuning is necessary. In this work, the commutation error terms in the filtered velocity equations are investigated to justify the method of additional filtering. It is shown that the additional filtering can be considered as a finite difference approximation to extra terms due to the non-commutivity between the hybrid filter and the spatial derivative. Moreover, an expression determining the filter width and its location for the additional filtering is obtained. To validate the expression, a hybrid simulation of channel flow is carried out. The additional filtering with the filter width derived is shown to be effective in eliminating the log-layer mismatch and improving the mean velocity profile.

Detached eddy simulation of a synthetic jet for flow control

Synthetic jets have been proposed for efficient flow control due to their unique zero-net-mass-flux feature and the conceptual simplicity of the system. However, the effectiveness of the control is heavily dependent upon the complicated flow interaction of the jets with the flow stream on which the jet is applied. To understand the flow control mechanism better, a numerical simulation method is investigated in this paper. Using a computational fluid dynamics technique, the detached eddy simulation for an isolated synthetic jet for flow control is presented. A cubic-root filter is introduced with continuous eddy viscosity variation at the switching point between the Reynolds-averaged Navier—Stokes solution region and the large eddy simulation region. The intention is to explore the effects of unstructured grids for such a filtering strategy. A dissipation-controlled Roe scheme is applied and a dynamic grid technique is employed to implement the periodically moving diaphragm. The results obtained are compared with the experimental data and a previous study using structured grids.

Improved turbulence generation techniques for hybrid RANS/LES calculations

Hybrid turbulence simulations that use a combination of Reynolds-averaged Navier–Stokes (RANS) equations and large-eddy simulation (LES) are becoming increasingly popular to increase predictive accuracy in complex flow situations without the cost of full large-eddy simulations. In a typical application, the RANS equations are solved in the equilibrium or near-equilibrium regions of the flow, while LES equations are used in regions where the flow is away from equilibrium. A difficulty in performing hybrid simulations stems from the mismatch between resolved turbulent scales between the RANS and LES regions: in the RANS region all turbulent scales are modeled, while in the LES region the energy containing scales are resolved. Interfacing these two regions requires the generation of turbulent eddies capable of supporting the Reynolds stresses in the LES field. Previous work by the authors proposed a combination of synthetic turbulence and a controlled forcing method that was capable of quickly generating these eddies. In this paper, we extend this earlier work by investigating the physical meaning of the model parameters to understand their effect on the flow development. We find that by matching the time-averaging window to the local integral time scale of the flow improved results can be obtained compared with previous work. We also determined the range of the controller parameters that give robust results. The method was applied to non-equilibrium boundary-layer flows far from the calibration case and was found to perform well. This technique gives best results if an outer-flow perturbation drives the flow; when the inner-layer dynamics are critical, less rapid establishment of the turbulence is achieved. Even in the worst result, however, the controller was able to establish a realistic flow within five boundary layer thicknesses of the end of the controlled region.

Hybrid RANS/LES Method for High Reynolds Numbers, Applied to Atmospheric Flow over Complex Terrain

The use of Large-Eddy Simulation (LES) to predict wall-bounded flows has presently been limited to low Reynolds number flows. Since the number of computational grid points required to resolve the near-wall turbulent structures increase rapidly with Reynolds number, LES has been unattainable for flows at high Reynolds numbers. To reduce the computational cost of traditional LES a hybrid method is proposed in which the near-wall eddies are modelled in a Reynolds-averaged sense. Close to walls the flow is treated with the RANS-equations and this layer act as wall model for the outer flow handled by LES. The well-known high Reynolds number two-equation k - ϵ turbulence model is used in the RANS layer and the model automatically switches to a two-equation k - ϵ subgrid-scale stress model in the LES region. The approach can be used for flow over rough walls.To demonstrate the ability of the proposed hybrid method, simulations of the wind flow over a complex terrain near Wellington in New Zealand are presented. Under certain conditions unsteady flow features have been measured at the site - flow features that could lead to high structural loads on a planned wind farm. These transient flow phenomena are reproduced with the new RANS/LES method. Additionally, the results from the hybrid method are compared with pure RANS results.