Engineering Turbulence Modelling and Experiments - 4 (eBook)
972 Seiten
Elsevier Science (Verlag)
978-0-08-053098-7 (ISBN)
The purpose of this series of symposia is to provide a forum for presenting and discussing new developments in the area of turbulence modelling and measurements, with particular emphasis on engineering-related problems.
Turbulence is still one of the key issues in tackling engineering flow problems. As powerful computers and accurate numerical methods are now available for solving the flow equations, and since engineering applications nearly always involve turbulence effects, the reliability of CFD analysis depends more and more on the performance of the turbulence models. Successful simulation of turbulence requires the understanding of the complex physical phenomena involved and suitable models for describing the turbulent momentum, heat and mass transfer. For the understanding of turbulence phenomena, experiments are indispensable, but they are equally important for providing data for the development and testing of turbulence models and hence for CFD software validation.
These proceedings contain the papers presented at the 4th International Symposium on Engineering Turbulence Modelling and Measurements held at Ajaccio, Corsica, France from 24-26 May 1999. It follows three previous conferences on the topic of engineering turbulence modelling and measurements. The purpose of this series of symposia is to provide a forum for presenting and discussing new developments in the area of turbulence modelling and measurements, with particular emphasis on engineering-related problems. Turbulence is still one of the key issues in tackling engineering flow problems. As powerful computers and accurate numerical methods are now available for solving the flow equations, and since engineering applications nearly always involve turbulence effects, the reliability of CFD analysis depends more and more on the performance of the turbulence models. Successful simulation of turbulence requires the understanding of the complex physical phenomena involved and suitable models for describing the turbulent momentum, heat and mass transfer. For the understanding of turbulence phenomena, experiments are indispensable, but they are equally important for providing data for the development and testing of turbulence models and hence for CFD software validation.
Front Cover 1
Engineering Turbulence Modelling and Experiments 4 4
Copyright Page 5
Contents 8
Preface 18
Part 1. Invited Lectures 20
Chapter 1. Strategies for turbulence modelling and simulations 22
Chapter 2. Modelling shock-affected near-wall flows with anisotropy-resolving turbulence closures 38
Chapter 3. Thermal hydraulics simulations: what turbulence model strategies? 56
Chapter 4. Modelling of production, kinematic restoration and dissipation of flame surface area in turbulent combustion 68
Chapter 5. Visualization and measurement of spatial structures in turbulent flow 82
Part 2. Turbulence Modelling 90
Chapter 6. Three-dimensional modelling of turbulent free-surface jets 92
Chapter 7. A-priori tests of Reynolds stress transport models in turbulent pipe expansion flow 102
Chapter 8. Thermodynamically consistent second order turbulence modelling based on extended thermodynamics 112
Chapter 9. A nonlinear stress-strain model for wall-bounded turbulent flows 122
Chapter 10. Accuracy and robustness of non-linear eddy viscosity models 132
Chapter 11. A proposal for taking into account the intermittency phenomenon in the calculation of wall-bounded turbulent flows 144
Chapter 12. Modelling the turbulent flow subjected to magnetic field 154
Chapter 13. Modeling separation and reattachment using the turbulent potential model 164
Chapter 14. A turbulence model for the pressure-strain correlation term accounting for the effect of compressibility 174
Chapter 15. Consistent modelling of fluctuating temperature- gradient-velocity-gradient correlations for natural convection 184
Chapter 16. On the modelling of the transport equation for the passive scalar dissipation rate 194
Part 3. Direct and Large-Eddy Simulations 204
Chapter 17. The direct influence of mean flow on subgrid stresses in LES of turbulent flows 206
Chapter 18. Large-eddy simulation of non-equilibrium inflow conditions and of the spatial development of a confined plane jet with co-flowing streams 216
Chapter 19. Influence of spatio-temporal inflow organization on LES of a spatially developing plane mixing layer 226
