This action might not be possible to undo. Are you sure you want to continue?
By M.K.AKILAN Reg. No: 71103402001 of KONGU ENGINEERING COLLEGE Perundurai, Erode-638 052 A PROJECT REPORT Submitted to the FACULTY OF MECHANICAL ENGINEERING In partial fulfillment of the requirements for the award of the degree of MASTER OF ENGINEERING IN CAD/CAM June 2005
KONGU ENGINEERING COLLEGE, PERUNDURAI ERODE – 638 052 DEPARTMENT OF MECHANICAL ENGINEERING (PG) CERTIFICATE OF PROJECT APPROVAL
This is to certify that the project report titled PERFORMANCE ANALYSIS OF A PEBBLE BED AIR HEATER USING CFD TECHNIQUE is the approved record of work done by M.K.AKILAN
Reg. No: 71103402001
IN PARTIAL FULFILLMENT FOR THE AWARD OF THE DEGREE OF MASTER OF ENGINEERING IN CAD/CAM OF ANNA UNIVERSITY, CHENNAI – 600 025 DURING THE YEAR 2004 –2005
Head of the Department
Submitted for the University examination held on ____________
Certified that this project report titled PERFORMANCE ANALYSIS OF A PEBBLE BED AIR HEATER USING CFD TECHNIQUE is the bonafide work of Mr.M.K.AKILAN who carried out the research under my supervision. Certified further, that to the best of my knowledge the work reported here in does not form part of any other project report or dissertation on the basis of which a degree or award was conferred on an earlier occasion on this or any other candidate.
Head of the Department
Bharat Heavy Electr
Research & Development
MHD Campus, Boiler Project Tiruchirapalli- 620 014; Ph: 2
This is to certify that M.K.Akilan – Roll. no. 23M 638052 has executed his curriculum project work titled, PERFORMANCE ANALYSIS OF PEBBLE BED AIR TECHNIQUE I certify that the work was carried out under my guidance and he excellent and I wish him all success in his future endeavors.
M.E -Cad/Cam., (P.G.Mechanical Engineering Dept., Kongu Eng
scope of work during the period 12-01-04 and 12-04-05. Hi
Tuesday, June 07, 2005 Tiruchirapalli-14
The Pebble bed air heaters are used to provide high temperature process gases in glass and steel industries, power plants and waste heat recovery systems. In all these situations, the temperature required is obtained by a regenerative air Pre-heater system consisting of a packed bed of low thermal conductivity materials like Alumina or Zirconia. Performance analysis of pebble bed heat exchanger which includes, Heat transfer between the gas to pebble and Pebble to air, Pressure drop suffered by the gas/air while passing through the packing and the temperature distribution within the core of the pebble. A 1D thermal analysis would be done using Matlab programming language. The code will be generated for the performance prediction of the pebble bed air heater for transient condition. The temperature profiles of the pebble bed with respect to time will also be arrived. The above said analysis will be validated using computational fluid dynamics (CFD).The air heater will be modeled and analyzed for transient conditions. The results will be verified and compared with the matlab results.
Sathiyamoorthy and our beloved Principal Dr.A.M.Natarajan for their support and facilities provided for us to do the project in a better manner. I also thank wholeheatedly Dr.A.Velluswamy, Dean (R&D), for his timely advice and help in the completion of the project. It is my duty to render my gratitude and thanks to our beloved head of the department Prof.E.Vallinayagam and Prof.S.P.Arul who has supported us and guided us throughout our project work. I also thank the project co-ordinator Prof.V.Hariharan, who has given valuable information and support for the successful completion of the project.
I thank with sincere gratitude Prof.K.Krishnamoorthy who was a true friend, and philosopher, with whose kind help and able guidance the project work is in the right form and shape. I also thank him for his timely help in collecting materials and details for doing the project in a successful manner. I also thank all faculty members and supporting staff who have helped me in my work and thank all my friends you have been behind me for the deeds that made me complete my project. Last but not the least I thank my parents whose blessings and wishes have given me the boost to accomplish my project in a successful manner.
TABLE OF CONTENTS List of tables List of figures Nomenclature ix x xii
INTRODUCTION 1.1. Magneto Hydrodynamics 1.2. Mhd In B.H.E.L - Trichy (R&D) 1.3. High Temperature Air Preheaters 1.4. Types Of The Regenerator 1.4.1. Recuperator 1.4.2. Regenerator 18.104.22.168. Pebble Bed Regenerator 22.214.171.124. Characteristics Of The Pebble Bed
1 1 3 5 5 6 6 8 10
11 20 20 21 22 22 23 24 24 24
PROBLEM FORMULATION AND METHODOLOGY 3.1. Problem Formulation 3.2. Problem Solving Methodology
SOLUTION PROCEDURE 4.1. Matlab Overview 4.1. Matlab 4.2. The Matlab System 4.2.1 Development Environment 4.2.2 The Matlab Mathematical Function Library
4.2.3 The Matlab Language 4.2.4 Matlab Procedure And Steps 4.3.Computational Fluid Dynamics 4.3.1 Basic Steps In Cfd 126.96.36.199. Preprocessing Phase 188.8.131.52. Solution Phase 184.108.40.206. Post Processing Phase RESULT AND DISCUSSION 5. 5.1. TECHNIQUE-I 5.1.1. Matlab 5.1.2. Heating Phase 5.1.3. Cooling Phase 5.2. TECHNIQUE-II
24 25 28 32 32 40 41 42 42 42 43 44 46 46 47 49 52 52 52
5.2.1. Computational fluid dynamics. 5.2.2. Heating Phase 5.2.3. Cooling Phase CONCLUSION 6. 6.1. Conclusion 6.2. Scope Of The Future Work
LIST OF TABLES
5.1.1 5.1.2 5.1.3 5.2.1
Time Period in matlab Input variables for the heating phase Input variables for the cooling phase Time Period in Fluent
42 43 45 46
LIST OF FIGURES
1.1. 1.2. 1.3. 2.1. 2.2. 2.3. 4.1. 4.2. 4.3. 4.4. 4.5. 5.1.1. 5.1.2. 5.2.1. 5.2.2. 5.2.3. 5.2.4.
