According to a study Earth’s temperature has increased by 20 C in last 15 years. The major cause behind this problem is Green House gasses which causes greenhouse effect. With rapid growth of world’s industrialization, the levels of greenhouse gases being dumping into the atmosphere has increased which is making the problem worst. Industries are dumping waste heat into the atmosphere which is again increasing the temperature of atmosphere and at the same time useful energy is being dumped. This waste heat can be utilized which can solve both the problems i.e., firstly it will cool off the atmosphere and secondly it will decrease the energy input requirement of the industry since the waste heat is being utilized.
Table of Contents
INTRODUCTION
CHAPTER 1
REVIEW OF LITERATURE
CHAPTER 2
MATHEMATICAL & COMPUTATIONAL MODELING
CHAPTER 3
COMPUATIONAL SIMULATION ON ANSYS FLUENT
CHAPTER 4
BOOK AT A GLANCE
CHAPTER 5
LITERATURE CITED
INTRODUCTION
CHAPTER 1
According to a study Earth’s temperature has increased by 20C in last 15 years. The major cause behind this problem is Green House gasses which causes greenhouse effect. With rapid growth of world’s industrialization, the levels of greenhouse gases being dumping into the atmosphere has increased which is making the problem worst. Industries are dumping waste heat into the atmosphere which is again increasing the temperature of atmosphere and at the same time useful energy is being dumped. This waste heat can be utilized which can solve both the problems i.e., firstly it will cool off the atmosphere and secondly it will decrease the enegy input requirement of the industry since the waste heat is being utilized.
The present thesis is a step taken in field of waste heat utilization. It includes the design and simulation of thermal regenerator which is used to utilize the waste heat. Thermal regenerator is a type of heat exchanger which is used to transfer heat from one fluid to another with the help solids as an intermediate medium. To study the details of thermal regenerator one has to have a good understanding of heat exchangers.
1.1 Heat Exchangers
Heat Exchangers are the devices which are used to transfer heat from one fluid to another. Heat exchangers are classified on the basis of how the heat is transferred between the fluids.If there is direct contact between the two fluids then the heat exchangers are classified as direct contact heat exchangers and if there is no contact between the fluids they are classified as indirect contact heat exchangers. Further indirect contact heat exchangers are classified as Recuperators, storage type exchangers and fluidized bed heat exchangers. The complete classification of heat exchanger is shown in Fig. 1.1
This figure is not part of this excerpt.
Fig.1.1 Classification of Heat exchangers on the basis of heat transfer process
Panwar (2016)
1.2 Heat Regenerators
Thermal heat regenerator is a type of heat exchanger which is filled with solids (metals or ceramics) of different shapes called bed of regenerator that have high volumetric heat capacity i.e, it can absorb and store relatively large amounts of heat. The complete working of thermal regenerator consists of two cycles, namely heating cycle and cooling cycle. During the heating cycle hot gases that can be exhaust/flue gases of any manufacturing industry such as glass manufacturing industry is made to pass through the regenerator. The heat from the flue gases is transferred to the solids and flue gases at lower temperature exist from regenerator. After the completion of heating cycle cooling cycle starts with cold air entering the same regenerator bed. The solids now transfer the heat to the cold air and cold air gets heated up. On the other hand in recuperators, the fluid between which the heat has to be transferred is separated by a wall through which heat is transferred. In recuperators there is no mixing of fluids between which the heat is to be transferred. Thus the heat in storage type heat exchanger (thermal regenerator) is not transferred through the wall as in recuperators but it is stored and rejected by solids. Figure 1.2 shows a parallel heat regenerator which is required for continuous flow of heated air. The continuous flow can achieved only by parallel arrangement in which heating and cooling cycle are carried out simultaneously in different beds.
This figure is not part of this excerpt.
Fig. 1.2Fixed bed heat regeneratorWillmott (1969)
1.3 Heat Regenerator Types
In the previous section regenerator is defined as a heat storing heat exchanger, in which the heat transferred through solid particles referred as a matrix or bed of the regenerator. For a regenerator to work continuous, either moving bed is used through which air and flue gas can periodically move into and out, which is a Rotary Regenerator or the flows of gases is operated through valves to and from the fixed matrices/beds as shown in Fig. 1.2 which is a fixed bed regenerator.
The fixed bed heat regenerator is also sometimes referred to as a periodic-flow regenerator or swing regenerator.
1.4 Advantages of Regenerator
Heat regenerator has following major advantages:
· Regenerator is compact in design as compared to recuparator, which makes it lighter in weight and economical as compared to recuperator.
· The manufacturing cost of regenerator is substantially lower than recuperators used to transfer same amount of heat.
· Material cost is higher in recuperators than in regenerators.
· Since regenerator has different mode of operation, it does not require leak-proof core as required in recuperator.
· Depending on the application, regenerators are made of metals, ceramics, alloys, plastics, many more.
· Design of regenerator is simpler as compared to design of components in recuperators.
· In regenerator the matrix/bed surface has self-cleaning characteristics, which results in low fouling due to flue gases and also reduces corrosion as compared with recuperators.
All these advantages make heat regenerator superior over recuperators for gas-gas heat exchange applications requiring effectiveness exceeding 85%.
1.5 Applications of Regenerator
Regenerators have very large area of applications such as in metallurgical industries, glass manufacturing, air separation plants, storage of solar energy, incineration of VOC’s and many more.
