An Euler solver for three-dimensional turbomachinery flows

A time-marching finite volume numerical procedure is presented for three-dimensional Euler analysis of turbomachinery flows. The proposed scheme is applied to the conservative form of the Euler equations written in general curvilinear co-ordinates. A simple but computationally efficient grid is constructed. Numerical solution results for three 3D turbine cascade flows have been presented and compared with available measurements as well as with another state-of-the-art 3D Euler analysis numerical solution in order to demonstrate the accuracy and computational efficiency of the analysis method. Also, the predicted results are compared with a 3D potential flow solver and comparison is made with the analytical solution. The proposed method is an accurate and reliable technique for solving the compressible flow equations in turbomachinery geometries.     

Overlapping grids and multigrid methods for three-dimensional unsteady flow calculations in IC engines

A new computational methodology with emphasis on using an overlapping grid technique and a multigrid method has been developed. The main feature of the present overlapping-grid system is of extended flexibility to deal with three-dimensional complex multicomponent geometries. The multigrid method is incorporated into this technique to accelerate the convergence of the numerical solution. The current scheme has been applied for computations of the laminar flows in the multicomponent configuration of internal combusion engines. The flow is governed by three-dimensional, time-dependent, incompressible Navier-Stokes equations with the continuity equation. A time-independent grid system is constructed for the moving boundary, i.e. the moving piston in the engine. This grid system is entirely different from others for the same problem in previous works. The performance of the present method has been validated by comparing the results with those from an equivalent, single-grid method and those from experiments. In addition, the flexibility and potential of the method has been demonstrated by calculating several cases which would be very difficult to be handled by other schemes.     

Simulation of two- and three-dimensional internal subsonic flows using a finite element method

In this paper a finite element method is presented to predict internal subsonic flows. Using a low-Mach-number approximation, the pressure is decomposed into a mean thermodynamic contribution and a dynamic fluctuation to deal with the complex role of the pressure in internal aerodynamics. A semi-implicit time integration and a finite element method with a moving mesh are described to take into account complex geometries and moving boundaries. An Uzawa algorithm accelerated by a preconditioned residual method is introduced to solve the coupled non-symmetric linear system for the velocity components and the pressure. An efficient conjugate gradient method combined with an incomplete LU preconditioning is used to solve the non-symmetric linear systems arising from the discretization. The implementation of the numerical scheme on parallel supercomputers is also discussed. Efficient algorithms for the finite element assembly phase and for the solution of linear systems are described which take advantage of the parallel architecture of the new generation of supercomputers. With this technique a global speed-up of 10 is achieved on a supercomputer with eight processors. To illustrate the capabilities of the numerical method, 2D and 3D simulations of flows in the combustion chamber of a reciprocating engine and around the combustor dome of a gas turbine engine are presented.     

Finite element analysis of two- and three-dimensional flows around square columns in tandem arrangement

Two- (2D) and three-dimensional (3D) finite element analyses for flow around two square columns in tandem arrangement were performed with various column spacings and Reynolds numbers. The computed values were compared with the wind-tunnel results in terms of the aerodynamic characteristics of the leeward column. In most 2D computations, strong vortices were formed behind the windward column, irrespective of widely changed Reynolds numbers. This was different from the experimental phenomena of equivalent spacing, so that the computed time-averaged pressure coefficients were not identical to the experimental values except when the distance between the two columns was adequately wide or narrow. On the other hand, in 3D computation, distinct differences in flow structures behind the column were observed between Reynolds numbers of 10³ and 10⁴ and the pressure coefficient in the 3D analysis with Re=10⁴ agreed well with the experimental value. Thus, the effectiveness of 3D computations and Reynolds number effects on the flow around two square columns have been confirmed. © 1998 John Wiley & Sons, Ltd.     

Variational formulation of three-dimensional viscous free-surface flows: Natural imposition of surface tension boundary conditions

We present a new surface-intrinsic linear form for the treatment of normal and tangential surface tension boundary conditions in C⁰-geometry variational discretizations of viscous incompressible free-surface flows in three space dimensions. The new approach is illustrated by a finite (spectral) element unsteady Navier-Stokes analysis of the stability of a falling liquid film.     