MHD layout in BHEL. Air Preheaters. Pebble bed air heater. Test Facility. Matrices of Lucite spheres of refractive. Experimental set up. Flowchart for Matlab programming. Symmetric pebble bed air heater model. Grid model of the heater. Boundary Conditions of Pebble Bed Air Heater. Continuum Conditions of Pebble Bed Air Heater. Results for 2 Hours in the matlab for temperature profile of pebble in heating phase. Results for 12 minutes in the matlab for temperature profile of pebble in cooling phase. Heating phase at different time steps. Heating phase at different time steps A, B, C, D. Cooling phase at different time steps A, B, C, D. Cooling phase at different time steps.
4 7 9 15 17 18 30 34 36 38 40 44 45 47 48 50 51
NOMENCLATURE Cp d V R Re T u h k H n Nu Pr A Specific Heat (J/kgK) Diameter (m) Volume (m^3) Radius (m) Reynolds Number Temperature (C) Superficial Velocity In Pebble Bed (m/sec) Heat Transfer Coefficient (W/m2K) Thermal Conductivity (W/mK) Height (m) Total Number Of Particles Contained In The Column Nusselt Number, hd/k Prandtl Number, μCp/k Total Surface Area (m^2)
Greek ρ Δ μ γ ε Subscript C p Cylinder Pebble Particle Density (kg/m3) Difference Dynamic Viscosity (N sec/m2) Kinematic Viscosity (m2/sec) Porosity/Void Fraction
CHAPTER 1 1. INTRODUCTION
1.1. MAGNETO HYDRODYNAMICS:
Magneto hydrodynamics (MHD) power generation is a method of generating electric power by passing an electrically conducting fluid through a magnetic field. By means of the interaction of the conducting fluid with the magnetic field, the MHD generator transforms the internal energy of the conducting fluid into electric power in much the same way as a convectional turbogenerator does by means of the interaction of a solid conductor with a magnetic field.
In principle the working fluid can be any electrically conducting fluid, such as salt water, liquid metal (or) hot ionized gas. For central station power generation applications, the most suitable working fluid is a hot ionized gas. This can be a relatively clean gas, (e.g.) a noble gas heated in an externally fired heat exchanger, or it can be composed of combustion products from any fossil fuel.
There are two basic types of MHD energy conversion system • • Closed cycle energy conversion system. Open cycle energy conversion system.
The closed cycle typically operates using a clean gas which is recycled; open cycle generally operates using combustion products which are then
discarded. In the most MHD system designs the gas exiting the topping cycle exhausts either into a radiant boiler and is used to raise steam, (or) it exhausts into a direct - fired air heater and is used to preheat the primary combustion air.
An alternative use of the exhaust gas is for chemical regeneration, in which the exhausts gases are used to process the fuel from its as received form into a more beneficial one. Chemical regeneration has been proposed for use with natural gas and oil as well as with coal.
In a coal based system the generator exhaust gas is used in a multistage process in which the incoming coal undergoes devolatilization, gasification, and partial combustion at atmospheric pressure to produce a low heating value fuel gas for the primary combustor. The thermal energy recovered from the exhaust gas and stored in the fuel gas is roughly 40% of the combustion energy of the fuel into the primary combustor, so that a substantial increase in fuel energy is achieved by use of chemical regeneration.
In the most advanced MHD power plants are expected to use the preheater temperature of up to 1400 c. This can be achieved by the Regenerative heat exchangers only. In some areas the recuperator were also used.
1.2. MHD IN B.H.E.L - TRICHY (R&D): The Indian National MHD programme was initiated in 1977 with funding from the Department of science and technology, Government of India. This programme was jointly executed by BHARAT HEAVY ELECTRICALS LIMITED and BHABHA ATOMIC RESEARCH CENTRE in close co-operation with HIGH
TEMPERATURE INSTITUTE, MOSCOW, USSR – the pioneers in large scale MHD activities.
Under this programme an MHD pilot plant facility has been established at TIRUCHIRAPPALLI with 5 mw (thermal) input with possible extension up to 15mw input to yield research and operating inputs towards future open cycle MHD power generation systems in the Indian context.
The research facility in the pilot plant has a flow train consisting of regenerative high temperature air preheaters and related combustors operating in the cyclic mode, high temperature values, expansion joints, FRP duct for electrical isolation of the MHD channel, a high magnet, and a gas cooling system.
The pilot plant is elaborately instrumented. Non-conventional diagnostic equipment is used for measuring high temperatures, electrical conductivities etc. the flow train is fed by auxiliary systems such as oxygen plants, gas plants, low conductivity water systems, seed recovery systems, etc.The layout of 200MW Open cycle coal fired MHD steam pilot plant has been given.
Fig .1.1. MHD layout in BHEL
Regenerators are likely to be the key pre-heat source in the MHD systems. So there were no specified correlation or formulae to calculate the performance of the pebble bed air heater. Here coding has been returned to predict the 1D performance of the pebble bed air heater using MATLAB version 6.5.And also analysis was done in the COMPUTATIONAL FLUID DYNAMICS.
1.3. HIGH TEMPERATURE AIR PREHEATERS
accomplished in one of two ways. One way is to use the heat energy of the MHD generator exhaust gas directly; in this case, the preheater is classified as • • Directly fired Separately fired
In the case of directly fired is located in the MHD generator exhaust as part of the bottoming plant. This type offer the potential of higher cycle efficiencies than can be achieved with separately fired preheats at the same oxidizer temperature. Airs preheat temperature requirements of 2250 -2300 k are anticipated for natural gas - fired system, and about 2000 k for oil or coal fired system. Use of 32-40% oxygen enrichment lowers the preheat temperature requirement to a moderate 900 - 1000 k, which can be attained with conventional metal type tubular heat exchangers.
1.4. TYPES OF THE REGENERATOR Depending on the use of separately fired high temperature regenerative heat exchangers, where heat transferred for a time from a hot fluid, were we have two types to achieve the similar temperature are:
• • 1.4.1. Recuperator
This is a type of heat exchanger, in which heat is transferred continuously from one fluid to another through a solid wall which separates the two fluids. In the below diagram it is given.
Metallic recuperators are used widely in industry, but are limited for MHD use to about 1250 k, because of corrosive problems caused by seed and ash, as well as mechanical strength problems caused by the pressure requirements. Whereas ceramic recuperator can operate at high temperatures, development of ceramic recuperator for MHD has not been pursued because of severe problems related to • • • Fabrication. Fluid leakage. Mechanical strength.