1.6 Pressure drop and Drag coefficient in Fixed bed Regenerator
The pressure drop inside the regenerator in case of a fully developed flow is given by Ergun’s equation given by Ergun (1952). The limitation of this equation is that it holds good for large D/dp ratio (>15), where condition of uniformity in void fraction prevails.
The Ergun’s equation is:
This figure is not part of this excerpt.
The coefficients of Ergun’s equation in Eq. 1.1(150 and 1.75) are universally disputed and controversial. Hicks (1970) has also given an equation for pressure drop in fixed bed regenerator with spherical particle but the coefficients in his equation are not constant but are the function of Reynolds number. In another research conducted by Handly and Heggs(1968)it was found that Ergun’s equation was unable to predict the pressure in irregular packed bed regenerator. Another equation for pressure drop in fixed bed regenerator is given by MacDonald (1979) as:
This figure is not part of this excerpt.
All the above mentioned pressure equation are for fixed bed regenerator with D/d p greater than 15. Since the regenerators with D/d > 15 are considered having uniformity in void fraction in the bed, the flow complexities is very low in these cases.
The complete review of wall effects in regenerator done by Einsfeld and Schnitzlein(2001) concluded that correlation for pressure drop byReichelt(1972)is most promising one.
The detailed understanding of flow structure in spaces near the particles in these beds can only be gathered by highly sophisticated flow analysis tool like Ansys Fluent.
The one of the objectives of the present thesis is focused on studying the flow complexity with in the regenerator with different operating conditions and calculating the pressure drop and temperature variation for fixed bed regenerator with the help of CFD simulations.
1.7 Thermal characteristics of Fixed bed regenerator
The other major objective of the present thesis is to study the thermal characteristics of the regenerator and investigate the effects of various factors on the thermal characteristics of the fixed bed regenerator. In the present work various factors affecting the thermal characteristics of the regenerator such as, bed height/length, regenerator diameter, particle diameter, heat storage capacity, switching time, residence time, and gases flow direction are numerically investigated.
1.8 Transient CFD analysis of Fixed bed regenerator
The heat regenerator works on two cycles i.e. heating cycle and cooling cycle. During heating cycle the flue gases enters the regenerator and flows through the bed. The heat is absorbed by the packing material bed and the flue gases at lower temperature flow out from the regenerator.
In cooling cycle the air from opposite direction is made to enter the sane regenerator through the bed and heat is transferred from the packing materials to the air. The time for which the flues gases are remained in the regenerator is known as thermal residence time of flue gas. Thermal residence time has a great effect on the effectiveness of the regenerator.
The time at which the heating cycle is switched with cooling cycle is called Switching time. Switching time is very important factor in regenerator working, only by properly adjusting the switching time one can improve the efficiency of the regenerator. For the study of temperature variation throughout the regenerator bed length, the regenerator is modeled in the Ansys design modular and after proper meshing of the model in GAMBIT it is imported in Ansys Fluent, where the transient simulation with time step of 10e-3 and switching time or cycle time of 1 minute is carried out for regenerator with D/d p 3, 8, 12. This transient CFD simulation of flue gases in heating cycle and ambient air in cooling cycle is continued till the steady state in temperature flow is reached.
REVIEW OF LITERATURE
CHAPTER 2
A vast literature is available in area of design and performance analysis of fixed-bed regenerator. Different researchers have done theoretical and experimental and simulation work in this context as given below.
P. C. Carman (1937)conducted experiments on pebble bed regenerators to calculated the pressure drop along the bed and wall effect by calculating drag coefficient through pebble bed under creeping flow conditions. The column to particle diameter ratio (D/ dp) for the experiment was kept less than 15. Under these conditions, a significant deviation in pressure drop was observed as compared to Ergun’s equation. It was concluded that Ergun equation does not holds good for D/d p less than 15.
M. Leva(1947) conducted experiments to calculate pressure drop and drag through packed bed for turbulent flow condition (Reynolds number greater than 1000) and for D/dp ratio 2.67, 5.32 and 9.91. It was concluded from the experimental results that for D/d p greater than 15 pressure drop follows the Ergun’s equation but if D/d p ratio is less than 15, the pressure drop decreases with decreasing D/d p ratio for any Reynolds number. Similarly J. M. Coulson 1949 conducted experiment tostudy wall effect under creeping flow condition for D/d p 6.2, 8, 12.8 in fixed bed regenerator.
S. Ergun etal ,(1952)conducted experiment for the case of fully developed flow in a fixed bed and proposed semi empirical correlation by linking Kozeny-Carman equation for the creeping flow regime and Burke-Plummer equation for turbulent regime. The equationis accepted widely to calculate the pressure drop in the thermal regenerator with its two coefficients, 150 and 1.75
D. Mehta and M.C. Hawley(1969)studied the wall effect on pressure drop in packed bed having 7< D/dp<91 and Reynolds number less than 10.
W. Reichletet al,(1972)investigated the wall effect and drag coefficient in creeping, transition and turbulent regime with in a regenerator.
Macdonald et al, (1979)has provided an expression for regenerator having the D/dpratio greater than 15 to calculate the pressure drop in the thermal regenerator with its two coefficients, 180 and 1.8
O. Levenspiel (1983)investigated behavior of long packed bed heat regenerator in terms of simple axial dispersion model and analyzed the spreading of temperature front in fixed bed regenerator.
