3/2003 Computational Fluid Dynamics

Contents:

• J.Badur, K.Kosowski, R.Stepien and M.Piwowarski, Research into Flows in Turbine Blade Seals Part I: Research Methods - abstract | full text
• R.Stepien, K.Kosowski and J.Badur, Research into Flows in Turbine Blade Seals Part II: Numerical Analysis - abstract | full text
• K.Kosowski, R.Stepien, M.Piwowarski and J.Badur, Research into Flows in Turbine Blade Seals Part III: Numerical Calculations versus Experiment - abstract | full text
• J.Badur, M.Karcz and T.Wysocki, Chemistry-Turbulence Coupling in a Model of Inhomogenously Premixed Combustion - abstract
• J.Baranski, J.Stasiek and W.Blasiak, Physical and Numerical Modeling of Heat-flow Processes in Tangentially Pulverized Fuel-Fired Boiler - abstract | full text
• P.Cyklis and R.Kantor, CFD Simulation as a Tool for the Identification of the Manifold Element Reaction to Pressure Pulsations Excitation (One- and Two-phase Flow) - abstract | full text
• M.Karcz and J.Badur, A Turbulent Heat Flux Two-Equation θ'²-ε_θClosure Based on the V2F Turbulence Model - abstract
• T.Michalek and T.A.Kowalewski, Simulations of the Water Freezing Process: Numerical Benchmarks - abstract | full text
• W.Podgorska and J.Baldyga, Drop Break-up and Coalescence in a Stirred Tank - abstract | full text
• A.Rozen and J.Baldyga, Influence of Viscosity Difference on the Instability of the Core-Annular Flow - abstract | full text
• J.Smolka, A.J.Nowak and L.C.Wrobel, Numerical Modelling and Analysis of Cooling System of Electrical Transformer Dipped into Polymerised Resin - abstract | full text
• W.Sobieski, Performance of an Air-Air Ejector: An Attempt at Numerical Modelling - abstract | full text
• P.Synowiec, J.Wojcik and J.Bigda, CFD Methods as a Modern Tool in Optimisation of Hydrodynamic Conditions in Magma Crystallizers - abstract | full text

• A.Januszajtis, Family Background, Birth and Baptism of Daniel Gabriel Fahrenheit

Abstracts:

• J.Badur, K.Kosowski, R.Stepien and M.Piwowarski, Research into Flows in Turbine Blade Seals Part I: Research Methods

  Flows in the shroud clearance affect leakage losses and, thus, the turbine stage efficiency. However, at the same time, the distribution of pressure in the seal gaps plays an important role in the generation of the so-called "pressure forces", which may cause self-excited rotor vibrations of the aerodynamic type. Our investigations were carried out in order to show how a CFD code can cope with determining the pressure field in a rotor-blade shroud clearance. Experimental investigations of the pressure field in the shroud clearance were performed on a one-stage air model turbine of the impulse type. Measurements of pressure distribution were carried out for various rotor speeds and turbine loads. 3D calculations of flows in the model turbine were performed using the FLUENT CFD code. The calculations were carried out for the same variants which had been measured experimentally. In this part of the paper, the experimental stand and the numerical methods are described, while a detailed numerical analysis and a comparison between the experimental and the calculated results are presented in parts II and III, respectively.

• R.Stepien, K.Kosowski and J.Badur, Research into Flows in Turbine Blade Seals Part II: Numerical Analysis

  3D calculations of an axial model turbine of the impulse type were performed using the FLUENT CFD code. The calculations were carried out for variants which had been measured experimentally. Special attention was paid to the pressure field in the rotor blade shroud clearance. The Multiple Reference Frame method, the Mixing Plane method and the Sliding Mesh method were applied, and meshes of different types and configurations were used for calculations. Only the Sliding Mesh technique appeared to describe non-stationary effects and pressure pulsations in the turbine flow channels and clearances. In this part of the paper, numerical analysis is described, while the comparison between the experimental and the calculated results is presented in part III.

