No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Modeling and simulation of an oxygen-blown bubbling fluidized bed gasifier using the Computational Particle- Fluid Dynamics (CPFD) approach
Abstract
Fluidized beds are conventional components of many industrial processes, such as coal gasification for energy generation and syngas production. Numerical simulations help to properly design and understand the complex multiphase flows occurring in these reactors. Two modeling approaches are usually adopted to simulate multiphase flows: the two fluids Eulerian-Eulerian model and the continuous/discrete Eulerian-Lagrangian model. Since fluidized beds account for an extremely large number of particles, tracking each of them could not assure to get results within a reasonable computational time. The Computational Particle-Fluid Dynamics (CPFD) approach, which belongs to the Eulerian-Lagrangian models class, groups together particles with similar key parameters (e.g. composition, size) into computational units (parcels). Parcel collisions are modeled by an isotropic solid stress function, depending on solid volume fraction. In this paper, the bubbling fluidized bed (BFB) upstream gasifier of the EU research infrastructure ZECOMIX (Zero Emissions of Carbon with Mixed technologies) has been simulated using a CPFD approach via Barracuda® software. The effect of different fluidizing agent injection strategies on bed bubbling and mixing, for non-reacting cases, has been studied. The numerical results for a reacting case have been compared to the available experimental data, gathered during the coal gasification campaign. The model has proved to be very useful in the choice of the more efficient injection configuration that assures a more effective contact of the gas with the solid bed and a good bubbling fluidization regime, together with a satisfactory prediction of the outlet gas composition. The numerical approach has turned out to be robust and time-saving and allowed to dramatically reduce the computational cost with respect the classical two fluids Eulerian-Eulerian models.
... The results supported that the two-dimensional CPFD model could effectively predict the flow pattern of particles near the feed zone. Nardo et al. (2018) found that CPFD model could assure a more effective contact of gas with solid bed and a good bubbling fluidization mechanism, and the prediction of the outlet gas composition was satisfactory. Raheem et al. (2019) conducted experimental and CPFD numerical research on CFB combustor under cold flow conditions. ...
Aggregation of fluidization media may appear at the dense phase region of the pant-leg fluidized bed near the incline walls. When the particles flow along the inclined wall, the friction and drag force will cause the particles to accumulate on the inclined wall, resulting in an uneven distribution of particles. The stagnant zones can be minimized by correctly arranging secondary air. Computational particle fluid dynamics (CPFD) method was used to simulate the gas-solid two-phase flow pattern in the dense phase region of pant-leg fluidized bed. Cold tests were performed on a benchtop pant-leg fluidized bed. A high speed imaging technology was used to monitor the flow pattern in the dense phase area, whereas the bubble size and residence time were compared to verify the accuracy of the simulation. The gas-solid flow patterns under various models were simulated. The influence of different secondary air velocities on the reduction of stagnant zone in the dense phase zone of the fluidized bed were predicted. The results indicated that the introduction of secondary air could effectively promote the mixing of particles, and weaken the accumulation of particles on the inclined wall surface. Moreover, secondary air can effectively promote the flow between the gas-solid two-phases and improve the combustion characteristics in the furnace.
A 3D Computational Particle Fluid Dynamic (CPFD) model is validated against experimental measurements in a lab-scale cold flow model of a Circulating Fluidized Bed (CFB). The model prediction of pressure along the riser, downcomer and siphon as well as bed material circulation rates agree well with experimental measurements. Primary and secondary air feed positions were simulated by varying the positions along the height of the reactor to get optimum bed material circulation rate. The optimal ratio of the height of primary and secondary air feed positions to the total height of the riser are 0.125 and 0.375 respectively. The model is simulated for high-temperature conditions and for reacting flow including combustion reactions. At the high temperature and reaction conditions, the bed material circulation rate is decreased with the corresponding decrease in pressure drop throughout the CFB for the given air feed rate.