Dalmanet al.(1986) were one of the first researchers who used CFD to study the regenerator. They developed a two dimensional CFD model and solved it using finite element technique. The analysis was conducted for a range of Reynolds number up to 200 for regenerator with spherical particles. The study explained the flow pattern inside the regenerator.
C. F. Chu, K.M. Ng (1989)made a similar study by conducting experiments in fixed bed having D/d p ratio 2.9 to 24 and Reynolds number less than 5.
E. A. Foumenyet al.(1993)conducted experiments on pebble bed regenerators to calculate pressure drop along the bed and examine the wall effect by calculating drag coefficient through pebble bed under turbulent flow conditions with D/d p ratio in range of 3 to 24. A study with 8 spheres in a linear array inside a regenerator using FIDAP was done by
Lloyd and Boehm (1994). The simulation helped to explain the flow and heat transfer through the regenerator bed. They conducted the simulation for a range of Reynolds number of 40, 80, 120 and Prandtl number ranging from 0.73 to 7.3.
Derxet al.(1996) were the first researchers to study three dimensional CFD model on a regenerator with spherical particles in a linear array.
Logtenberg and Dixon (1998)with the help of CFD codes in FOTRAN modeled a regenerator bed with 8 spherical particles in two layers with particles perpendicular to the flow direction. The simulation was conducted for Reynolds number ranging 9 to 1450. A similar work with higher number of spherical particles was presented by Logtenberget al.(1999) . He modeled 10 spheres with aspect ratio 2.43 with the help of FEM. The study was focused to observe and predict the flow pattern and heat transfer inside the regenerator.
H.P.A. Caliset al.(2001)simulated the pressure drop and drag coefficient in square channels with particles, having D/d p ratio in range 1-2.A quantitative comparison between experimental results and CFD predictions were presented by MichielNifemeisland and Dixon (2001). They simulated a regenerator model using CFD having 44 spherical particles. The D/d p ratio was kept 2 to maintain the uniformity in void fraction. The simulation was focused to predict the velocity variation and temperature variation especially at wall-particle interface. The simulated predictions were compared with experimental setup of similar geometry.
Calliset al.(2001)used CFX CFD software to simulate the pressure drop and drag coefficient in a square regenerator channel. The hydraulic diameter of regenerator to diameter of particle ratio ( D/dp) for the simulation was ranged from 1 to 2. The number particles were also increased upto 40 particles than the previous studies. The simulation done by them was validated with experimental results calculated used Laser Doppler Aneometer. The limitation of the study was its inability to explain wall effects in low D/d p ratio.
Y. Juan et al. (2003)investigated performances of ball packed regenerator experimentally and numerically and gave revised Ergun’s equation with its two coefficients as 203 and 1.95. Further thermal characteristics of the regenerator were studied for one dimension, transient numerical model. The effect of variation of fuel on exit flue gas temperature and the heat recovery was also studied.
P. M. Park et al.(2003)studied unsteady thermal flow of regenerator with spherical particles using one dimensional, two phase fluid dynamics model. The effect of inlet velocity of exhaust gas and air, configuration of regenerator and diameter of regenerative particles on thermal flow in heat regenerator was investigated in his work.
L. Zong et al. 2004developed three dimensional transient mathematical model for heat transfer in regenerator matrix.
M.T. Zarrinehkafsh and S.M. Sadrameli(2004) developed a mathematical model to investigate the performance of fixed bed regenerator which accommodates for the convection and conduction heat transfer inside ceramic balls. The investigation considered both approaches in the analysis of the experimental data obtained for a regenerator packed with aluminium balls.
R. D. Feliceand L. Gibilaro (2004)proposed the model for predicting wall effect on pressure drop for low D/d p ratio fixed bed.The work of Derxet al.(1996)was extended by Nifemeisland and Dixson(2004) using similar 3 dimensional CFD model using commercial Fluent software. The simulation was conducted for fixed bed regenerators with 72 spherical particles which is higher than used by Derxet al.(1996) and D/d p ratio was kept to be constant as 4. The flow complexities and temperature variation along the regenerator was studied by them near the ball/particle and wall vicinity.
A comparative study of performance of five different turbulent models of fixed bed regenerator having 44 homogenous spheres as storage material was given byGuardoet al.(2005).The analysis was done using Fluent 6.0 with 44 particles placed in the bed in fixed order while in conventional beds the particles are randomly distributed throughout the bed.
The air was taken as working fluid and variation of pressure drop, velocity, and temperature was predicted within the regenerator.A. Guardoet al.(2005)investigated the wall to fluid heat transfer and pressure drop for the case of D/d p = 3.92 and over a Reynolds number range 100-1000.
N. Rafidi and W. Blasiak (2005)investigated thethermal performance analysis on a two composite material honeycomb heat regenerators used for HiTAC burners.
Reddy and Joshi (2008) presented CFD analysis of regenerator with column to particle diameter ratio as 5. The wall effect on the flow and heat transfer was studied in the paper.