• K.Kosowski, R.Stepien, M.Piwowarski and J.Badur, Research into Flows in Turbine Blade Seals Part III: Numerical Calculations versus Experiment

  Experimental and theoretical investigations of the pressure field in the shroud clearance were performed on a one-stage air model turbine of the impulse type. Measurements of pressure distribution were carried out for various rotor speeds and turbine loads. 3D calculations of flows in this turbine were performed using the FLUENT CFD code. The calculations were carried out for variants which had been measured experimentally. The experimental data have been compared to theoretical results obtained with 3D codes for turbomachinery calculations. The Sliding Mesh and Multiple Reference methods have given very similar results of average values of pressure distribution and the velocity field in the shroud clearance. These results correspond to the experimental data. The pressure pulsations were determined only by the Sliding Mesh method, and these results have also been compared with the experiment. Stage flow calculations carried by the Sliding Mesh method with a structural shroud mesh and with a minimum number of 2-2.5 million cells have given a range of non-stationary pressure pulsations corresponding to the experimental data.

• J.Badur, M.Karcz and T.Wysocki, Chemistry-Turbulence Coupling in a Model of Inhomogenously Premixed Combustion

  Results of work on an extension of the turbulent flame speed model used for a mathematical description of partially premixed combustion are presented. The approach is based on the concept of internal coupling between the turbulence of the mixed fuel and oxidizer stream with the reaction progress variable. The model after implementation has been calibrated and tested on a BERL benchmark experiment.

• J.Baranski, J.Stasiek and W.Blasiak, Physical and Numerical Modeling of Heat-flow Processes in Tangentially Pulverized Fuel-Fired Boiler

  The paper presents results of three-dimensional physical modeling and computer simulations of fluid-flow structures, mixing and combustion processes in a 125 MW tangentially fuel-fired boiler and additional fuel - natural gas. This method is commonly called the reburning process, with an emphasis on the reduction of CO, NO_x and SO_x. The co-firing process is realized between the main coal burners and additional fuel nozzles. To improve the mixing and combustion processes, a physical technique, the so-called acid/alkali technique, is used to optimize the placement and direction of additional air and fuel nozzles. The best result obtained from physical modeling experiments is studied using numerical simulations with the FLUENT commercial code. Numerical modeling results are then used to analyze the performance of an industrial boiler. These results, compared with measurements in a real boiler, seem to be in good agreement with each other.

• P.Cyklis and R.Kantor, CFD Simulation as a Tool for the Identification of the Manifold Element Reaction to Pressure Pulsations Excitation (One- and Two-phase Flow)

  The periodicity of compressor operation is a source of pressure pulsations in volumetric compressor manifolds. An analysis of pressure pulsations is important for several reasons.
  The Helmholtz model, applied in all commercial programs offered by the companies professionally dealing with damping pressure pulsations, contains numerous simplifying assumptions; a straight pipe segment, with an ideal gas isentropic flow assumption, substitutes each element of the piping system. In many cases this model is insufficient. The existing experimental methods could not be used in the design of a muffler. The aim of this paper is to show a new method to identify an arbitrary fragment of a manifold, i.e. a method of identification of the appropriate complex transmittance matrix elements using CFD simulation. This method allows the liquid phase dispersed in the compressed gas and non ideal gas as a working medium to also be considered.
  The most important conclusion of this work is that identification of acoustic element parameters in the manifold, based on multi-dimensional simulation model, is feasible. The author obtained much better results from the developed method than those yielded by the classic Helmholtz model. A comparison between pure gas and gas with oil contamination is also shown in the paper.

• M.Karcz and J.Badur, A Turbulent Heat Flux Two-Equation θ'²-ε_θClosure Based on the V2F Turbulence Model

  The paper deals with the proposition of a two-equation turbulent heat flux closure without any damping function. The model has been based on Durbin's V2F dynamic turbulence closure and the Deng-Wu-Xi thermal turbulence model. Both models have been implemented into the FLUENT code by a User Defined Function. Results of numerical computation have been compared with experimental data for developing a thermal field in a pipe by Nagano and DNS heat transfer prediction for a two-dimensional channel flow by Kasagi.