Download at http://authors.elsevier.com/a/1Snha1H~czy4FV
This paper presents experimental and modelling results of fuel gas obtained in a steam-oxygen blown fluidised bed pilot scale (500 kWth) coal gasifier that is part of the ZECOMIX (Zero Emission of CarbOn with MIXed Technologies) research infrastructure run by ENEA. As there is poor information about start up and steady state operation of fluidised bed gasifiers at such a scale, in this work we investigated the influence of some important parameters such as coal ignition temperature, feeding rate of particulate bed material (olivine sand) and transition from oxidant to reducing environment. The experimental runs confirmed the crucial importance of steam/oxygen ratio (S/O) on the product gas composition. High quality syngas (low content of methane, below 1.0 v/v %, and high content of H2 and CO: CO+H2 61.0 v/v%) was obtained with S/O=1.1.
A shortcut model of the gasifier was also formulated, considering instantaneous coal pyrolysis, kinetics of coke steam reforming reactions in the fluidised bed, and achievement of thermodynamic equilibrium for the water gas shift reaction. The model takes into consideration the average residence time of coke particles into the gasifier: since this is a characteristic parameter of the reactor, the performance of different fluidised bed gasifiers can be simulated.
We first describe the main approaches used to model fluidized suspensions. Focusing on the multifluid approach, we overview the principal averaging techniques that consent to turn granular systems into continua; in particular, we discuss volume, ensemble and time averages. We then use volume averages to derive the Eulerian equations of motion for fluidized suspensions of a finite number of monodisperse particle classes. We introduce the closure problem, and overview some widely adopted closure equations used to express the granular stress and the interaction forces between the phases, giving emphasis to the fluid-particle interaction force, in particular to the buoyancy and drag contributions. We conclude the work by discussing some published CFD simulations of mono and bidisperse fluidized beds, spanning different fluidization regimes and commenting on the insight that these studies provide.
Morgantown Energy Technology Center is developing an advanced moving-bed gasifier, which is the centerpiece of the Integrated Gasifier Combined-Cycle (IGCC) system, with the features of good efficiency, low cost, and minimal environmental impact. A mathematical model of the gasifier, the METC-Gasifier Advanced Simulation (MGAS) model, has been developed for the analysis and design of advanced gasifiers and other moving-bed gasifiers. This report contains the technical and the user manuals of the MGAS model. The MGAS model can describe the transient operation of coflow, counterflow, or fixed-bed gasifiers. It is a one-dimensional model and can simulate the addition and withdrawal of gas and solids at multiple locations in the bed, a feature essential for simulating beds with recycle. The model describes the reactor in terms of a gas phase and a solids (coal or char) phase. These phases may exist at different temperatures. The model considers several combustion, gasification, and initial stage reactions. The model consists of a set of mass balances for 14 gas species and three coal (pseudo-) species and energy balances for the gas and the solids phases. The resulting partial differential equations are solved using a finite difference technique.
An efficient utilization of biomass fuels in power plants is often limited by the melting behavior of the biomass ash, which causes unplanned shutdowns of the plants. If the melting temperature of the ash is locally exceeded, deposits can form on the walls of the combustion chamber. In this paper, a bubbling fluidized bed combustion chamber with 50 MW biomass input is investigated that severely suffers deposit build-up in the freeboard during operation. The deposit layers affect the operation negatively in two ways: they act as an additional heat resistance in regions of heat extraction, and they can come off the wall and fall into the bed and negatively influence the fluidization behavior. To detect zones where ash melting can occur, the temperature distribution in the combustion chamber is calculated numerically using the commercial CPFD (computational particle fluid dynamics) code, Barracuda Version 15. Regions where the ash melting temperature is exceeded are compared with the fouling observed on the walls in the freeboard. The numerically predicted regions agree well with the observed location of the deposits on the walls. Next, the model is used to find an optimized operating point with fewer regions in which the ash melting temperature is exceeded. Therefore, three cases with different distributions of the inlet gas streams are simulated. The simulations show if the air inlet streams are moved from the freeboard to the necking area above the bed a more even temperature distribution is obtained over the combustion chamber. Hence, the areas where the ash melting temperatures are exceeded are reduced significantly and the formation of deposits in the optimized operational mode is much less likely.