Chapter 20. On the influence of turbulence characteristics at an inlet boundary for large-eddy simulation of a turbulent boundary layer 236
Chapter 21. A dynamic one-equation subgrid model for simulation of flow around a square cylinder 246
Chapter 22. Numerical simulation and modelling of a wall-bounded compressed turbulence 256
Chapter 23. Influence of curvature and torsion on turbulent flow in helically coiled pipes 266
Chapter 24. Large eddy simulations of stirred tank flow 276
Part 4. Applications of Turbulence Models 286
Chapter 25. Higher order turbulence modelling for industrial applications 288
Chapter 26. Flow and turbulence modelling in a motored reciprocating engine using a cubic non-linear k-e turbulence model 298
Chapter 27. A spectral closure for inhomogeneous turbulence applied to the computation of an engine related flow 308
Chapter 28. Prediction of turbulent oscillatory flows in complex systems 318
Chapter 29. Numerical computation of turbulent flow around radome structures 328
Chapter 30. Computations of turbulent flows using the V2F model in a finite element code 338
Chapter 31. Intake/S-bend diffuser flow prediction using linear and non-linear eddy-viscosity and second-moment closure turbulence models 348
Chapter 32. Modelling of the flow in rotating annular flumes 358
Chapter 33. Second-moment closure predictions of turbulence- induced secondary flow in a straight square duct 368
Part 5. Experimental Techniques 378
Chapter 34. Experimental investigation of coherent structures using digital particle image velocimetry 380
Chapter 35. Measurements on the mixing of a passive scalar in a turbulent pipe flow using DPIV and LIF 390
Chapter 36. Wavelet patterns in the near wake of a circular cylinder and a porous mesh strip 400
Part 6. Experimental Studies 410
Chapter 37. An investigation of the engulfment mechanism in a turbulent wake 412
Chapter 38. Investigation on boundary effects in jet flows" 422
Chapter 39. Influence of shallowness on growth and structures of a mixing layer 434
Chapter 40. Influence of stationary and rotating cylinders on a turbulent plane jet 442
Chapter 41. Turbulence measurements of an inclined rectangular jet in a boundary layer 452
Chapter 42. Pressure velocity coupling in a subsonic round jet 462
Chapter 43. Scalar mixing in variable density turbulent jets 472
Chapter 44. Decay of round turbulent jets with swirl 480
Chapter 45. Correlating structure of tip vortices and swirl flows induced by a low aspect ratio rotor blade 490
Chapter 46. Effects of adverse pressure gradient on quasi-coherent structures in turbulent boundary layer 500
Chapter 47. The semi-deterministic approach as way to study coherent structures: Case of a turbulent flow behind a backward-facing step 510
Chapter 48. Experimental investigation of the coolant flow in a simplified reciprocating engine cylinder head 520
Chapter 49. Secondary flow in compound sinuous/meandering channels 530
Chapter 50. Diffused turbulence distortion by a free-surface 540
Part 7. Transition 550
Chapter 51. Surface curvature and pressure gradient effects on boundary layer transition 552
Chapter 52. Calculating turbulent and transitional boundary- layers with two-layer models of turbulence 562
Chapter 53. Turbulence modeling and computation of viscous transitional flows for low pressure turbines 574
Chapter 54. A general model for transition in wall-bounded compressible flows 586
Chapter 55. The use of a turbulence weighting factor to model by-pass transition 596
Chapter 56. Modelling of separation-induced transition to turbulence with a second-moment closure 606
Chapter 57. The effect of a single roughness element on a flat plate boundary layer transition 616
Chapter 58. Features of laminar-turbulent transition in a free convection boundary layer near a vertical heated surface 626
Part 8. Turbulence Control 636
Chapter 59. On active control of high-lift flow 638
Chapter 60. A demonstration of MEMS- based active turbulence transitioning 646
Chapter 61. Evolution of instabilities in an axisymmetric impinging jet 656
Part 9. Aerodynamic Flows 666
Chapter 62. Assessment of eddy viscosity models in 2D and 3D shock/boundary-layer interactions 668
Chapter 63. Assessment of explicit algebraic stress models in transonic flows 678
Chapter 64. Detached-eddy simulation of an airfoil at high angle of attack 688
Chapter 65. Transition and turbulence modelling for dynamic stall and buffet 698
Chapter 66. Scrutinizing flow field pattern around thick cambered trailing edges: experiments and computation 708
