No description available.

The average and fluctuating components in a model ring-type gas turbine combustor are characterized using a Large Eddy Simulation at a Reynolds number of 11,000, based on the bulk velocity and the mean channel height. A spatial filter is applied to the incompressible Navier-Stokes equations, and a high pass filtered Smagorinsky model is used to model the sub-grid scales. Two cases are studied: one with only the swirler inlet active, and one with a single row of dilution jets activated, operating at a momentum flux ratio J of 100. The goal of both of these studies is to validate the capabilities of the solver NEK5000 to resolve important flow features inherent to gas turbine combustors by comparing qualitatively to the work of Jakirlic. Both cases show strong evidence of the Precessing Vortex Core, an essential flow feature in gas turbine combustors. Each case captures other important flow characteristics, such as corner eddies, and in general predicts bulk flow movements well. However, the simulations performed quite poorly in terms of predicting turbulence shear stress quantities. Difficulties in properly emulating the turbulent velocity entering the combustor for the swirl, as well as mesh quality concerns, may have skewed the results. Overall, though small length scale quantities were not accurately captured, the large scale quantities were, and this stress test on the HPF LES model will be built upon in future work that looks at more complex combustors.

In the context of Large Eddy Simulation (LES) of turbulent flows, there is a current need to compare and evaluate different proposed subfilter-scale models. In order to carefully compare subfilter-scale models and compare LES predictions to Direct Numerical Simulation (DNS) results (the latter would be helpful in the comparison and validation of models), there is a real need for a "grid-independent" LES capability and explicit filtering methods offer one means by which this may be achieved. Advantages of explicit filtering are that it provides a means for eliminating aliasing errors, allows for the direct control of commutation errors, and most importantly allows a decoupling between the mesh spacing and the filter width which is the primary reason why there are difficulties in comparing LES solutions obtained on different grids. This thesis considers the design and assessment of discrete explicit filters and their application to isotropic turbulence prediction.

In the context of Large Eddy Simulation (LES) of turbulent flows, there is a current need to compare and evaluate different proposed subfilter-scale models. In order to carefully compare subfilter-scale models and compare LES predictions to Direct Numerical Simulation (DNS) results (the latter would be helpful in the comparison and validation of models), there is a real need for a "grid-independent" LES capability and explicit filtering methods offer one means by which this may be achieved. Advantages of explicit filtering are that it provides a means for eliminating aliasing errors, allows for the direct control of commutation errors, and most importantly allows a decoupling between the mesh spacing and the filter width which is the primary reason why there are difficulties in comparing LES solutions obtained on different grids. This thesis considers the design and assessment of discrete explicit filters and their application to isotropic turbulence prediction.

Large Eddy Simulation is a method of obtaining high accuracy computational results for modelling fluid flow. Unfortunately it is computationally expensive limiting it to users of large parallel machines. However, it may be that the use of LES leads to an over-resolution of the problem because the bulk of the computational domain could be adequately modelled using the Reynolds averaged approach. A study has been undertaken to assess the feasibility, both in accuracy and computational efficiency of using a parallel computer to solve both LES and RANS type turbulence models on the same domain for the problem flow over a circular cylinder at Reynolds number 3 900 To do this the domain has been created and then divided into two sub-domains, one for the LES model and one for the kappa - epsilon turbulence model. The hybrid model has been developed specifically for a parallel computing environment and the user is able to allocate modelling techniques to processors in a way which enables expansion of the model to any number of processors. Computational experimentation has shown that the combination of the Smagorinsky model can be used to capture the vortex shedding from the cylinder and the information successfully passed to the kappa - epsilon model for the dissipation of the vortices further downstream. The results have been compared to high accuracy LES results and with both kappa - epsilon and Smagorinsky LES computations on the same domain. The hybrid models developed compare well with the Smagorinsky model capturing the vortex shedding with the correct periodicity. Suggestions for future work have been made to develop this idea further, and to investigate the possibility of using the technology for the modelling of mixing and fast chemical reactions based on the more accurate prediction of the turbulence levels in the LES sub-domain.