Reddy and Joshi (2010) extendedtheir previous work by taking three different D/d p 3, 5, 10 for a range of Reynolds number 0.1 – 10,000 with water as the fluid flowing through the regenerator. In their simulation they modeled 3 different regenerators with approximately same voidage (0.44) and column diameter as 76.3 mm, 127 mm and 254 mm. Number of particles with in the regenerator were taken as 55 particles in 8 layers, 151 particles in 8 layers and 1120 particles in 15 layers respectively. The height of bed was 179.2 mm, 177.3 mm and 329 mm respectively. The simulation was focused to predict the pressure drop and drag coefficient for all three regenerators with D/dp= 3, 5, 10. The pressure drop results calculated by CFD were compared with Ergun’s equation and variation was studied. It was observed that for Reynolds number 0.1 to 100, the pressure drop were overestimated and for 100 to 10,000 the pressure drop was under estimated. The simulated CFD pressure drop result was compared with experimental results of Recheltet al.(1972). The comparative study showed that for Reynolds number 0.1 to 100 the CFD results are comparable with experimental results of Recheltet al.(1972). Further they also calculated the drag coefficient for all three regenerators with D/d p = 3, 5, 10 for the same range of Reynolds number. The results of drag coefficient calculated by CFD were further compared with experimental results of Coulson (1949), Mehta & Hawley (1969), Chu and Ng (1989) for the rage of Reynolds number 0.1 to 100 and Leva(1947), Foumenyet al.(1993) for the Reynolds number 100 to 10000. The results were found to be in good agreement for Renolds number 0.1 to 100 and a deviation of 10 % was found in results for Reynolds number range from 100 to 10000. The cause of deviation was the unstructured tetrahedral meshing in the interstitial voids between particles, where very high flow velocities occurs.
R. K. Reddy and J. B. Joshi (2010)simulated fixed beds having column to particle diameter 5 for creeping, transition and turbulent flow regime.
P. Pinelet al.(2011)reviewed various methods available for the seasonal storage of thermal energy in residential applications.
Another CFD analysis of regenerator of a thermo caustic engine was done by Davidet al.(2013). They considered two modes of heat transfer mainly convection and radiation within the regenerator. Ansys Fluent CFD tool was used for the analysis. The important findings of the paper included the effect of convection and its effects on amplitude and vibration of oscillating flow and effect of radiation on the performance of regenerator.
Z. Wen et al. (2014)developed a three dimensional model for regenerator and studied interaction between fluid flow and inter phase heat transfer within the regenerator. Detailed information about fluid flow and heat transfer within regenerator was presented with geometrical optimization of regenerator.
MATHEMATICAL & COMPUTATIONAL MODELING
CHAPTER 3
The present chapter presents a mathematical model of the fixed bed regenerator and CFD analysis using Ansys Fluent software.
3.1 Mathematical Modeling of Fixed bed regenerator
Fixed bed regenerator can have large and small particles. Mathematical model (Dispersion Model) of Fixed-bed with small particles accounts for:
• Effect of Intra-particle Conduction:
Temperature of solid particle is no longer assumed to be uniform, and depends upon convective heat transfer coefficient and intra-particle conduction.
• Dispersion Effect:
Mixing of fluid and axial molecular conduction of heat within fluid due to eddy generation. The mathematical model for fixed bed regenerator is given below.
The energy balance equation for the gas, is
This figure is not part of this excerpt.
The energy balance for solid particles is,
This figure is not part of this excerpt.
The initial and boundary conditions are:
This figure is not part of this excerpt.
3.2 CFD Analysis
Computational Fluid Dynamics (CFD) is used to solve the complex problems in fluid mechanics and heat transfer. In cases where the geometries are very complex to predict the flow and temperature distribution CFD have been proved to be a very handy, useful and efficient.
In experimental work lot of labor is required to decide the appropriate condition for stratification, which can be easily done with the help of CFD for the regenerators.
Ansys fluent is a state of art computer software which is used to model the heat transfer and fluid flow process complex engineering situations.
3.3 Computational Model
The computational model for the current study was made in Ansys-Design Modular and meshing was done in ICEM of Ansys Fluent. 5499 hexahedral cells comprising 6160 nodes were used for meshing to capture the effect of stratification accurately within the regenerator. ICEM view of different mesh is shown in Fig. 3.1.
This figure is not part of this excerpt.
Fig. 3.1 ICEM view of surface meshing
Table 3.1 Geometrical details of Computational Model
This figure is not part of this excerpt.
3.3.1 Skewness Factor
The quality of mesh is very important in the CFD for getting the accurate results. To measure the quality of the mesh skewness is one of the important factors. Skewness of a grid is an indicator of quality and suitability.
This figure is not part of this excerpt.
For hexahedral cells, skewness should not exceed 0.85. The worst value of skewness factor in the present geometrical model was 0.41 and the average skewness of the geometric model was 0.16. The detail of skewness factors for computational model is given in Table 3.2.
Table 3.2 Skewness factor for regenerator bed modeled geometries
This figure is not part of this excerpt.
3.3.2 Aspect Ratio
For measuring the quality of meshing aspect ratio of mesh should also be calculated. Aspect ratio is defined as the ratio of longest to shortest side in a cell. Ideally it should be 1. Table 3.3 shows the details of computational model of fixed-bed regenerator.
Table 3.3 Aspect ratio for regenerator bed modeled geometries
This figure is not part of this excerpt.
3.4 Governing Equations
The governing equations with which the physical phenomenon of flow through porous medium can be governed are as follows:
This figure is not part of this excerpt.
The viscous loss coefficient and the inertial loss coefficients of the porous zone are calculated with the help of following equations:
This figure is not part of this excerpt.
3.5 Boundary Conditions
Appropriate boundary conditions applied were defined as follows:
· Inlet velocity and inlet temperature of flue gas and air were specified at the inlet zone for computational model as Re= 0.1 and 473 K for flue gas and 300 K for air.
· Zero gauge pressure was given at the outlet boundary.