Chapter 67. Turbulent structure in the three-dimensional boundary layer on a swept wing 718
Chapter 68. Flowfield characteristics of swept struts in supersonic annular flow 728
Part 10. Turbomachinery Flows 738
Chapter 69. Flow in a radial outflow impeller rear cavity of aeroengines 740
Chapter 70. Experimental investigation of turbulent wake-blade interaction in axial compressors 750
Chapter 71. An experimental study of the unsteady characteristics of the turbulent wake of a turbine blade 760
Chapter 72. Experimental guidelines for retaining energy-efficient axial flow rotor cascade operation under off-design circumstances 770
Part 11. Heat Transfer 780
Chapter 73. On prediction of turbulent convective heat transfer in rib-roughened rectangular cooling ducts 782
Chapter 74. The measurement of local wall heat transfer in stationary U-ducts of strong curvature, with smooth and rib roughened walls 792
Chapter 75. Studies of turbulent jets impinging on moving surfaces 802
Part 12. Combustion Systems 812
Chapter 76. Turbulence modelling in joint PDF calculations of piloted-jet flames 814
Chapter 77. Towards a general correlation of turbulent premixed flame wrinkling 824
Chapter 78. Mixing in isotropic turbulence with scalar injection 834
Chapter 79. Modelling turbulent diffusion flames with full second-moment closures using cubic, realizable models 840
Chapter 80. Advanced modeling of turbulent non-equilibrium swirling natural gas flames 850
Chapter 81. Effects of turbulence length scale on flame speed: a modelling study 860
Chapter 82. Application of a Lagrangian PDF method to turbulent gas/particle combustion 870
Chapter 83. Large eddy simulation of a nonpremixed turbulent swirling flame 880
Chapter 84. Large eddy simulation of a bluff body stabilised flame 890
Chapter 85. Investigation of the effect of turbulent flow behaviour and mixing conditions on the combustion process in the homogeneous burnout zone of a small scale wood heater by numerical simulations and measurements 900
Part 13. Two-Phase Flows 910
Chapter 86. Proposal of a Reynolds stress model for gas-particle turbulent flows and its application to cyclone separators 912
Chapter 87. A CRW model for free shear flows 922
Chapter 88. Test of an Eulerian Lagrangian simulation of wall heat transfer in a gas-solid pipe flow 932
Chapter 89. Analysis and discussion on the Eulerian dispersed particle equations in non-uniform turbulent gas-solids two-phase flows 942
Chapter 90. Experimental research of modification of grid- turbulence by rough particles 952
Chapter 100. Interaction between leading and trailing elongated bubbles in a vertical pipe flow 962
AUTHOR INDEX 972
Strategies for turbulence modelling and simulations
P.R. Spalarta a Boeing Commercial Airplanes, P.O. Box 3707, Seattle, WA 98124, USA
Abstract
This is an attempt to clarify the many levels possible for the numerical prediction of a turbulent flow, the target being a complete airplane, gas turbine, or car. These levels still range from a solution of the steady Reynolds-Averaged Navier-Stokes equations to a Direct Numerical Simulation, with Large-Eddy Simulation in-between. However recent years have added intermediate concepts, dubbed “VLES”, “URANS” and “DES”. They are in experimental use and, although more expensive, threaten complex traditional models especially for bluff-body flows, where three-dimensional simulations in two-dimensional geometries are flourishing. Turbulence predictions face two principal challenges: (I) predicting growth and separation of the boundary layer, and (II) providing accurate Reynolds stresses after separation. (I) is simpler, but makes much higher accuracy demands, and appears to give models of higher complexity almost no advantage. (II) is the arena for complex RANS models and the newer strategies. With some strategies, grid refinement is aimed at numerical accuracy; in others it is aimed at richer turbulence physics. In some approaches the empirical constants play a strong role even when the grid is very fine; in others, their role vanishes. For several decades, practical methods will necessarily be hybrid, and their empirical content will remain substantial. The law of the wall will be particularly resistant. Estimates are offered of the grid resolution needed for the application of each strategy to full-blown aerodynamic calculations, feeding into rough estimates of the feasibility date, based on computing-power growth.