Hybrid Unsteady RANS and PDF Method for Turbulent Non-Reactive and Reactive Flows

A hybrid unsteady Reynolds-averaged numerical simulation (U-RANS) and probability density function (PDF) method is developed for turbulent non-reactive and reactive flows. The resulting modeled equations are solved by a consistent hybrid finite volume and Lagrangian Monte-Carlo particle method. Both turbulent non-reactive and reactive flows in a rectangular channel containing a triangular-shaped bluff-body are simulated. One-step and two-step mechanisms for propane/air combustion are used for the reactive case. The time-averaged results are compared with both experimental data and numerical results from the literature using large eddy simulation (LES) and steady RANS. The results of the present method are in good agreement with the experimental data, and they improve the numerical results available in the literature.     

Treatment of numerical diffusion in strong convective flows

A three-dimensional extension of the QUICK scheme adapted for the finite volume method and non-uniform grids is presented to handle convection-diffusion problems for high Peclet numbers and steep gradients. The algorithm is based on three-dimensional quadratic interpolation functions in which the transverse curvature terms are maintained and the diagonal dominance of the coefficient matrix is preserved. All formulae are explicitly given in an appendix. Results obtained with the classical upwind (UDS), the simplified QUICK (transverse terms neglected) and the present full QUICK schemes are given for two benchmark problems, one two-dimensional, steady state and the other three-dimensional, unsteady state. Both QUICK schemes are shown to give superior solutions compared with the UDS in terms of accuracy and efficiency. The full QUICK scheme performs better than the simplified QUICK, giving even for coarse grids acceptable results closer to the analytical solutions, while the computational time is not affected much.     

Edge fracture in cone-plate and parallel plate flows

Edge fracture is an instability of cone-plate and parallel plate flows of viscoelastic liquids and suspensions, characterised by the formation of a `crack' or indentation at a critical shear rate on the free surface of the liquid. A study is undertaken of the theoretical, experimental and computational aspects of edge fracture. The Tanner-Keentok theory of edge fracture in second-order liquids is re-examined and is approximately extended to cover the Criminale-Ericksen-Filbey (CEF) model. The second-order theory shows that the stress distribution on the semi-circular crack is not constant, requiring an average to be taken of the stress; this affects the proportionality constant, K in the edge fracture equation −N _2c = KΓ/a, where N _2c is the critical second normal stress difference, Γ is the surface tension coefficient and a is the fracture diameter. When the minimum stress is used, K = 2/3 as found by Tanner and Keentok (1983). Consideration is given to the sources of experimental error, including secondary flow and slip (wall effect). The effect of inertia on edge fracture is derived. A video camera was used to record the inception and development of edge fracture in four viscoelastic liquids and two suspensions. The recorded image was then measured to obtain the fracture diameter. The edge fracture phenomenon was examined to find its dependence on the physical dimensions of the flow (i.e. parallel plate gap or cone angle), on the surface tension coefficient, on the critical shear rate and on the critical second normal stress difference. The critical second normal stress difference was found to depend on the surface tension coefficient and the fracture diameter, as shown by the theory of Tanner and Keentok (1983); however, the experimental data were best fitted by the equation −N _2c = 1.095Γ/a. It was found that edge fracture in viscoelastic liquids depends on the Reynolds number, which is in good agreement with the inertial theory of edge fracture. Edge fracture in lubricating grease and toothpaste is broadly consistent with the CEF model of edge fracture. A finite volume method program was used to simulate the flow of a viscoelastic liquid, obeying the modified Phan-Thien-Tanner model, to obtain the velocity and stress distribution in parallel plate flow in three dimensions. Stress concentrations of the second normal stress difference (N ₂) were found in the plane of the crack; the velocity distribution shows a secondary flow tending to aid crack formation if N ₂ is negative, and a secondary flow tending to suppress crack formation if N ₂ is positive. Received: 4 January 1999 Accepted: 19 May 1999     

Simulation of bluff body stabilized flows with hybrid RANS and PDF method

The motivation of this study is to investigate the turbulence–chemistry interactions by using probability density function (PDF) method. A consistent hybrid Reynolds Averaged Navier–Stokes (RANS)/PDF method is used to simulate the turbulent non-reacting and reacting flows. The joint fluctuating velocity–frequency–composition PDF equation coupled with the Reynolds averaged density, momentum and energy equations are solved on unstructured meshes by the Lagrangian Monte Carlo (MC) method combined with the finite volume (FV) method. The simulation of the axisymmetric bluff body stabilized non-reacting flow fields is presented in this paper. The calculated length of the recirculation zone is in good agreement with the experimental data. Moreover, the significant change of the flow pattern with the increase of the jet-to-coflow momentum flux ratio is well predicted. In addition, comparisons are made between the joint PDF model and two different Reynolds stress models. The project supported by the National Natural Science Foundation of China (50506028), and Action Scheme for Invigorating Education Towards the twenty-first century.     