Useful as a reference for engineers in industry and as an advanced level text for graduate engineering students, Multiphase Flow and Fluidization takes the reader beyond the theoretical to demonstrate how multiphase flow equations can be used to provide applied, practical, predictive solutions to industrial fluidization problems. Written to help advance progress in the emerging science of multiphase flow, this book begins with the development of the conservation laws and moves on through kinetic theory, clarifying many physical concepts (such as particulate viscosity and solids pressure) and introducing the new dependent variable-the volume fraction of the dispersed phase. Exercises at the end of each chapterare provided for further study and lead into applications not covered in the text itself.
A three-dimensional CFD model was developed to simulate the full-loop of a dual fluidized-bed biomass gasification system consisting of a gasifier, a combustor, a cyclone separator, and a loop-seal. This full-loop simulation includes the chemical kinetic modeling of biomass drying and pyrolysis, heterogeneous char reactions, and homogeneous gas-phase reactions. In the model, the gas phase is described using Large Eddy Simulation (LES) and the particle phase is described with the Multiphase Particle-In-Cell (MP-PIC) method. The simulation was performed using the GPU-accelerated computing and the simulation results were compared with the gas composition and temperature measurements from a pilot-scale biomass gasification power plant (1MWth, 6tonsbiomass/day). The independence of the accuracy of the model on mesh resolution and computational particle number was determined. The impacts of the particle size distributions (PSD) and drag models on the reactive flows were also investigated.
Both two-fluid models embedding the kinetic theory of granular flow for particulate phase stress (TFM) and discrete particle models (DPM) are widely used for the numerical simulation of gas fluidization. In this study, a detailed comparison between results obtained from both TFM and DPM is reported, including axial and radial solid concentration profiles, solids circulation patterns, pressure drop and its standard deviation and granular temperature. It was shown that good agreement can be obtained even in cases of low restitution coefficient, which suggests the possible applicability of kinetic theory of granular flow beyond its nominal range of validity and clearly indicates that the continuum treatment of the solids phase in TFM provides a good approximation of its discrete nature.
A critical validation study of CPFD model was carried on simulating bubbling gas–solid fluidized bed. A previous study on a pseudo-2D fluidized bed with accurate experimental data and TFM modeling results were used to identify its strength and weakness. Although some advantages of the CPFD model were discerned or proved again, an inherent disadvantage was also identified, i.e. its incapability in simulating the right bubble coalescence phenomenon in bubbling fluidized beds. These results can be used as good references for other users and future model improvements.
Energy transport and chemistry are modeled in an extension of the Eulerian–Lagrangian computational particle fluid dynamics (CPFD) methodology. The CPFD methodology is based on the MP-PIC method, which uses a stochastic particle method for the particle phase and an Eulerian method for the fluid phase, to solve equations for dense particle flow. In our extension of CPFD, an enthalpy equation describes energy transport for fluid, and provides for transfer of sensible and chemical energy between phases and within the fluid mixture. Homogenous and heterogeneous chemistry are described by reduced-chemistry, and the reaction rates are implicitly solved numerically on the Eulerian grid. Inter-phase momentum and energy transfer are also implicitly calculated, giving a robust numerical solution from the dilute flow to close-pack limits. A three-dimensional example of a hot fluidized bed coal gasifier is presented with homogeneous and heterogeneous chemistry. The inter-dependencies of fluidization, thermal, and chemistry behaviors in this complex three-dimensional calculation are described.
The capability of the Multiphase Particle in Cell (MP-PIC) approach for modeling a bubbling fluidized bed of Geldart A particles has been investigated. Four different simulation cases, which include three different mesh sizes and two drag models with a realistic particle size distribution, have been designed and tested. Bubble properties have been extracted from the model predictions and compared with the predictions of empirical correlations as well as experimental data. The results show a promising predictive capability of the MP-PIC approach without the need to modify the drag force or other constitutive relationships in the model. This is in contrast to other studies by our group and others using the two-fluid model (TFM) approach. The use of the commercial code Barracuda to solve the MP-PIC system of equations limits our ability to understand the physical or numerical reasons behind this and therefore further work using an open-source code would be beneficial.