1.4.2. REGENERATOR Regenerators are compact heat exchangers in which heat is alternately stored and removed using a heat storage matrix. During the heating period, the hot gas passes through the regenerator and transfers heat to the matrix. After a certain time (hot period), the hot gas flow stops and the cold gas flow initiates, normally in the opposite direction to that of the hot gas. The cold gas picks up the heat stored in the matrix.
Fig .1.2. AIR PREHEATER
Regenerators employed in the glass and aluminum industries are designed such that they can withstand entrance gas temperature of about 1400 0c. At this level of temperature the matrix must be constructed from ceramic materials, which introduce conduction effects into the overall heat transfer in addition to the convection and radiation.
This necessitates the prediction of an accurate value for the heat transfer coefficient to accommodate the effects of all mechanisms of heat transfer. In order to have continuous operation, the installation must comprise at least two distinct matrix, beds. So that at all period time's one matrix is being heated while the other is being cooled. Theoretical performance of regenerators can be predicted by solving a set of partial differential equations governing heat transfer between the two fluid streams and the solid matrix. Based on the simplest mathematical model of the
regenerator effectiveness is only dependent upon four dimensionless parameters. They can be used to determine the important independent variables as well as the design and performance of this type of equipment.
Regenerators may be divided into two groups; • • Fixed-bed regenerator. Rotary bed regenerator.
In fixed-bed regenerators the storage material is stationary and valves are employed to alternately direct the hot and cold gas streams through the storage material. Such systems have usage in the steel, glass making and gas turbine plants as waste heat recovery systems, particularly for the stack gases.
220.127.116.11. PEBBLE BED REGENERATOR The pebble bed air heater is also known as fixed (or) packed bed regenerator. The fixed-bed regenerator have been used in the glass and steel industries to preheat air to 1350 -1650 k, but these operate with relatively clean gases compared to the MHD combustion gases
One design, fabrication and testing of a regenerator for MHD use, involved a regenerator matrix (or) bed packing, of 8.5 m in height and 7.9 m in dia. The matrix geometry consisted of cored bricks, or 0checkers, made of fusion cast magnesia - sipnel, steel pall rings, steel rasching rings, ceramic saddles.
Cycle times were 1280 seconds on MHD gas flow, 760 seconds on oxidant, with 360 seconds for switching. Operating times up to 1470 hours were achieved. System heat, leakage and pressure losses were all within acceptable limits.
A pebble bed instead of a cored brick matrix is used. The pebbles are made of alumina spheres, 20 mm in diameter. Heat transfer co-efficient 3-4 times greater than for checker work matrices are achieved. A prototype device 400 m 3 in volume has been operated for three years at an industrial blast furnace, achieving preheats temperatures of 1670 to 1770 k.
18.104.22.168. CHARACTERISTICS OF THE PEBBLE BED • • • The heat transfer co-efficient between the air and solid is high in bed. The conductivity of the bed is low, where is no air flow. The pressure drop through the bed is low
CHAPTER 2 LITERATURE REVIEW
For the literature review the journal papers and books related to the heat transfer in the pebble bed air heater has been referred. From these materials the results has been taken as the guidelines.
Earlier Problems in Fixed Beds As was mentioned before a large deficiency in the modeling of packed beds in industry is the assumption of plug flow. It is already generally accepted that the void fraction in unstructured beds is large near the wall and fluid flow is channeled in these areas causing radial inhomogenities in the overall flow profiles (Kalthoff and Vortmeyer,1980; Haidegger et al., 1989; Lerou and Froment, 1977; Froment and Bischoff, 1979; Papageorgiou and Froment, 1995). These radial distributions of the axial flow have been measured by different groups outside the bed (Morales
et al., 1951; Price, 1968; Schuster and Vortmeyer, 1981; Ziolkowska and Ziolkowski, 1993; Daszkowski, 1991). Direct measurements of gas flow inside the bed have not been possible yet. Other groups measured radial profiles below the packing, averaging to get a general velocity profile by repacking a 3 < N < 11 column several times. The measured results were extrapolated using an extended Brinkman equation to get radial flow profiles in the bed (Bey and Eigenberger, 1997).
McGreavy et al. (1984) used Laser Doppler Velocimetry (LDV) in low N (6.5) packed beds. To be able to access the internals of the packed bed several optical access arrangements were made, these arrangements disturbed the continuity of the packing. They showed (McGreavy et al., 1986) that flow profiles in the bed were different from profiles above the bed; they found a region in the beginning of the bed where flow development took place and a region near the bed exit where flow degradation took place. This indicates that measurement of flow profiles outside the bed is inadequate.
Recently the group of Vortmeyer reported velocity measurements taken inside a packed bed using LDV (Giese et al., 1998). Their setup consisted of all glass particles in a glass tube with a tube to particle ratio of approximately 9 using an organic mixture with a refraction index similar to the glass as a fluid. In these measurements the focus was on low Reynolds numbers, <100 and velocities were averaged to get superficial velocities. The main focus of these groups has been the behavior of the superficial velocity and the porosity of the bed. Most groups use a number of different packing materials to find the effects of the packing on radial distributions of superficial velocities.
Thompson and Fogler (1997) used a network-model to model flow in a packed bed. The smallest element in a network model was created in a Delauney tessellation using basically one element per void in the packing. The different beds were established using a computer simulation method for creating a random bed. Computer simulation and experimental data were compared using a N = 12 bed of approximately 7500 particles. The computer simulation treated the bed as a porous medium, neglecting specifics in the flow profile. It was, however, not a homogeneous porous bed, the porosity was established using the computer generated random packed bed. .
Fluid flow and heat transfer analysis through packed beds is a common occurrence especially in the field of chemical engineering. Some of the limitations we encounter in experimental studies can be overcome by CFD techniques. Harris et. al. presents a good overview on the capabilities of CFD in chemical reactor engineering. Some of the previous works like Logtenburg and Dixon , Dixon and van Dongeren , Logtenburg, et. al. , mainly deal with the heat transfer analysis in packed beds using the CFD techniques. This work discusses a CFD technique that involves multi-component flow of gases through packed beds.