Finite volume methods for laminar and turbulent flows using a penalty function approach

A penalty function, finite volume method is described for two-dimensional laminar and turbulent flows. Turbulence is modelled using the k-? model. The governing equations are discretized and the resulting algebraic equations are solved using both sequential and coupled methods. The performance of these methods is gauged with reference to a tuned SIMPLE-C algorithm. Flows considered are a square cavity with a sliding top, a plane channel flow, a plane jet impingement and a plane channel with a sudden expansion. A sequential method is employed, which uses a variety of dicretization practices, but is found to be extremely slow to converge; a coupled method, evaluated using a variety of matrix solvers, converges rapidly but, relative to the sequential approach, requires larger memory.     

Application of the finite volume method in the simulation of saturated flows of binary Mixtures with particular emphasis on Non-Darcy motions through porous media

Non-Darcy flows in saturated porous media with significative boundary and inertia effects are modelled applying the Continuum Theory of Mixtures approach and simulated by discretization of the governing equations by the finite volume method.

---

Sommario Flussi di tipo ‘Non-Darcy’ in mezzi porosi saturi, con significativi effetti di bordo ed inerziali, vengono modellati applicando l'approccio della Teoria delle Miscele per il Continuo e simulati mediante discretizzazione delle equazioni governanti con il metodo del volume finito.

     

Asymptotics, structure, and integration of sound-proof atmospheric flow equations

Relative to the full compressible flow equations, sound-proof models filter acoustic waves while maintaining advection and internal waves. Two well-known sound-proof models, an anelastic model by Bannon and Durran’s pseudo-incompressible model, are shown here to be structurally very close to the full compressible flow equations. Essentially, the anelastic model is obtained by suppressing ? _t ρ in the mass continuity equation and slightly modifying the gravity term, whereas the pseudo-incompressible model results from dropping ? _t p from the pressure equation. For length scales small compared to the density and pressure scale heights, the anelastic model reduces to the Boussinesq approximation, while the pseudo-incompressible model approaches the zero Mach number, variable density flow equations. Thus, for small scales, both models are asymptotically consistent with the full compressible flow equations, yet the pseudo-incompressible model is more general in that it remains valid in the presence of large density variations. For the relatively small density variations found in typical atmosphere–ocean flows, both models are found to yield very similar results, with deviations between models much smaller than deviations obtained when using different numerical schemes for the same model. This in agreement with Smolarkiewicz and Dörnbrack (Int J Numer Meth Fluids 56:1513–1519, 2007). Despite these useful properties, neither model can be derived by a low-Mach number asymptotic expansion for length scales comparable to the pressure scale height, i.e., for the regime they were originally designed for. Derivations of these models via scale analysis ignore an asymptotic time scale separation between advection and internal waves. In fact, only the classical Ogura and Phillips model, which assumes weak stratification of the order of the Mach number squared, can be obtained as a leading-order model from systematic low Mach number asymptotic analysis. Issues of formal asymptotics notwithstanding, the close structural similarity of the anelastic and pseudo-incompressible models to the full compressible flow equations makes them useful limit systems in building computational models for atmospheric flows. In the second part of the paper, we propose a second-order finite-volume projection method for the anelastic and pseudo-incompressible models that observes these structural similarities. The method is applied to test problems involving free convection in a neutral atmosphere, the breaking of orographic waves at high altitudes, and the descent of a cold air bubble in the small-scale limit. The scheme is meant to serve as a starting point for the development of a robust compressible atmospheric flow solver in future work.     