1 INTRODUCTION
The turbulence problem is of course far from solved, whether in terms of mathematical and intuitive understanding, or in terms of obtaining engineering accuracy in machines that depend on viscous fluid dynamics. Technological fields of global importance such as the airliner and automobile industries revolve around such devices. This economic stake motivates relentless, imaginative, and expensive efforts at turbulence prediction by any plausible approach. This should not defeat common sense, as argued elsewhere [1], and we must have visibility of when a method may progress from experimental to established and useful in engineering (research uses of simulation are another matter). Chapman made such predictions in 1979 [2], which still carry weight although his view of turbulence prediction in the 1990’s is now recognized as optimistic.
This paper focuses on the numerical prediction of turbulent flow regions. The equally difficult problem of transition prediction is mentioned only in passing. Physical testing methods in the transportation industry are beset by their own severe transition- and turbulence-related difficulties. Tests with scale models usually imply both lower Reynolds numbers and higher freestream turbulence levels (in addition to blockage, bracket and mounting issues, and aero-elastic differences). The resulting scale effects can be misleading, with unforeseen reversals of the normal trend (by which higher Reynolds numbers bring better performance), especially as competing companies seek optimal aerodynamic designs; such designs have narrow margins and magnify the sensitivity to viscous effects. Thus, industry demands accuracy from Computational Fluid Dynamics (CFD), but not perfection.
Also note that, whether in the airframe, turbine engine, or automotive industry, turbulence is not the only obstacle in CFD. Major numerical challenges remain between the state of the art and the routine calculation of flows over even moderately complex 3D geometries. These challenges relate not only to computing cost, but also to solution quality, particularly in terms of gridding. Presenting turbulence as the only “pacing item” in CFD might benefit research funding, but it is not accurate. On the other hand, many more capable people are engaged in grid generation, solvers, and pre- and post-processors, than in turbulence. Our effort may be unbalanced, although more duplication occurs on the programming side (it is easier to show progress in programming, let alone in code exercising, than in modelling). Sharing large codes is more difficult than sharing turbulence models, for which the equations (normally!) fit on one page. With a few exceptions, models have been freely published.
The numerical strengths of CFD increase by the year thanks to the progress of computers and algorithms, whereas turbulence modelling can stagnate. If that is the case long enough, modelling will become the pacing item in important types of CFD in a matter of a decade or less, at least in the Reynolds Averaged Navier-Stokes (RANS) mode. It is then very sensible to examine approaches that trade “intelligence” (in the sense of powerful turbulence theories) for computing effort (unfairly described as “brute force”). It is my purpose to provide a viewpoint on such methods, which I predict will proliferate, and make a major contribution. The most stimulating issue may be the share between empiricism and numerical force in the eventually successful methods (§2). The concrete cost issues are addressed through a table attached to §3.
2 PHYSICAL ASPECTS
2.1 RANS models
The field of classical RANS turbulence modelling is active. At a recent biennial international symposium, about twenty-five papers presented new models or new versions of models [3]. These were offered for outside use, with varying degrees of sincerity and completeness in the description. No student of turbulence has the time to give each of these serious consideration. The full range of RANS methods is receiving work; this unfortunately testifies that no class of models has emerged as clearly superior, or clearly hopeless. Activity is not even restricted to differential methods; isolated groups are refining integral boundary-layer solvers, to allow more three-dimensionality and more separation. The same seems to apply to algebraic models. Eddy-viscosity transport models, being the simplest models that can be applied with a general grid structure, are now used extensively. The step back from two equations to a single equation has clearly not crippled the approach [4-6], while tangibly reducing the true cost of solutions. Conversely, models with up to four equations are in contention [7,8]. Perot & Moin’s is especially intriguing.