The commercial CFD package FLUENT™  is used for numerical simulations. The packed bed of monomer particles is modeled using a porous medium approach. Since the gas flows at a very low velocity inside the packed bed, the flow is treated as laminar flow. The flow rate and hence the velocity of the carrier gas is restricted by conditions that the superficial velocity in the packed bed should be much lower than the minimum fluidization velocity of the monomer particles. Fluidization has to be prevented inside the packed bed to eliminate any undesirable effects such as particle entrapment. The porous medium is assumed to be isotropic.
R.C. Hendricks  et al have presented a videotape presentation of flow in a packed bed of spheres. The refraction indices of the oil and the test array of spheres were closely matched, and the flow was seeded with aluminum oxide particles. Planar laser light provided a two-dimensional projection of the flow field, and traverse simulated a three-dimensional image of the entire flow field. Light focusing and reflection rendered the spheres black, permitting visualization of the planar circular interfaces in both the axial and transverse directions. Flows were observed near the wall-sphere interface and within the set of spheres.
The CFD model required that a representative section of a packed bed be formed and griddled, enclosing and cutting six spheres so that symmetry conditions could be imposed at all cross-boundaries. Simulations had to be made with the flow direction at right angles to that used in the experiments, however, to take advantage of flow symmetry. Careful attention to detail was required for proper gridding.The flow field was three-dimensional and complex to describe, yet the most prominent finding was flow threads, as computed in the representative ‘cube' of spheres with face symmetry and conclusively.
The test facility consisted of an oil tunnel, flow system components, video recording equipment, a 4-W continuous-wave argon-ion laser, lens systems, data recorders, and a test configuration .The matrix of spheres and entry ramp spacer was fabricated to fill the tunnel cross section. Both the laser light sheet and the viewing port walls were Lucite.
Fig.2.1. Test facility
The point of contact yet permitted construction of a regular array of staggered spheres and a clear view through the spheres when they were immersed in the oil, which also had the same refraction index as Lucite. The spheres in the bottom layer were ground flat, and their height was adjusted to exactly fit the tunnel. No other spheres were altered. The width was adjusted by using a gradual entry ramp that blocked a portion of the tunnel and at the same time eliminated any recalculating flow at the entry of the array.
The packed sphere bed contained 172 spheres. The 7-by-7 longitudinal array had transverse array of four 12.7-mm diameter spheres. The volume of the test array envelope was Vt = 305 cm3, the solid volume of spheres was Vs = 184 cm3. The void Vo = Vt – Vs = 121 cm3, with an effective open cross section for flow of Ae = 17.9 cm2. The coherent beam of the 4CW argon-ion laser was directed by micro metrically adjustable mirrors through two cylindrical lenses positioned at 90° to each other, through the Lucite tunnel window, and into the test section.
The light sheet was approximately 0.1 mm thick, and the flow was seeded with magnesium oxide flow tracers. Micrometric adjustments controlled the position of the light slices scanning across the test section and provided, through composition, a three-dimensional visualization of the flow field. The light was naturally focused and reflected at the boundary interfaces between the working fluid and the spheres, rendering the spheres black and permitting visualization of the planar circular interfaces in both the axial and transverse directions. The images Were recorded on a videocassette recorder. Other instrumentation included static pressure taps and thermocouples upstream and downstream of the test matrix and two spheres coated with liquid crystal with a narrow band response about 21 °C.
A videotape presentation is provided that enables the viewer to compare experimental and virtual computational fluid dynamic flows through packed beds of spheres or porous media. For the experiment the refraction indices of the oil working fluid and the spheres were closely matched so that only the boundaries of the physical spheres remained faintly visible to the eye and the cameras.
The pressure drops per unit length in the packed bed were also calculated and compared with results from Ergun’s theory for porous media, and good correlation between the two sets was obtained.
Fig.2.2. Matrices of Lucite spheres of refractive
C.Briens  et al this work is concerned with finding out an effective way of eliminating oxygen from a packed bed of monomer particles. This process finds application in industries involved in the manufacture of Nylon12. In the manufacture of the polymer Nylon12, the polymerization reaction is hindered by the presence of oxygen. This work involves the numerical simulation and experimental verification of the flow in a packed bed. In addition, a parametric study is carried out for the parameters such as the number of injectors, the radial position of injectors, and the position of the injectors along the circumference of the packed bed to find out the best possible combination for effective elimination of the oxygen.
The packed bed will be modeled using a porous medium approach available in the commercial computational fluid dynamics (CFD) software FLUENT™. The fluid flow inside the packed bed will be a multi-component gas
flow through a porous medium. The simulation results are validated by comparing with the experimental results.
Fluid flow and heat transfer analysis through packed beds is a common occurrence especially in the field of chemical engineering. Some of the limitations we encounter in experimental studies can be overcome by CFD techniques. Its presents a good overview on the capabilities of CFD in chemical reactor engineering. Its mainly deal with the heat transfer analysis in packed beds using the CFD techniques. This work discusses a CFD technique that involves multicomponent flow of gases through packed beds.
The commercial CFD package FLUENT™  is used for numerical simulations. The packed bed of monomer particles is modeled using a porous medium approach. Since the gas flows at a very low velocity inside the packed bed, the flow is treated as laminar flow.
Fig.2.3. Experimental set up
The flow rate and hence the velocity of the carrier gas is restricted by conditions that the superficial velocity in the packed bed should be much lower than the minimum fluidization velocity of the monomer particles. However, the numerical
approach used to model the fluid flow in the packed bed needs to be validated against the experimental data. Therefore, experiments are also carried out in this study for an experimental packed bed to validate the numerical results.
A numerical procedure to simulate the fluid flow in a packed bed is presented in this study. The simulations are carried out for an experimental packed bed and the numerical results are compared with the experimental data. The comparison indicates that the numerical results agree well with the experimental data. The parametric studies on the effect of the number of injectors, the radial position of injectors, and the position of the injectors along the circumference of the packed bed on the effectiveness of oxygen elimination from the packed bed are also carried. There are some parameter will affect the performance of the pebble bed air heater, so to predict it.
PROBLEM FORMULATION AND METHODOLOGY
3.1. PROBLEM FORMULATION Pebble bed air heater are used to provide high temperature process gases in the
• • •
Glass and Steel industries, Power plants Waste heat recovery systems.
For these applications there is no direct analytical method or correlations to predict the performance of the air heater with respect to time. In all these situations the temperature levels require the regenerator packing to be made from the low thermal conductivity materials such as ceramic. Performance of fixed bed heat exchangers must accommodate the heat transfer from the gas to the packing surface, packing surface to the air. and the temperature distribution within the core of the ceramic spheres.