Migration and deformation of leukocytes in pressure driven flows

Direct numerical simulations (DNS) are used to study the motion and deformation of leukocytes in pressure driven flows in parallel plate channels. The influence of the adhesion force between the leukocytes and the channel wall on such motion and deformation is also investigated. Leukocytes are represented by two composite fluid models, consisting of a membrane, a cytoplasm and a nucleus. The adhesion force is computed using two adhesion force models. In the first model, the adhesion force is given by a potential, and in the second one it is given by Dembo’s kinetic adhesion model. The numerical code is based on the finite element method and the level set technique is used to track the cell membrane position. In the absence of the adhesion force, the leukocyte moves away from the wall to an equilibrium location that depends on the ratio of the cell to plasma viscosities. In presence of the adhesion force, the leukocyte is attracted to the layer of endothelial cells and, as it gets closer, it flattens under the action of hydrodynamic forces. This deformation, in turn, further increases the adhesion force. The leukocyte, however, can be captured only when it is placed sufficiently close to the wall, which for the kinetic model is of the order of 30 nm. We also find that for the normal parameter values and flow rates the adhesive force given by the kinetic model is too small to capture the leukocyte.     

Anti-plane stress intensity, energy release and energy density at crack tips in a functionally graded strip with linearly varying properties

The fracture problem of a crack in a functionally graded strip with its properties varying in a linear form along the strip thickness under an anti-plane load is considered. The embedded anti-plane crack is located in the middle of strip half way through the thickness. The third mode stress intensity factor is derived using two different methods. In the first method, by employing Fourier integral transforms, the governing equation is converted to a singular integral equation, which is subsequently solved numerically by the collocation method based on Chebyshev polynomials. Then, the problem is solved by means of finite element method in which quadrilateral 8-node singular elements around each crack tip are used. After inspecting the validity of the solution technique, effects of crack geometry and non-homogeneous material parameter on the stress intensity, energy release and energy density are studied and the results of analytical and FEM solutions are compared.     

Application of adaptive mesh refinement method to complex flows in clean rooms

The adaptive mesh refinement (AMR) method is developed for three-dimensional turbulent complex flows in clean rooms using the finite volume method with a collocated grid arrangement. Clean rooms have many interesting and complex flow characteristics especially the secondary flows and the recirculation regions. The accurate numerical solution of the flows is important for the efficient design of clean rooms. The use of the conventional uniform grid requires such a high computational time and data storage capacity that they make computational fluid dynamics (CFD) less attractive for the design optimization. The AMR method is, therefore, applied by using the fine grid only in the required regions and using the coarse grid in the other regions. The velocity is chosen as the main parameter for the grid refinement because it is the most influential parameter in clean rooms. The results show that the present AMR method can reduce the computational time by eight times and the data storage requirement is only 37% of that using the conventional method, while the same order of accuracy can be maintained. The present AMR method is, therefore, proved to be a promising technique for solving three-dimensional turbulent complex flows in clean rooms.     

Metal matrix composite solidification in the presence of cooled fibers: numerical simulation and experimental observation

Attempts have been made to alter the solidification microstructures of fiber reinforced aluminum composites by cooling the ends of the fibers extending out of the mold. Experimental observations indicate that cooling the extended ends of the reinforcement results in finer microstructures in the matrix and changes the nature of the interface. In this paper, numerical simulation is performed on a two-dimensional axi-symmetric model to investigate the solidification process of metal matrix composite (MMC) with the extended ends of the fibers cooled by a heat sink. The numerical simulation is based on the source-based enthalpy method with finite volume discretization. The temperature profiles obtained by simulation are compared to the cooling curves measured experimentally in order to validate the current mathematical model. It is found that the simulation result matches the experimental data with reasonable agreement.     

Strain energy approach to compute stress intensity factors for isotropic homogeneous and bi-material V-notches

A strain energy approach (SEA) is developed to compute the general stress intensity factors (SIFs) for isotropic homogeneous and bi-material plates containing cracks and notches subject to mode I, II and III loading conditions. The approach is based on the strain energy of a control volume around the notch tip, which may be computed by using commercial finite element packages. The formulae are simple and easy to implement. Various numerical examples are presented and compared to corresponding published results or results that are computed using different numerical methods to demonstrate the accuracy of the SEA. Many of those results are new, especially for the cases of bi-material notches where the problem is quite complicated.