Having referred in the abstract to “Challenges (I) and (II)”, I could add the following “Challenge Zero”. Complete configurations often have laminar regions in their boundary layers; it is very helpful if a turbulence model can be “dormant” in such regions, meaning that its transport equations accept solutions with vanishing Reynolds stresses. Similarly, regions of irrotational and non-turbulent fluid, which are large in external aerodynamics, do not physically influence the turbulent regions such as boundary layers (weak freestream activity does have much influence on natural transition, but we leave transition prediction to a separate method). Again it is very helpful if the model accepts zero values in such regions, or small values without influence on the turbulent layers. At the same time, the model should allow the contamination of a laminar shear layer by contact with a strongly turbulent layer (contamination by moderate freestream turbulence is more subtle, and is within reach of only a few models). This all depends on the behaviour of the model at the turbulent/non-turbulent interface. In some models the stress level in the turbulent layer depends demonstrably on either the freestream values of the turbulence variables or, even worse, on the grid spacing at the interface. Few people have devoted attention to this question [9,10,5], and model descriptions sometimes make no mention of recommended freestream values (and also fail to demonstrate insensitivity). However, it happens that the k-∈, SST and S-A models, which all three pass the freestream-sensitivity test, can fairly be described as “popular”. Possibly, their tendency to give the same answer in different codes is valued by the users. In the perennial question of the choice of a second variable in two-equation models, freestream sensitivity should be given a high priority. It is much more important than the value of some high derivative at a solid wall.
The model activation or “transition” from laminar to turbulent boundary layer is in fact more troublesome than the interface with irrotational fluid. Even though the S-A model was designed and tested for it, users still encounter premature transition. This occurs even with first-order upstream differencing, which at first sight should guarantee that unwanted nonzero values of eddy viscosity are “washed out” of the region upstream of the numerical trip. A factor in this is the steep variation of the eddy viscosity at transition [11]: a grid may be fine enough in both the laminar and turbulent regions but much too coarse at transition, creating oscillations which then propagate upstream. This problem can be solved with grid adaptation, but it remains an embarrassment to the model designers, especially since the transition process is not modeled accurately to start with [11]. The failure...
| Erscheint lt. Verlag | 14.4.1999 |
|---|---|
| Sprache | englisch |
| Themenwelt | Informatik ► Software Entwicklung ► User Interfaces (HCI) |
| Naturwissenschaften ► Physik / Astronomie ► Angewandte Physik | |
| Naturwissenschaften ► Physik / Astronomie ► Strömungsmechanik | |
| Technik ► Bauwesen | |
| Technik ► Maschinenbau | |
| ISBN-10 | 0-08-053098-2 / 0080530982 |
| ISBN-13 | 978-0-08-053098-7 / 9780080530987 |
| Haben Sie eine Frage zum Produkt? |
PDF (Adobe DRM)Kopierschutz: Adobe-DRM
Adobe-DRM ist ein Kopierschutz, der das eBook vor Mißbrauch schützen soll. Dabei wird das eBook bereits beim Download auf Ihre persönliche Adobe-ID autorisiert. Lesen können Sie das eBook dann nur auf den Geräten, welche ebenfalls auf Ihre Adobe-ID registriert sind.
Details zum Adobe-DRM
Dateiformat: PDF (Portable Document Format)
Mit einem festen Seitenlayout eignet sich die PDF besonders für Fachbücher mit Spalten, Tabellen und Abbildungen. Eine PDF kann auf fast allen Geräten angezeigt werden, ist aber für kleine Displays (Smartphone, eReader) nur eingeschränkt geeignet.
Systemvoraussetzungen:
PC/Mac: Mit einem PC oder Mac können Sie dieses eBook lesen. Sie benötigen eine
eReader: Dieses eBook kann mit (fast) allen eBook-Readern gelesen werden. Mit dem amazon-Kindle ist es aber nicht kompatibel.
Smartphone/Tablet: Egal ob Apple oder Android, dieses eBook können Sie lesen. Sie benötigen eine
Geräteliste und zusätzliche Hinweise
Buying eBooks from abroad
For tax law reasons we can sell eBooks just within Germany and Switzerland. Regrettably we cannot fulfill eBook-orders from other countries.
aus dem Bereich