Most of the mathematical models employed in theory and practice assume either that the internal thermal resistance to heat flow within the core is negligible or that the resistance can be incorporated in a lumped convective heat transfer coefficient at the surface. The main aim was to predict and analysis the performance of the regenerator in transient state. The investigation was carried out separately for two phases.
Heating phase Cooling phase
In these two phases the heat transfer and pressure drop will be evaluated with respect to time. The matlab code will be developed for the performance prediction and the simulation results will obtained in computational fluid dynamics.
3.2. PROBLEM SOLVING METHODOLOGY In MATLAB the one dimensional heat transfer analysis is predicted. The heat transfer for the single pebble layer will be calculated, and then it will be iterated for whole height of the pebble bed using MATLAB 6.5.
The three dimensional pebble bed air heater will be modeled using GAMBIT and analyzed in computational fluid dynamics software FLUENT 6.1.
CHAPTER 4 SOLUTION PROCEDURE
In the performance analysis pebble bed air heater is solved by the two techniques. In the first technique, the matlab programming is carried out to predict
the performance one dimensional pebble bed air heater. In the second one the three dimensional analysis is done in commercial available computational fluid dynamics analysis software FLUENT. The solution procedures for these two techniques are given below.
4.1 MATLAB OVERVIEW
Numerical mathematics, a branch of computational mathematics, is a crucial tool for both qualitative and quantitative analysis. This branch of mathematics is devoted to the design and analysis of methods for solving problems in science, engineering, economics, and operations research. Accordingly, it deals with the mathematical foundations of methods of solution, and analysis of the theoretical properties of the methods, such as speed, accuracy, stability, and complexity. Also, it practically demonstrates the performances of the methods using examples and counter-examples. Thus, this discipline lays equal emphasis on both the theoretical and experimental aspects of computational mathematics.
4.1.1 MATLAB MATLAB is an interactive system whose basic data element is an array that does not require dimensioning. This allows you to solve many technical computing problems, especially those with matrix and vector formulations, in a fraction of the time it would take to write a program in a scalar no interactive language such as C or FORTRAN.
The name MATLAB stands for matrix laboratory. MATLAB was originally written to provide easy access to matrix software developed by the LINPACK and EISPACK projects. Today, MATLAB engines incorporate the LAPACK and BLAS libraries, embedding the state of the art in software for matrix computation.MATLAB has evolved over a period of years with input from many users. In university environments, it is the standard instructional tool for introductory and advanced courses in mathematics, engineering, and science. In industry, MATLAB is the tool of choice for high-productivity research, development, and analysis.
MATLAB features a family of add-on application-specific solutions called toolboxes. Very important to most users of MATLAB, toolboxes allow you to learn and apply specialized technology. Toolboxes are comprehensive collections of MATLAB functions (M-files) that extend the MATLAB environment to solve particular classes of problems. Areas in which toolboxes are available include signal processing, control systems, neural networks, fuzzy logic, wavelets, simulation, and many others
4.2 THE MATLAB SYSTEM The MATLAB system consists of five main parts:
4.2.1 DEVELOPMENT ENVIRONMENT
This is the set of tools and facilities that help you use MATLAB functions and files. Many of these tools are graphical user interfaces. It includes the MATLAB desktop and Command Window, a command history, an editor and debugger, and browsers for viewing help, the workspace, files, and the search path.
4.2.2 THE MATLAB MATHEMATICAL FUNCTION LIBRARY This is a vast collection of computational algorithms ranging from elementary functions like sum, sine, cosine, and complex arithmetic, to more sophisticated functions like matrix inverse, matrix eigenvalues, Bessel functions, and fast Fourier transforms.
4.2.3 THE MATLAB LANGUAGE This is a high-level matrix/array language with control flow statements, functions, data structures, input/output, and object-oriented programming features. It allows both “programming in the small” to rapidly create quick and dirty throwaway programs, and “programming in the large” to create complete large and complex application programs.
MATLAB has extensive facilities for displaying vectors and matrices as graphs, as well as annotating and printing these graphs. It includes high-level functions for two-dimensional and three-dimensional data visualization, image processing, animation, and presentation graphics. It also includes low-level functions that allow you to fully customize the appearance of graphics as well as to build complete graphical user interfaces on your MATLAB applications.
This is a library that allows US to write C and FORTRAN programs that interact with MATLAB. It includes facilities for calling routines from MATLAB (dynamic linking), calling MATLAB as a computational engine, and for reading and writing MAT-files.
4.2.4 MATLAB PROCEDURE AND STEPS
In matlab the one dimensional analytical analysis will be predicated. In programming certain assumptions were made,
The pebbles will be packed uniformly There is no conduction between the pebbles. The conduction between the wall and the pebble are not considered.
It calculates the superficial velocity, Reynolds number, nusselt number, the outlet temperatures of a pebble, air and flue gas are calculated using lumped parameter analysis and the energy balance equation. These predictions are in transient state. These parameters will be calculated for the single layer in the bed. Then the outlet temperature of the first layer will be given as the inlet temperature for the second layer, so it should be carried out for each layer in the bed. We have two separate programming for heating and cooling phases. There are seven steps for this programming, which follows.
Initially the porosity, height to diameter ratio of the pebble bed, diameter of the pebbles, material properties of the pebbles, inlet temperatures of the flue gas and air, will be given as an input to the programming.
Step 2 By using the input variable it calculates the geometries of the bed and the properties of air and flue gas with respect to the temperature change.
Step 3 If the number of layers is more than the assigned variable N then, it will be move directly to the step 7.otherwise it moves the step 4.
Again a conditional statement will appear for reducing the errors, in this statement the difference of the inlet and outlet temperature will be calculated were it is assigned to a variable ’tmp’. And then it will be compared to the previous temperature. When the difference is same it move to the step7, otherwise it move to the next step 5.
Step 5 Here its calculates the dimensionless parameters and the temperatures for a layer. If the number of layers is less than the assign variable it moves to step3 otherwise it move the next step. The formulas used to calculate the dimensionless parameters are given below.
