Begell House Inc.
Computational Thermal Sciences: An International Journal
CTS
1940-2503
4
1
2012
NUMERICAL SIMULATION OF A TURBULENT FREE JET ISSUING FROM A RECTANGULAR NOZZLE
1-22
Mohsen
Akbarzadeh
Department of Mechanical and Manufacturing Engineering, University of Manitoba, Winnipeg, Manitoba, R3T 5V6, Canada
Madjid
Birouk
Department of Mechanical and Manufacturing Engineering, University of Manitoba, Winnipeg, MB, R3T 5V6 Canada
Brahim
Sarh
ICARE-CNRS, 1С Avenue de la Recherche Scientifique, Orléans 4571, France
The aim of this paper is to demonstrate that the simple Reynolds-averaged Navier-Stokes (RANS) two-equation standard k − ε turbulence model is capable of predicting the main characteristics of turbulent free jets issuing from three-dimensional complex rectangular nozzle geometry. This paper also investigates the main characteristics of turbulent rectangular free shear jets by varying the nozzle aspect ratio. The computations were performed by using RANS equations and the turbulence terms were handled by testing the standard two-equation k − ε or k −ω turbulence models. The solution of the governing equations was obtained by using the finite volume method (FVM), where the SIMPLEC algorithm was adopted with a staggered grid to prevent decoupling velocity and pressure fields. The inflow boundary conditions including the inflow velocity profile and the turbulence intensity level were adopted from published experimental data. The main results demonstrate that the standard k − ε model, when applied with appropriate inflow boundary conditions, provides successful predictions of the main features of turbulent free jets issuing from rectangular nozzles, including the decay rate in the near- and far-field region, the spreading rate except in the very near field, vena contracta, and axis switching. The numerical results show also that increasing the nozzle aspect ratio leads to an increase in fluid entrainment closer to the nozzle exit in the near field, which is in complete agreement with experimental observations.
ON THE VORTEX SHEDDING MECHANISM BEHIND A CIRCULAR CYLINDER SUBJECTED TO CROSS BUOYANCY AT LOW REYNOLDS NUMBERS
23-38
Dipankar
Chatterjee
Academy of Scientific and Innovative Research (AcSIR), CSIR-Central Mechanical
Engineering Research Institute, Durgapur-713209, India; Advanced Design and Analysis Group, CSIR-Central Mechanical Engineering Research
Institute Durgapur-713209, India
Bittagopal
Mondal
Simulation and Modeling Laboratory, CSIR- Central Mechanical Engineering Research Institute, Durgapur-713209, India
The paper discusses the influences of thermal buoyancy on the vortex shedding mechanisms behind a heated circular cylinder in an infinite medium for cross flow at low Reynolds numbers. The major contribution is the quantification and assessment of the characteristic behavior of the critical buoyancy parameter (Richardson number) for transition from steady to unsteady periodic flow. A two-dimensional numerical simulation is performed in this regard by using a finite volume method based on the PISO (pressure-implicit with splitting of operators) algorithm in a collocated grid system. The range of Reynolds number is chosen to be 10−45. In this range the flow field is found to be steady and separated without the superimposed thermal buoyancy (i.e., for pure forced convection). However, as the intensity of buoyancy increases (i.e., Richardson number > 0), the flow becomes unstable and eventually, at some critical value of the Richardson number, periodic vortex shedding is observed to characterize the flow and thermal fields. The effects of superimposed thermal buoyancy are studied for the Richardson number range 0−2. The critical Richardson number for the onset of vortex shedding is found to decrease (which is in clear contrast with other findings) and the dimensionless frequency of vortex shedding (Strouhal number) is found to increase with Reynolds number in the chosen range.
AN IMPROVED LUMPED MODEL FOR TRANSIENT HEAT CONDUCTION IN DIFFERENT GEOMETRIES
39-48
Santosh K.
Sahu
Discipline of Mechanical Engineering, School of Engineering Indian Institute of
Technology Indore, Madhya Pradesh, 453552, India
Pritinika
Behera
Department of Mechanical Engineering, National Institute of Technology, Rourkela-769008, India
This paper deals with the analysis of transient heat conduction in various geometries, namely, large plate, long cylinder, and sphere by employing the polynomial approximation method. A wide variety of guess profiles have been tried to obtain the temporal variation of temperature in the solid and the best profile is identified for different geometry. The effect of convection has been expressed by a unique parameter termed as the modified Biot number, the exact value of which depends on the geometry and process parameters. This parameter enables us to provide a generalized three-parameter relationship for transient conduction and eliminates the development of different models for different geometry. Compared to other analytical models the present prediction exhibits excellent agreement with the exact solution (0.67% error in large plate, 3.84% error in long cylinder, and 1.47% in sphere).
COMPUTATIONAL FLUID DYNAMICS MODELING TOWARD CLEAN COMBUSTION
49-65
K. K. J. Ranga
Dinesh
School of Engineering, Cranfield University, Cranfield; Engineering Department, Lancaster University Lancaster, LA1 4YR, UK
Michael P.