Heterogeneous catalytic chemistry is used throughout the chemical and petro-chemical industry. In predicting the performance of a reactor, knowing the gases and solids flow dynamics is as important as having good chemical rate expressions. This paper gives the solution of ozone decomposition in a bubbling bed using the CPFD numerical scheme which is a Eulerian–Lagrangian solution method for fluid–solid flows. The ozone decomposition can be described by a single stoichiometric equation and has a first order reaction rate. The ozone decomposition is a standard problem for chemical analysis and has been used to characterize gas–solid contacts in fluidized beds. The accuracy of predicting the ozone decomposition comes from correctly predicting the bed dynamics. The solution in this study is three-dimensional and predicts the coupled motion of both solids and gas. The chemical rate equation uses solids volume fraction, but the numerical method could calculate chemistry on the discrete catalyst, including a variation in size (surface area) if such a rate equation was available. The numerical results compare well with an analytic solution of the decomposition rate, and calculated results compare well with the experiment by Fryer and Potter [Fryer, C. and Potter, O.E, (1976), “Experimental investigation of models for fluidized bed catalytic reactors,” AIChE J., 22.].
In this paper, we consider the evolution of an initially small voidage disturbance in a gas-fluidized bed. Using a one-dimensional model proposed by Needham & Merkin (1983), Crighton (1991) has shown that weakly nonlinear waves of voidage propagate according to the Korteweg–de Vries equation with perturbation terms which can be either amplifying or dissipative, depending on the sign of a coefficient. Here, we investigate the unstable side of the threshold and examine the growth of a single KdV voidage soliton, following its development through several different regimes. As the size of the soliton increases, KdV remains the leading-order equation for some time, but the perturbation terms change, thereby altering the dependence of the amplitude on time. Eventually the disturbance attains a finite amplitude and corresponds to a fully nonlinear solitary wave solution. This matches back directly onto the KdV soliton and tends exponentially to a limiting size. We interpret the series of large-amplitude localized pulses of voidage formed in this way from initial disturbances as corresponding to the ‘voidage slugs’ observed in gas fluidization in narrow tubes.
Although great progress has been made in modeling the gas fluidization of Geldart B and D particles and dilute gas−solid flow by standard Eulerian approach, researchers have shown that, because of the limitation of computational resources and the formation of subgrid-scale (SGS) heterogeneous structures, Eulerian model with a suitable SGS model for constitutive law is necessary to simulate the hydrodynamics of large-scale gas-fluidized beds containing Geldart A particles. In this article, a state-of-the-art review of Eulerian modeling of Geldart A particles in gas-fluidized beds is presented. The available methods for establishing SGS models are classified into six categories, that is, empirical correlation method, scaling factor method, structure-based method, modified Syamlal and O’Brien drag correlation method, EMMS-model-based method, and correlative multiscale method. The basic ideas of those methods, as well as their advantages and disadvantages, are reviewed. Finally, directions for future research are indicated.
The high-temperature rate of reaction of the homogeneous, reverse water–gas shift reaction (rWGSR) has been evaluated in quartz reactors with rapid feed preheating under both low- and high-pressure conditions. The form of the power-law rate expression was consistent with the Bradford mechanism. The Arrhenius expressions for the reaction rate constant, corresponding to the empty reactor, were in very good agreement with the low-pressure results of Graven and Long, but yielded rate constants roughly four times greater than those obtained in our packed reactor and those reported by Kochubei and Moin and by Tingey. Reactor geometry was not responsible for these differences because computational fluid dynamics simulations revealed similar residence time distributions and comparable conversions when the same kinetic expression was used to model the rWGSR in each reactor. Most likely, the empty NETL reactor and the Graven and Long reactor did not attain an invariant value of the concentration of the chain carrier (H) at low reaction times, which led to an overestimation of the rate constant. Conversions attained in an Inconel® 600 reactor operating at comparable conditions were approximately two orders of magnitude greater than those realized in the quartz reactor. This dramatic increase in conversion suggests that the Inconel® 600 surfaces, which were depleted of nickel during the reaction, catalyzed the rWGSR. © 2004 American Institute of Chemical Engineers AIChE J, 50: 1028–1041, 2004
A new model is presented for numerical simulations of collisional transfer of mass, momentum and energy in gas/liquid/solid fluidized beds. The mathematical formulation uses a collision model similar to that of Bhatnagar, Gross, and Krook (BGK), in a particle distribution function transport equation, in order to approximate the rates at which collisions bring about local equilibration of particle velocities and the masses, compositions, and temperatures of liquid films on bed particles. The model is implemented in the framework of the computational-particle fluid dynamics (CPFD) numerical methodology, in which the particle phase is represented with computational parcels and the continuous phase is calculated on Eulerian finite-difference grid. Computational examples using the Barracuda® code, a commercial CFD code owned by CPFD Software, LLC, show the ability of the model to calculate spray injection and subsequent liquid spreading in gas/solid flows.