Superficial velocity, Uf = Reynolds number, Re = DP uf ρf / µf (1- ε). Nusselt number, NU = 0.8 Re 0.7 Pr 0.333 Nusselt number, NU = hDP /k. Lumped parameter analysis Tg_1- Tp_2 / Tg_1 –Tp_1 = EXP ((- h A /ρf V µf) * t). Energy lost by pebble (or) gas = Energy gained by the pebble (or) air m_ p*Cp_ p*(Tp_1-Tp_2) = t*m_ a*Cp_ a*(Ta_1-Ta_2). Step 6 It stores the temperature in the assigned variable “tmp”. And moves it to the step 4. m_ a /ρf Abed.
Step 7 In this step it plots the results for single layers in the bed. Then the outlet temperature will be assigned as inlet temperature to the next layer.
A flowchart is also presented for the matlab programming and the results in temperature’s of the pebbles
and air/gas the matlab has been discussed in the chapter 5.In this program the properties for
To start with, assume the initial
flue gas and air will be calculated by assigning the polynomial equations directly in the program.
And get other input variables Get the material properties of the pebble
4.3. Computational Fluid Dynamics
The application field for CFD as a modeling tool is constantly expanding. Improvements in computer hardware, computational speed and memory size, as heater. well as improvements in software capabilities make CFD a feasible and accessible tool in a large range of applications.
If no. of layers = < n Calculate the properties of the air/gas used in the air
Calculate the geometries of the air heater
If (tmptmp1) <= 0.001
Then assign Tmp = Ta_1cel –Ta_21cel/2
Calculate Uf, Re, Nu, h and temp’s For the single layer of pebble
Then assign Tmp1 = Ta_1cel – Ta_21cel/2
Calculates the final output temperature of the single layers.
B Plot the results for the both temperatures from obtain results.
Then assign the output temperature of fluid/gas in layer 1 to the initial temperature of fluid/gas to next layer
Fig.4.1.Flowchart for Matlab Computational Fluid Dynamics (CFD) is the science of predicting fluid flow, heat transfer, mass transfer, chemical reactions, and related phenomena by
solving the mathematical equations which govern these processes using a numerical process. For engineers CFD gives the safe hands on Unconventional Geometry • Complex duct networks, flow through unconventional shaped conduits, changing flow cross section, flow through Turbo machines, tube banks, river bed simulation etc. Simultaneous flow with heat & mass transfer • Heat exchangers with large temp changes, non-std configuration, diffusion of heat and mass, radiation etc. Multi-phase flows • Water-oil mixtures, ash laden flue gas, coal-air mix, aerosols, foams, mists, smoke, slurries. Flows with phase changes • Vaporization, condensation, boiling, flashing, cavitations, ablation etc. Flows with heat transfer & Combustion • Combustion of fuels, open and confined flames, smoke, soot formation, pollutants formation
Solutions in CFD are obtained by numerically solving a number of balances over a large number of control volumes or elements. The numerical solution is obtained by supplying boundary conditions to the model boundaries and iteration of an initially guessed solution. The balances, dealing with fluid flow, are based on
the Navier Stokes equations for conservation of mass (continuity) and momentum. These equations are modified per case to solve a specific problem.
The control volumes or elements, the mesh, are designed to fill a largescale geometry, described in a CAD file. The density of these elements in the overall geometry is determined by the user and affects the final solution. Too coarse a mesh will result in an oversimplified flow profile, possibly obscuring essential flow characteristics. Too fine a mesh will necessarily increase iteration time.
4.3.1 BASIC STEPS IN CFD
The following are the basic steps used in the CFD software for the analysis of the intake manifold model, 1. Preprocessing Phase 2. Solution Phase 3. Post processing Phase
22.214.171.124. PREPROCESSING PHASE Pre-processor consists of the input of a problem to a CFD program by means of an operator friendly interface and the subsequent transformation of this input into a form for use by the solver .The user activities at the pre-processing stage involve
a) Definition of geometry of the region of interest: the computational domain. b) Grid generation –the sub division of the domain into a number of smaller,
no overlapping domains: a grid (or mesh) of cells (or control volumes or elements)
c) Selection of the physical and chemical phenomena that need to be modeled d) Definition of fluid properties
e) Specification of appropriate boundary conditions at cells, which include with or touch the domain boundary.
A. Geometry Definition The preprocessing stages initially defines the geometry of the pebble bed air heater. In order to perform a computational analysis in an efficient manner, it is important to take advantage of symmetry and geometry simplification wherever possible. Heat transfer and pressure drop in the bed has to be analyzed. Due to uniform packing of pebbles in the bed, its cylindrical structure and time dependent flow patterns, the model for the pebble bed can take advantage of symmetry. This will decreases the time required for processing the calculation. For this pebble bed model we had created three volumes, and then it had been connected by using the connectivity option. The three volumes are bed, inlet pipe, outlet pipe. This has been shown in Fig 4.2.
Fig.4.2. Symmetric pebble bed air heater model
B. Gird Generation
One of the most important parts of CFD modeling is the construction of the mesh topology. The mesh establishes the accuracy of the simulation. It has to be chosen with enough detail to describe the processes accurately and with a degree of coarseness that enables solution within an acceptable amount of time. When an optimal density has been found, refining this will increase the model size without displaying more flow detail. When it is coarsened the mesh will obscure, possibly essential, parts of the flow detail. The mesh determines a large part of creating an acceptable simulation. As the governing equations form a coupled system of non-linear partial differential equations, it is not possible to find closed-form solutions for anything but the simplest geometries. Thus in order to be able to solve the equations, the domain must be broken down into smaller volumes or areas. The equations
themselves are then approximated over these volumes by a discretization process. There are three common types of discretization in use today: finite difference, finite volume and finite element.
The three methods differ in how the equations are discretized, and in how they are applied to the elements in the flow domain. The finite differences method calculates the variables of the flow equations at discrete points in the domain. These points are defined by the nodes that are at the corners of the elements in the mesh. The finite volumes method calculates the average values of the flow variables within each element. These values are then assigned to a central node in each element. Lastly, the finite elements method interpolates the variables within the elements. The order of interpolation determines how accurately the method approximates the continuous solutions to the governing equations. The enlarged finite volume meshed view of pebble bed air heater is shown in Fig.4.3.
For the meshing of the pebble bed air heater, it has been started with the face meshing. The face between the inlet pipe and the bed layer will be meshed with quadrilateral element and pave type mesh is used. Then the model is volume meshed.