• T.Michalek and T.A.Kowalewski, Simulations of the Water Freezing Process: Numerical Benchmarks

  Three numerical benchmarks concerning the freezing of water in small enclosures are analysed using the commercial FLUENT code. The first case is a steady-state natural convection in a differentially heated cavity for temperatures near the freezing point. In the second case, the freezing of water in a differentially heated cavity is simulated. The third case describes a simulation of freezing water in the presence of forced convection and a free surface flow. Two finite-differences numerical codes are used to verify results of the FLUENT simulations for the natural convection and solidification in the differentially heated cavity. It is found that the simulation of water solidification requires very fine meshes and short time steps, extending the computational time to the extreme.

• W.Podgorska and J.Baldyga, Drop Break-up and Coalescence in a Stirred Tank

  It is shown in the paper that drop size distribution in liquid-liquid dispersions is affected by both the fine-scale and the large-scale inhomogeneity of turbulence. Fine-scale inhomogeneity is related to the phenomenon of local intermittency and described using a multifractal formalism. Large-scale inhomogeneity is related to inhomogeneous distributions of the locally averaged properties of turbulence, including the rate of energy dissipation and the integral scale of turbulence. Large-scale distributions of the properties of turbulence in a stirred tank are considered with a network of well-mixed zones. CFD methods are used to compute the properties of turbulence in these zones. A model taking into account inhomogeneity of both types explains the effect of the system's scale on drop size; it predicts smaller maximum stable drop sizes than the classic Kolmogorov theory of turbulence. The model predictions agree well with experimental data.

• A.Rozen and J.Baldyga, Influence of Viscosity Difference on the Instability of the Core-Annular Flow

  The process of destabilisation of an axi-symmetric, core-annular flow (CAF) of two Newtonian fully miscible liquids widely differing in viscosity is investigated. Formation of periodic structures is observed in experiments, predicted by the linear stability theory and simulated numerically using the volume of fluid method (VOF). Possible influence of this phenomenon on mixing on the molecular scale is discussed.

• J.Smolka, A.J.Nowak and L.C.Wrobel, Numerical Modelling and Analysis of Cooling System of Electrical Transformer Dipped into Polymerised Resin

  This paper discusses a numerical model of the heat dissipation processes in an electrical transformer dipped into polymerised resin. The transformer is cooled by both natural convection (via the ambient air) and forced convection (via the water cooling system attached to one of the transformer casing walls). Two cases have been compared, i.e. the cooler connected to the bottom or the top wall of the transformer container, respectively. In order to improve the modelling of the natural convection problem, an independent geometrical model of the surrounding air was created and considered separately. The continuity of temperature and heat flux along the interface between the transformer and air was enforced by an iterative procedure. This procedure allowed one to calculate and then prescribe local heat fluxes to the external walls of the transformer. The numerical results obtained in this project have yielded information on the efficiency of the analysed cooling system.

• W.Sobieski, Performance of an Air-Air Ejector: An Attempt at Numerical Modelling

  The paper describes a numerical attempt to model the operation of an air-air ejector by means of a commercial numerical code. The obtained results are compared with the results of a laboratory experiment - particularly in relation to the diffuser of the modelled ejector. The substantial discrepancies between the performance characteristics obtained in the laboratory experiment and by numerical modelling are interpreted as a result of the absence of appropriate models of turbulence and mixing in the numerical code used. The paper also presents additional numerical simulations that emphasize the importance of unsteady phenomena (bifurcations of working fluid jet, turbulent mixing at the boundary of components) in the modelling of ejector equipment.

• P.Synowiec, J.Wojcik and J.Bigda, CFD Methods as a Modern Tool in Optimisation of Hydrodynamic Conditions in Magma Crystallizers

  Selected parameters concerning the optimisation of hydrodynamics in magma crystallizers are discussed. At this stage, results of CFD (Computational Fluid Dynamics) simulations are shown, focused on the effect of the geometrical configuration of a crystallizer on energy dissipation rate, axial velocity field and general hydraulic efficiency.
  The influence of the shape of the apparatus' bottom, diameter of the stirrer, its location and presence of the draft tube on, respectively: (i) unit power input distribution, (ii) the average mixing power, and (iii) pumping capacity have been takeninto consideration.
  The results obtained from simulations have been compared with experiments and literature data.