· Table 3.4 shows the details of viscous resistance and inertial resistance for creating porosity in the computational model of fixed-bed regenerator.
· Fluid porosity was kept 0.4 for the regenerator.
Table 3.4 Details of viscous and inertial resistances used for CFD simulations
This figure is not part of this excerpt.
3.6 Solution Method
In the fluent solver setting the velocity formulation was kept absolute, space was 3D and for unsteady simulations transient option was selected. Fig. 3.2 shows the details of solver settings. Option of energy equation was chosen to get temperature field and porous formulation was chosen in cell zone condition.
This figure is not part of this excerpt.
Fig. 3.2 Details of solver settings
Second order upwind scheme was used for momentum equation. Pressure was solved using simple pressure interpolation. First order implicit scheme was used for transient formulation. Residuals for the continuity, velocity were set 10-3, Residual for the energy were set to convergence up to 10-6. Details of solution method are shown in Fig. 3.3.
This figure is not part of this excerpt.
Fig. 3.3 Details of solution method used for CFD simulation
COMPUATIONAL SIMULATION ON ANSYS FLUENT
CHAPTER 4
The present chapter shows the simulated results for creeping flow of flue gases and air in both heating and cooling cycle of computational model of a fixed-bed regenerator. After the meshing the computational domain was divided into 5499 hexahedral elements with 6160 nodes. The captures the effects of heat and fluid flow in 3 D, Ansys fluent software was used and results were studied once the conversation criteria was reached. Governing equations combined with initial and boundary conditions are discretized in implicit form using the control volume method and solved using SIMPLE-based approach with k-ε turbulence model.
After the CFD simulation results have been presented in the chapter for the cases:
· Temperature variation of flue gases along the regenerator height.
· Pressure contour variation of flue gases along the regenerator height.
· Effect of flue gas exit temperature and regenerator height on its effectiveness.
· Transient CFD analysis of regenerator with D/dp=10.
4.1 Analysis of Thermal Characteristics of Fixed-bed Regenerator
The computation model of regenerator is simulated in fluent software for investigating its thermal characteristics. Flue gas was made to flow through the regenerator with a uniform porosity of 0.4 and regenerator diameter of 0.20 m and the height of 1 m. During heating cycle flue gas enters the regenerator at 473 K which is followed by air at 300 K during the cooling cycle. This cycle goes on till the steady state is achieved.
Following assumptions were made to carry out the CFD simulation:
- Intra-particle conduction is neglected.
- The effect of radiation is neglected.
- The regenerator wall is insulated.
4.2 CFD Analysis of Fixed-bed Regenerator
For transient CFD simulation the regenerator was modelled in Ansys fluent. The simulation is started by making the flue gases at high temperature (473 K) to enter the regenerator bed for cycle time of 60 sec. Once the heating cycle is over the cooling cycle starts with ambient at 300 K entering the regenerator in counter flow direction for the cycle time of 60 sec. This alternate flow of flue gases in heating cycle and ambient air in cooling cycle is continued till the steady state in temperature flow is reached.
Fig. 4.1 to Fig. 4.39 shows the temperature and pressure contours along the regenerator bed length at plane y = 0 respectively each after 60 seconds so that a clear view about the temperature variation along the bed can be made.
During the heating cycle the temperature at inlet flue gases entering the regenerator bed was decreases because of heat energy being absorbed by the solids and for cooling cycle the temperature was raised due to heat energy being released by the solids and transferred to the ambient air.
Table 4.1 Details of Geometrical models of Regenerator for CFD Analysis
This figure is not part of this excerpt.
Fig. 4.1 Temperature variation along regenerator length at plane y=0 for heating cycle at t= 1 min
This figure is not part of this excerpt.
Fig. 4.2 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 1 min
This figure is not part of this excerpt.
Fig. 4.3 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 2 min
This figure is not part of this excerpt.
Fig. 4.4 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 2 min
This figure is not part of this excerpt.
Fig. 4.5 Temperature variation along regenerator length at plane y=0 for heating cycle at t=3 min
This figure is not part of this excerpt.
Fig. 4.6 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 3 min
This figure is not part of this excerpt.
Fig. 4.7 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 4 min
This figure is not part of this excerpt.
Fig. 4.8 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 4 min
This figure is not part of this excerpt.
Fig. 4.9 Temperature variation along regenerator length at plane y=0 for heating cycle at t= 5 min
This figure is not part of this excerpt.
Fig. 4.10 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 5 min
This figure is not part of this excerpt.
Fig. 4.11 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t=6 min
This figure is not part of this excerpt.
Fig. 4.12 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t=6 min
This figure is not part of this excerpt.
Fig. 4.13 Temperature variation along regenerator length at plane y=0 for heating cycle at t= 7 min
This figure is not part of this excerpt.
Fig. 4.14 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 7 min
This figure is not part of this excerpt.
Fig. 4.15 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 8 min
This figure is not part of this excerpt.
Fig. 4.16 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 8 min
This figure is not part of this excerpt.
Fig. 4.17 Temperature variation along regenerator length at plane y=0 for heating cycle at t= 9 min
This figure is not part of this excerpt.
Fig. 4.18 Pressure variation along regenerator length at plane y=0 for heating cycle at t=9 min
This figure is not part of this excerpt.
Fig. 4.19 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 10 min
This figure is not part of this excerpt.
Fig. 4.20 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 10 min
This figure is not part of this excerpt.
Fig. 4.21 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 11 min
This figure is not part of this excerpt.