Figure.4.3. Grid model of the heater
The bed volume is meshed with three dimensional hexahedron meshes and the type is copper mesh. The inlet and outlet pipe had been meshed using the tetrahedron element and the tgrid types have been used. Gambit software has been used for modeling and meshing.
APPLY BOUNDARY CONDITIONS The next step in the performance analysis is to determine the boundaries conditions. The boundary conditions (BC-s) was determined by project goals and accuracy required
According to this model. We had given mass flow inlet, pressure oulet, symmetry 1, symmetry 2, bed, and wall.These boundry conditions has been explained in the Fig 4.4
Inlet Mass flow inlet is specified as inlet boundary conditions. It is used define the fluid flow rate at the inlet. These boundary conditions are used when the flow rate is known. Mass flux will be given. The atmospheric pressure is applied for the pressure inlet boundary condition.
Outlet Pressure outlet is specified as outlet boundary conditions. This condition will specify the static pressure at the outlet boundary. The pressure 1 bar is applied at the outlet.
Symmetry We had been given symmetric conditions to the system. According to our model we can apply two symmetric conditions. Normally in the symmetric conditions it creates the negative pressure to the symmetric sides. It is important to take advantage of symmetry due to uniform packing of pebbles in the bed and its cylindrical structure, this will decreases the time required for processing the calculation.
Figure 4.4. Boundary Conditions of Pebble Bed Air Heater BED In this boundary conditions we had applied porous zone by specifying the direction-1 vector and the direction-2 vector. According to the porous media we have to specify the viscous and inertia resistance in this bed. With these resistances it calculates for the heat transfer and pressure drop in the packed bed.
In addition to the BC-s specified above, the continuum condition has to be specified for this model. By this continuum condition we are specifying the software that whether the volume is a solid (or) fluid. In this we specify the bed volume as a fluid, outer wall and the insulating wall as the solid. In Fig 4.4 Continuum Conditions has been given for the Pebble Bed Air Heater.
D. SETTING CFD PARAMETERS After assign the boundary and continuum conditions we had to export the model in .msh format. This .msh file can open directly in the fluent software. Before opening the fluent we had to decide whether it going to be a three dimensional (or) three dimensional double precision, we used the three dimensional double precision for the pebble bed model. Initial step in the fluent is to define the model.
The initial step is to decide the solver, for this model we had selected the segregated solver and this model is a time dependent model i.e. transient model. The energy equation has been used for this model because the heat transfer is involved in this performance analysis.
Then we have to define the flow in the viscous option, two-equation (k-ε) turbulence model is employed. In the preprocessor stage mention all the physical, chemical phenomena and also specify the fluid properties of the air heater. We have to defining the fluid properties such as fluid density, fluid viscosity, thermal conductivity and specific gravity as a input. Here we had to specify the air and the pebble properties to the cooling phase and in the heating phase we had specify the flue gas and pebble properties.
Figure 4.5. Continuum Conditions of Pebble Bed Air Heater.
126.96.36.199. Solution phase In the solution phase of the analysis, the computer takes over and solves the simultaneous set of equations. With the advent of high-speed digital computers and numerical technique in recent years there has been a major impact on the design process. The numerical solution techniques that form the basic of the solution procedure involved:
a. Approximation of the unknown flow variables by means of simple functions as Under-relaxation factor which plays a vital role in the numerical solutions. b. Discretization by substitution of the approximations into the flow equations and Subsequent mathematical manipulations. c. Solution of the algebraic equation.
188.8.131.52. Post processing phase Post processing means reviewing the results of an analysis. It is probably the most important step in the analysis, because it helps in understanding how the applied boundary conditions affect the design how the meshing accuracy affects the results. As in pre-processing a huge amount of development work has recently taken place in post-processing field. Owing to the increased popularity of engineering workstations many of which have outstanding graphics capabilities, the leading CFD packages are now equipped with versatile data visualization tools. In the heat transfer problem the CFD results to give various plots like temperature plot, velocity plot, pressure field and etc. CFD vector plot gives good informative picture of the flow field, which can help in the proper configuration and design of the Pebble bed air heater and animation of results help to find the flow paths. The Results of these two will be given in the next chapter.
CHAPTER 5 RESULT AND DISCUSSION
The work was carried out by using the computational fluid dynamics (FLUENT) and Matlab techniques. The Performance of the pebble bed air heater was predicted using the computational fluid dynamics software and the one dimensional heat transfer code was written using Matlab. The result obtained from the CFD analysis and Matlab are discussed here.
5.1.1. MATLAB. In the first technique Matlab code has been written for one dimensional heat transfer prediction of pebble bed air heater. The lumped parameter analysis and the energy balance equation have been used to predict the heat transfer. There are certain assumptions have been used to predict, which was explained in the chapter4.
S.NO 1 2
PHASE Heating Phase Cooling Phase
DURATION 3 hours 15 minutes
Table 5.1.1 Time Period in matlab
The coding has been written for cooling and the heating phase of the air heater. Here certain input variable has to given. The obtained was given in the table 5.1.1.
5.1.2. HEATING PHASE In the Heating phase the pebble will be heated by flue-gas. For this phase the input parameters are porosity, pressure of inlet air, diameter of the cylinder, diameter of the pebble, Mass flow rate of the air. The values are given below in the table 5.1.2
Porosity Pressure Diameter of the cylinder Diameter of the Pebble Mass flow rate of air Time taken
0.45 30 bar 3m 0.02 m 0.5 kg/sec Its kept variable
Table. 5.1.2 Input variables for the heating phase
According to the Results, bed will be heated within five hours duration. The beds have been heated from its initial condition. The results have been provided in the graph shown below in Fig.5.1.1.The graph shown the temperature profile of the pebbles in the bed.
5.1.3. COOLING PHASE In the cooling phase the pebble will be cooled by air. For this phase the input parameters are porosity, pressure of inlet air, diameter of the cylinder, diameter of the pebble, Mass flow rate of the air. The values are shown below table 5.1.3.
Fig .5.1.1 Results for 2 Hours in the matlab for temperature profile of pebble in heating phase
Porosity Pressure Diameter of the cylinder Diameter of the Pebble Mass flow rate of air Time taken
0.45 30 bar 3m 0.02 m 50 kg/sec Its kept variable
Table. 5.1.3 Input variables for the cooling phase
With respect to time the results will be displayed. Here the temperature in the pebble is assumed for every 100 layers. For the respective time given the temperature of the pebble and the temperature of the air will be given as the output in the coding. The results have been provided in the graph shown below in Fig.5.1.2.The graph shown the temperature profile of the pebbles in the bed.