Kirkpatrick
School of Aerospace, Mechanical and Mechatronic Engineering, The University of Sydney, NSW 2006, Australia
A.
Odedra
Hamworthy Combustion Engineering Limited, Fleets Corner, Poole, Dorset, BH17 0LA, United Kingdom
A turbulent hydrogen−air non-premixed jet flame is studied using three-dimensional large eddy simulation (LES) and a laminar flamelet model based on detailed chemical kinetics. The LES solves the governing equations on a structured Cartesian grid using a finite volume method, with turbulence and combustion modeling based on the localized dynamic Smagorinsky model and the steady laminar flamelet model, respectively. The LES results are validated against experimental measurements and overall the LES yields good qualitative and quantitative agreement with the experimental observations. Analysis showed that the LES gives good prediction of the flow field, flame temperature, and major species. The three-dimensional transient LES demonstrates the variation of low and high temperature structures and both transient and mean predictions show that the high temperature regions and combustion product appear close to the jet centerline. The present findings provide useful details on fundamental issues of turbulence−chemistry interactions of hydrogen combustion and help to identify potential pathways for combustion modeling toward clean combustion.
NUMERICAL SIMULATION OF LAMINAR DIFFUSION FLAME WITH FINITE RATE CHEMISTRY AND VARIABLE PROPERTY FORMULATION
67-76
A. K.
Chowdhuri
Department of Mechanical Engineering, Bengal Engineering and Science University, Shibpur, Howrah 7111103, India
A. J.
Bhowal
Department of Mechanical Engineering, Heritage Institute of Technology, East Kolkata Township, Kolkata 700107, India
S.
Chakraborti
Department of Mechanical Engineering, Bengal Engineering and Science University, Shibpur, Howrah 7111103, India
Bijan Kumar
Mandal
Department of Mechanical Engineering, Bengal Engineering and Science University, Shibpur, Howrah 7111103, India
A numerical model is used for simulation of a confined axisymmetric laminar jet diffusion flame under normal gravity and pressure conditions to predict the velocity, temperature, and species distributions. An explicit finite difference technique has been adopted for the numerical simulation of reacting flow with finite rate chemistry and variable thermo-dynamic and transport properties. The predictions match well with the experimental results available in the literature. A recirculation of ambient air is observed to extend from the exit plane into the domain adjacent to the wall. Radial velocity is never positive (away from the axis) in the solution domain. High temperature and high CO2 concentration zones are confined to a small radial distance. The H2O distribution shows a similar pattern as that of CO2 distribution. On the other hand, CO is almost absent above 10 cm axial height.
BOUNDARY CONDITION EFFECTS ON NATURAL CONVECTION OF BINGHAM FLUIDS IN A SQUARE ENCLOSURE WITH DIFFERENTIALLY HEATED HORIZONTAL WALLS
77-97
Osman
Turan
Deptartment of Mechanical Engineering, Karadeniz Technical University, Turkey
Robert J.
Poole
School of Engineering, University of Liverpool, Brownlow Hill, Liverpool, L69 3GH, UK
Nilanjan
Chakraborty
School of Mechanical and Systems Engineering, Newcastle University, Newcastle-Upon-Tyne, NE17RU, United Kingdom
Natural convection of Bingham fluids in square enclosures with differentially heated horizontal walls has been numerically analyzed for both constant wall temperature (CWT) and constant wall heat flux (CWHF) boundary conditions for different values of Bingham number Bn (i.e., nondimensional yield stress) for nominal Rayleigh and Prandtl numbers ranging from 103 to 105 and from 0.1 to 100, respectively. A semi-implicit pressure-based algorithm is used to solve the steady-state governing equations in the context of the finite-volume methodology in two dimensions. It has been found that the mean Nusselt number Nu increases with increasing Rayleigh number, but Nu is found to be smaller in Bingham fluids than in Newtonian fluids (for the same nominal values of Rayleigh and Prandtl numbers) due to augmented flow resistance in Bingham fluids. Moreover, Nu monotonically decreases with increasing Bingham number irrespective of the boundary condition. Bingham fluids exhibit nonmonotonic Prandtl number Pr dependence on Nu and a detailed physical explanation has been provided for this behavior. Although variation of Nu in response to changes in Rayleigh, Prandtl, and Bingham numbers remains qualitatively similar for both CWT and CWHF boundary conditions, Nu for the CWHF boundary condition for high values of Rayleigh number is found to be smaller than the value obtained for the corresponding CWT configuration for a given set of values of Prandtl and Bingham numbers. The physical reasons for the weaker convective effects in the CWHF boundary condition than in the CWT boundary condition, especially for high values of Rayleigh number, have been explained through a detailed scaling analysis. The scaling relations are used to propose correlations for Nu for both CWT and CWHF boundary conditions and the correlations are shown to capture Nu satisfactorily for the range of Rayleigh, Prandtl, and Bingham numbers considered in this analysis.