Fig. 4.22 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 11 min
This figure is not part of this excerpt.
Fig. 4.23 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 12 min
This figure is not part of this excerpt.
Fig. 4.24 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 12 min
This figure is not part of this excerpt.
Fig. 4.25 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 13 min
This figure is not part of this excerpt.
Fig. 4.26 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 13 min
This figure is not part of this excerpt.
Fig. 4.27 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 14 min
This figure is not part of this excerpt.
Fig. 4.28 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 14 min
This figure is not part of this excerpt.
Fig. 4.29 Temperature variation along regenerator length at plane y=0 for heating cycle at t= 15 min
This figure is not part of this excerpt.
Fig. 4.30 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 15 min
This figure is not part of this excerpt.
Fig. 4.31 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 16 min
This figure is not part of this excerpt.
Fig. 4.32 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 16 min
This figure is not part of this excerpt.
Fig. 4.33 Temperature variation along regenerator length at plane y=0 for heating cycle at t= 17 min
This figure is not part of this excerpt.
Fig. 4.34 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 17 min
This figure is not part of this excerpt.
Fig. 4.35 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 18 min
This figure is not part of this excerpt.
Fig. 4.36 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 18 min
This figure is not part of this excerpt.
Fig. 4.37 Temperature variation along regenerator length at plane y=0 for heating cycle at t= 19 min
This figure is not part of this excerpt.
Fig. 4.38 Pressure variation along regenerator length at plane y=0 for heating cycle at t= 19 min
This figure is not part of this excerpt.
Fig. 4.39 Temperature variation along regenerator length at plane y=0 for Cooling cycle at t= 20 min
This figure is not part of this excerpt.
Fig. 4.40 Pressure variation along regenerator length at plane y=0 for Cooling cycle at t= 20 min
This figure is not part of this excerpt.
Fig. 4.41 Velocity Vector along Regenerator Length at t=1min
This figure is not part of this excerpt.
Fig. 4.42 Velocity Vector along Regenerator Length at t=2min
This figure is not part of this excerpt.
Fig. 4.43 Velocity Vector along Regenerator Length at t=3min
This figure is not part of this excerpt.
Fig. 4.44 Velocity Vector along Regenerator Length at t=4min
This figure is not part of this excerpt.
Fig. 4.45 Velocity Vector along Regenerator Length at t=5min
This figure is not part of this excerpt.
Fig. 4.46 Velocity Vector along Regenerator Length at t=6min
This figure is not part of this excerpt.
Fig. 4.47 Velocity Vector along Regenerator Length at t=7min
This figure is not part of this excerpt.
Fig. 4.48 Velocity Vector along Regenerator Length at t=8min
Further, it is seen from the Fig. 4.38 at t=19 min that there are linear temperature profile without steep gradient and the thermal flow reaches the steady state i.e. heat absorption of solids is equal to heat released to air. At t= 20 min the ambient air is preheated to 472.7 K from which the thermal efficiency can be calculated by the given formula
η = (Ta,o- Ta,i)/(Tg,i- Ta,i) (4.4)
where, T a,o is the temperature of air at outlet (i.e. 472.7 K), T a,i is the temperature of ambient air at inlet (i.e. 300 K) and Tg,iis temperature of flue gas at inlet (i.e. 473 K) which comes out as 99%.
This figure is not part of this excerpt.
Fig 4.49 Variation of temperature with regenerator length during heating cycle
Figure 4.48 and Fig. 4.49shows the variation of simulated temperature of Flue gas and air during heating and cooling cycle. During the heating cycle the temperature of the flue gas is transferred to the solids in the regenerator-bed due to which the temperature of flus gases decreases as it passes through the regenerator. After 60 sec heating cycle is discontinued and cooling cycle starts with air entering the regenerator at 300 K from counter flow direction. The effectiveness of the regenerator can be calculated using exit air temperature by the help of Eq 4.4.
This figure is not part of this excerpt.
Fig. 4.50 Variation of temperature with regenerator length during cooling cycle
4.3 Variation of Effectiveness of Fixed-bed Regenerator With its length
The comparative study of effectiveness of regenerator and exit flue gas temperature shows that as the regenerator length increases the exit flue gas temperature decreases as shown in Fig. 4.51 which is due to the fact that as the flue gas flows through the regenerator the heat from flue gas is transferred to the solids due to which the exit flue gas temperature decreases.
This figure is not part of this excerpt.
Fig. 4.51 Variation of effectiveness of regenerator with regenerator length
It means for higher length the thermal mean residence time of flue gas increases which means the flue gas remains in the regenerator for longer period hence more heat is transferred to the solids. With decrease in exit flue gas temperature the effective of the fixed-bed regenerator increases as shown in Fig. 4.50.
This figure is not part of this excerpt.
Fig. 4.52 Variation of Exit flue gas temperature with regenerator length
But is seen from the same figure 4.50 that after a certain length (0.62 m) the thermal efficiency of the regenerator remains constant i.e., it does not increases with regenerator length. Hence it is very important for the designer to have an optimal regenerator length such that the thermal efficiency is highest and at the same time material cost is also low or optimum.
BOOK AT A GLANCE
CHAPTER 5
In the present book Regenerators are studied as an effective heat recovery system being used in industries to recover heat from flue gases.
The thermal characteristics of regenerator were studied and effect of various factors such as regenerator height, D/d p ratio, and porosity was studied. With the help of Ansys Fluent, contours of pressure and temperature were studied for regenerator with different length.