A novel computational approach, Two Level Simulation (TLS), was developed based on the explicit reconstruction of the small-scale velocity by solving the small-scale governing equations on the domain with reduced dimension representing a collection of one-dimensional lines embedded in the three-dimensional flow domain. A coupled system of equations, that is not based on an eddy-viscosity hypothesis, was derived based on the decomposition of flow variables into the large-scale and the small-scale components without introducing the concept of filtering. Simplified treatment of the small-scale equations was proposed based on modeling of the small-scale advective derivatives and the small-scale dissipative terms in the directions orthogonal to the lines. TLS approach was tested to simulate benchmark cases of turbulent flows, including forced isotropic turbulence, mixing layers and well-developed channel flow, and demonstrated good capabilities to capture turbulent flow features using relatively coarse grids.

No description available.

<p>The aim of the present study is to develop subgrid-scale models that are relevant for complex flows and combustion. A stochastic model based on a stochastic Smagorinsky constant with adjustable variance and time scale is proposed. The stochastic model is shown to provide for backscatter of both kinetic energy and scalar variance without causing numerical instabilities. A new subgrid-scale scalar flux model is developed using the same kind of methodology that leads to the explicit algebraic scalar flux model, EASFM, for RANS. The new model predicts the anisotropy of the subgrid-scales in a more realistic way than the eddy diffusion model. Both new models were tested in rotating homogeneous shear flow with a passive scalar. Rogallo’s method of moving the frame with the mean flow to enable periodic boundary conditions was used to simulate homogeneous shear flow.</p>

Optimal large eddy simulation (OLES) is an approach to LES sub-grid modeling that requires multi-point correlation data as input. Until now, this has been obtained by analyzing DNS statistics. In the finite-volume OLES formulation studied here, under the assumption of small-scale homogeneity and isotropy, these correlations can be theoretically determined from Kolmogorov inertial-range theory, small-scale isotropy, along with the quasi-normal approximation. These models are expressed as generalized quadratic and linear finite volume operators that represent the convective momentum flux. These finite volume operators have been analyzed to determine their characteristics as numerical approximation operators and as models of small-scale effects. In addition, the dependence of the model operators on the anisotropy of the grid and on the size of the stencils is analyzed to develop idealized general operators that can be used on general grids. The finite volume turbulence operators developed here will be applicable in a wide range of LES problems. / text

Accurate models for the mixing of fuel and oxidizer at small, unresolved flow length scales are critical to the predictive skill of large eddy simulation (LES) of turbulent combustion. Subfilter scalar variance and subfilter scalar dissipation rate are important parameters in combustion modeling approaches based on a conserved scalar, but are prone to numerical and modeling errors due to the nature of practical LES computations. This work examines the errors incurred in these models using a novel method that couples LES scalar modeling with direct numerical simulation (DNS) of homogeneous isotropic turbulence and offers modeling and numerical techniques to address these errors. In the coupled DNS-LES method, DNS velocity fields are evolved simultaneously with LES scalar fields. The filtered DNS velocities are supplied to the LES scalar equations, instead of solving the LES momentum equations. This removes the effect of errors in the filtered scalar evolution from the scalar modeling analysis. Results obtained using the coupled DNS-LES approach, which permits detailed study of physics-related and numerical errors in scalar modeling, show that widely used algebraic dynamic models for subfilter scalar variance lack accuracy due to faulty equilibrium modeling assumptions and sensitivity to numerical error. Transport equation models for variance show superior performance, provided that the scalar dissipation rate model coefficient is set appropriately. For this purpose, a new dynamic approach for nonequilibrium modeling of subfilter scalar dissipation rate is developed and validated through a priori tests in an inhomogeneous jet flow and using the coupled DNS-LES method for assessment of numerical error effects. Explicit filtering is assessed as means to control numerical error in LES scalar modeling and the scalar equations are reformulated to account for the explicit filtering technique. Numerical convergence of the mean subfilter scalar variance prediction with increasing grid resolution is demonstrated. / text