Fig .5.1.2 Results for 12 minutes in the matlab for temperature profile of pebble in cooling phase
For 12 minutes the results has been taken because this is the optimum time for the heating cycle to begin,where if it operates above this time, then the required heat will not be a constant it suddenly varies.
5.2.1. COMPUTATIONAL FLUID DYNAMICS.
Fluent commercial available CFD software is used to predict the heat transfer in the second technique. Here the three dimensional analysis has been done. The analysis has been done for cooling and the heating phase of the air heater separately. Here the porous media concept is used. The result has been given in the table 5.2.1.
S.NO 1 2
PHASE Heating Phase Cooling Phase
DURATION 5 hours 10 minutes
Table 5.2.1 Time Period in Fluent
5.2.2. HEATING PHASE
In this phase, mass flux, inlet temperature for the flue gas and the pebbles, the properties for the flue gas and pebble are given. The Results is iterated for the unsteady state. The results had been discussed for first five hours.
Fig.5.2.1.Heating phase at different time steps
Time step at A=90 min
Time step at A=180 min
Time step at A=240 min
Time step at A=300 min
Fig.5.2.2Heating phase at different time steps A, B, C, D
The contour plot of the temperature had given in Fig 5.2.2 for the various time steps. Normally bed will be heated for the five hours to achieve the pebble temperature of 1400 c, if it is heated again then the temperature will rise gradually. But due to the conduction it will also lose some heat to wall and pebble.
The graph had been shown in fig.5.2.1 were it gives the result of whole system. Were the rise in the temperature profile can be noted.
5.2.2. COOLING PHASE
In this phase, mass flux, inlet temperature for the air and the pebbles, the properties for the air and pebble are given. The Results is iterated for the unsteady state. The results had been discussed for twelve minutes.
The contour plot of the temperature had given above in Fig 5.2.3 for the various time steps. Normally bed will be cooled for the twelve minutes to achieve the air of 1000 c, if it is cooled again then the temperature will rise gradually. But due to the conduction it will also lose some heat to wall and pebble.and also it will stared to heat the pebbles .
The graph had been shown in fig.5.2.4 were it gives the result of whole system. Were the rise in the temperature can be noted.
Time step at A=100 sec
Time step at B=200 sec
Time step at C=430 sec
Time step at D=610 sec
Fig.5.2.3.Cooling phase at different time steps A, B, C, D
Fig.5.2.4.cooling phase at different time steps
CHAPTER-6 CONCLUSION AND SCOPE FOR FUTURE WORK 6.1 CONCLUSION Performance analysis of a pebble bed air heater has been done by two technique i.e. 15 seconds of cooling and 2 hour of heating in initial stage of operation is recommend for this pebble bed heater . Using Matlab the one dimensional heat transfer code has been written with certain assumptions. In computational fluid dynamics technique, the three dimensional model has been modeled, meshed and analyzed using the porous media concept in transient state. And the performance was predicted for initial conditions of the pebble bed air heater.
6.2 SCOPE OF THE FUTURE WORK The matlab code can be developed by writing the two and three dimensional heat transfer prediction by assign the certain assumptions like conduction between pebbles, conduction between the wall and the pebbles. we can compare the results with the CFD results
In the CFD analysis we can try the conduction between the pebbles in the bed.
REFERENCES 1. Harris, C.K., Roekaerts, D., Rosendal, F.J.J.,Buitendjik, F.G.J., Daskopoulos, Ph., Vreenegoor, A.J.N., & Wang, H. (1996), Computational fluid dynamics for chemical reactor engineering, Chemical Engineering Science, 51,1569-1594. 2. Logtenberg, S.A. and Dixon A.G., (1998), computational fluid dynamics and studies of fixed bed heat transfer, Chemical Engineering and Processing, 37, 7-21. 3. Dixon, A.G. and van Dongeren, J.H. (1998), The Influence of the Tube and Particle Diameters at Constant Ratio on Heat Transfer in Packed Beds, Chemical Engineering Process., 37, 23. 4. Logtenberg, S.A., Nijemeisland, M., Dixon, A.G. (1999), Computational fluid dynamics of fluid flow and heat transfer at the wall-particle contact points in a fixed-bed reactor, 54, 2433-2439. 5. Fluent 6 User’s Guide, Fluent incorporated limited, 2001. 6. Bode, J. (1994) Computational fluid dynamics applications in the chemical industry.Computers & Chemical Engineering 18 SUPPL, S247-S251 7. Bey, O. and Eigenberger, G. (1997) Fluid flow through catalyst filled tubes. ChemicalEngineering Science 52, 1365-1376. 8. Carslaw, H.S. and Jaeger, J.C. (1959) Conduction of Heat in Solids. 2nd edition,University Press, Oxford. 9. Cheng, G.J., Yu, A.B. and Zulli, P. (1999) Evaluation of effective thermal conductivity from the structure of a packed bed. Chemical Engineering Science 54, 4199-4209. 10. Dalman, M.T., Merkin, J.H. and McGreavy, C. (1986) Fluid flow and heat transfer past two spheres in a cylindrical tube. Computers & Fluids 14, 267281. 11. Damköhler, G. (1937) Der Chemie Ingenieur, Eucken-Jacob Volume 3, Akadem.Verlag., Leipzig Daszkowski, T. (1991) Strömung, stoff- und
wärmetransport in schüttungsgefüllten rohrreaktoren. Ph.D. Thesis, Institute für Chemische Verfahrenstechnik, Universität Stuttgart. 12. Derkx, O.R. (1995) Wall heat transfer coefficient in a fixed bed reactor. M.S. ThesisWorcester Polytechnic Institute. 13. Derkx, O.R. and Dixon, A.G. (1996) Determination of the fixed bed wall heat transfer coefficient using computational fluid dynamics. Numerical Heat Transfer Part A 29,777-794. 14. Dixon, A.G. (1985) The length effect on packed bed effective heat transfer parameters.Chemical Engineering Journal 31, 163-173. 15. Verification Studies of Computational Fluid Dynamics in Fixed Bed Heat Transfer by ir. Michiel Nijemeisland
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.