Regenerator was modelled in Fluent and simulations were carried out to study the variation of temperature along the regenerator length. The exit flue gas temperature was calculated from the results of simulations and with the help of it the effectiveness of regenerator was calculated as a function of exit flue gas temperature. The relationship between exit flue gas temperature and bed height showed that for longer regenerator the exit flue gas temperature was low. The effectiveness of regenerator was calculated and was found varying from 20 % to 90% as the height of regenerator bed varies from 0.2 m to 1 m. The reason behind this increase in percentage is higher residence time for flue gas in regenerator with larger heights.
At last the transient CFD model of regenerator with D/d p 10 was modelled and solved in fluent. The simulation results of each time cycle (heating and cooling cycle) were analysed and effectiveness for each regenerator was calculated. This transient model of regenerator can be used for other applications of thermal regenerators.
LITERATURE CITED
· Calis, H.P.A. Nijenhuis, J. Paikert, B.C. Dautzenberg, F.M.. Van den Bleek, C.M.. 2001. CFD modeling and experimental validation of pressure drop and flow profile in a novel structured catalytic reactor packing. Chemical EngineeringScience.56: 1713–1720.
· Carman, P.C. 1937. Fluid flow through granular beds.Transactions of the Institutionof Chemical Engineers. 15: 150–166.
· Chu, C.F. and Ng, K.M. 1989. Flow in packed bed tubes with small tube to particlediameter ratio.AIChE Journal. 35: 148–158.
· Coulson, J.M. 1949. The flow of fluids through granular beds: Effect of particleshape and voids in streamline flow. Transactions of the Institution of ChemicalEngineers. 27: 237–257.
· Dalman, M. .T. Merkin, J.H..McGreavy, C.. 1986. Fluid flow and heat transfer past two spheres in a cylindrical tube. Computers & Fluids 14: 267–281.
· Derx, O. R. and Dixon, A. G. 1996. Determination of the fixed bed wall heat transfer coefficient using computational fluid dynamics. Heat Transfer Part A. 2: 777 – 749.
· Ergun, S. 1952. Fluid flow through packed columns. Chemical Engineering Progress .48: 89–94.
· Foumeny, E. A, Benyahia, F, Castro, J. A. A, Moallemi, H. A, Roshani, S. 1993.Correlations of pressure drop in packed beds taking into account the effect ofconfining wall. International Journal of Heat and Mass Transfer . 36 : 536–540.
· Guardo, A., Coussirat, M., Larrayoz, M.A., Recasens, F., Egusquiza, E., 2005. Influence of the turbulence modeling of wall to fluid heat transfer in packed beds. Chemical Engineering Science. 60: 1733-1742.
· Juan, Yu.,ZangMingechusan., Fan Weiotong., Zhou Yuegui, and Zhao Guofeng.2002. Study on performance of the ball packed-bed regenerator.Applied Thermal Engineering. 22: 641-657.
· Levenspiel, O.1983. Design of Long heat regenerators by use of dispersion model. Chemical Engineering Science. 38: No. 12. 2035-2045.
· Leva, M. 1947.Pressure drop through packed tubes. Part I. A general correlation.Chemical Engineering Progress. 43: 549–554.
· Liu, Y., Tao, S., Liu, X.,Wen, Z.. 2014. Three dimensional analysis of gas flow and heat transfer in a regenerator with alumina balls. Applied Thermal Engineering. 69: 113 - 122.
· Lloyd, B and Boehm, R.. 1994.Flow and heat transfer around a linear array of spheres.Numerical Heat Transfer. Part A: Applications. 26: 237–252.
· Logtenberg, S. A. and Dixon, A. G.. 1998. Computational fluid dynamics of fixed
bed heat transfer. Chemical Engineering Process. 37: 7 – 21.
· Mac Donald, T. F., El-Sayer, M. S., Mow. K., Dullen, F. A. L. 1979. Flow
through porous media: The Ergun equation revisited .Industrial and Engineering
Chemistry Fundamentals.18: 199–208.
· Mehta, D. and Hawley, M.C. 1969. Wall effect in packed columns. Industrial andEngineering Chemistry. Process Design and Development . 8: 280–282.
· Nijemeisland, M. and Dixon A. G. 2004. “CFD study of fluidflowand wall heat transferin a fixed bed of spheres”. A.I.Ch.E. Journal. 50: 906–921.
· Poo Park Min., Cho Chang Han., Shin Dong Hyun. 2003. Unsteady thermal flow analysis in a heat regenerator with spherical particles. Int. J. of Energy Research. 27: 161-172.
· Rafidi Nabil and BlasiakMlodzimierz. 2005. Thermal performance analysis on a two composite material honey comb heat regenerator used for HiTAC burners. J. Applied Thermal Engineering.25: 2966-2982.
· Reichelt,W.1972.Zur Berechnung des DruckverlusteseinphasigdurchstroKmterKugel- und ZylinderschuKttungen. Chemie-Ingenieur-Technik. 44: 1068–1071.
· Reddy, R. K and Joshi, J.B. 2010. CFD modeling of pressure drop and drag coefficient in fixed beds: wall effects. Particuology. 8: No. 1. 37- 43.
· Reddy, R. K and Joshi, J.B. 2008.CFD modeling of pressure drop and drag coefficientin fixed and expanded beds. Chemical Engineering Research and Design. 86:444–453.
· Willmott, A. J. 1969. The regenerative heat exchanger computer representation. Int. J. of Heat and Mass Transfer. 12: 997-1014.
· Zarrirelkafsh, M.T. and Sadarameli, S.M.. 2004. Simulation of fixed-bed regenerative heat exchangers for flue gas heat recovery. Applied Thermal Engineering . 24: 373-382.
Frequently Asked Questions About Regenerators
What is the main focus of the provided language preview on regenerators?
The language preview focuses on the design and simulation of a thermal regenerator for waste heat utilization. It covers aspects like heat exchanger types, regenerator types, their advantages, applications, pressure drop, thermal characteristics, and transient CFD analysis.
What are the key chapters covered in this language preview?
The key chapters include an Introduction, a Review of Literature, Mathematical & Computational Modeling, Computational Simulation on ANSYS Fluent, a Book at a Glance, and a Literature Cited section.
What are heat exchangers, and how are they classified?
Heat exchangers are devices used to transfer heat from one fluid to another. They are classified as direct contact (fluids directly interact) or indirect contact (fluids are separated). Indirect contact exchangers are further classified as Recuperators, storage type exchangers (like thermal regenerators), and fluidized bed heat exchangers.
What are heat regenerators, and how do they work?
Thermal heat regenerators are heat exchangers filled with solids (metals or ceramics) with high volumetric heat capacity. They work in two cycles: a heating cycle where hot gases transfer heat to the solids, and a cooling cycle where cold air passes through the heated solids, absorbing the heat.
What are the different types of heat regenerators?
There are two main types: Rotary Regenerators, which use a moving bed, and Fixed Bed Regenerators, where gas flows are controlled by valves through fixed matrices/beds.
What are the advantages of using a regenerator?
Regenerators are compact, lighter, and more economical compared to recuperators. They have lower manufacturing costs, lower material costs, don't require leak-proof cores, and offer self-cleaning characteristics, resulting in low fouling and reduced corrosion.
What are the applications of heat regenerators?
Regenerators have broad applications in metallurgical industries, glass manufacturing, air separation plants, solar energy storage, incineration of VOCs, and more.
What is Ergun's equation, and why is it important in regenerator design?
Ergun's equation is used to calculate pressure drop inside the regenerator for fully developed flow. However, its limitation is that it is valid for large D/dp ratios (>15).
What thermal characteristics of fixed-bed regenerators are being studied?
Factors affecting the thermal characteristics are being studied, such as bed height/length, regenerator diameter, particle diameter, heat storage capacity, switching time, residence time, and gas flow direction.
What is transient CFD analysis, and why is it used in studying regenerators?
Transient CFD analysis is used to study the temperature variation throughout the regenerator bed length during the heating and cooling cycles. This is done using software like Ansys Fluent to model and simulate the heat transfer process over time, allowing for the optimization of parameters like switching time for improved efficiency.
What are some research gaps identified by the literature review?
Some researchers have focused on column to particle diameter ratios (D/dp) that cause significant deviation from Ergun's equation. Wall effects can also be difficult to assess.
What mathematical models are used to describe fixed-bed regenerators?
The mathematical model (Dispersion Model) considers the effect of intra-particle conduction and dispersion effect.
What is the purpose of the computational modeling section?
The section describes the mathematical modeling of the fixed bed regenerator and the CFD analysis using Ansys Fluent software.
What geometrical details are used in the computational model?
The computational model uses specific dimensions for the regenerator length and diameter, as well as the diameter of the spherical particles.
What are some key settings and parameters used in the CFD simulation?
The CFD simulation used inlet velocity and inlet temperature of flue gas and air, with specified outlet pressure and other viscous/inertial parameters.
What does the book conclude about regenerators?
Regenerators are an effective heat recovery system being used in industries to recover heat from flue gases, and Ansys Fluent can be used to model and simulate the process. A larger regenerator height results in a greater thermal efficiency due to the flue gas having a higher residence time.
What are the governing equations used in the CFD analysis?
The equations are the continuity equation, the momentum equation, and the energy equation.
What boundary conditions were applied in the model?
Inlet velocity and temperature for flue gas and air were specified at the inlet, and zero gauge pressure was applied at the outlet.
How was the solution method implemented in Ansys Fluent?
The velocity formulation was kept absolute, the space was 3D, and for unsteady simulations a transient option was selected. The energy equation option was chosen to get the temperature field.
What were the initial conditions for Transient CFD simulation of Regenerator?
Flue gases at 473 K enter the regenerator bed for a cycle time of 60 seconds, then cooling cycle starts with ambient air at 300 K entering the regenerator in counterflow direction for 60 seconds, and this process alternates in heating cycle and cooling cycle till the steady state is achieved.
What can be gathered from pressure and temperature contours?
Temperature and pressure contours can give insight on the temperature and pressure variation along the bed after heating and cooling cycles.
What were the conclusion for "Effect of flue gas exit temperature and regenerator height on its effectiveness"?
As the regenerator length increases, the exit flue gas temperature decreases due to the heat transfer from flue gas to the solids. There should be an optimal regenerator length such that thermal efficiency is highest, and material cost is also low or optimum.
- Citation du texte
- Dr. Kuldeep Panwar (Auteur), Er. Kuldep Rawat (Auteur), Dr. Neeraj Bisht (Auteur), 2016, Packed Bed Regenerator. CFD Analysis, Munich, GRIN Verlag, https://www.grin.com/document/495135