A steady state model of moving-bed coal gasification reactors has been developed. Model predictions are in agreement with published commercial plant data for Lurgi pressurized gasification reactors and a pilot plant slagging gasifier. The dependence of reactor performance on operating variables has been studied for Illinois and Wyoming coals. For a given coal, maximum efficiency is determined by the coal-to-oxygen feed ratio. The location of the maximum temperature, which defines the combustion zone, is an important operating variable. Efficient operation of the dry ash reactor cannot be carried out below a critical feed gas temperature because of insufficient gasification and excessive carbon loss in the ash.
The kinetics of the high-temperature (1070–1134 K), low- and high-pressure gas-phase forward water–gas shift reaction (fWGSR) were evaluated in an empty quartz reactor and a quartz reactor packed with quartz particles. The power-law expression for the reaction rate was consistent with the Bradford mechanism and was invariant with respect to pressure. The experimental rate constant was lower than that published by Graven and Long, and slightly higher than estimates obtained using the reaction rate expression derived from the Bradford mechanism in conjunction with values of reaction rate constants obtained from the GRI database. Similar experiments conducted using a reactor composed of Inconel® 600, a representative reactor shell material, exhibited substantially enhanced rates of reaction. A simple power-law rate expression was incorporated into a surface-catalyzed plug flow reactor (PFR) model to correlate the results between 600 and 900 K. Palladium and palladium–copper alloy surfaces, representative of hydrogen membranes, were also shown to enhance the fWGSR rate, but not as much as the Inconel® 600 surfaces. © 2005 American Institute of Chemical Engineers AIChE J, 2005
A multiphase particle-in-cell (MP-PIC) method has been developed. This numerical technique draws upon the best of Eulerian/Eulerian continuum models and Eulerian/Lagrangian discrete models. The MP-PIC method uses an accurate mapping from Lagrangian particles to and from a computational grid. While on the grid, continuum derivative terms that treat the particle phase as a fluid are readily evaluated and then mapped back to individual particles. The result of this procedure is a computational technique for multiphase flows that can handle particulate loading ranging from dense to dilute, a distribution of particle sizes and a range of particle materials. The dense particulate model represents separated flows of particles and includes drag exerted by a gas phase, inter-particle stresses, particle viscous stresses and gas pressure gradients. Six problems are presented to demonstrate the MP-PIC method. This MP-PIC method has important applications in fluidized beds (combustion, catalytic cracking), sedimentation, separation and many other granular flows.
This paper describes several improvements to a numerical model introduced by O’Rourke et al. (2009) for collisional exchange and damping in dense particle flows. O’Rourke et al. (2009) use a Bhatnagar, Gross, and Krook (BGK) approximation to the collision terms in a particle distribution function transport equation to model the effects of particle collisions on damping fluctuating particle velocities and, in gas/liquid/solid beds, fluctuating temperatures and compositions of liquid films on particle surfaces. In this paper we focus on particle flows in which the particles have no liquid films and report on an improved expression we have developed for the collision damping time of particle velocity fluctuations used in the BGK approximation. The improved expression includes the effects on the collision damping time of the particle material coefficient of restitution and of non-equilibrium particle velocity distributions. The collision model improvements are incorporated into the general-purpose computational-particle fluid dynamics (CPFD) numerical methodology for dense particle flows. Three computational examples show the benefits of using the new collision time in calculations of particle separation in polydisperse dense particle flows and calculations of colliding particle jets.
This paper reviews the use of discrete particle models (DPMs) for the study of the flow phenomena prevailing in fluidized beds. DPMs describe the gas-phase as a continuum, whereas each of the individual particles is treated as a discrete entity. The DPMs accounts for the gas–particle and particle–particle interactions. This model is part of a multi-level modeling approach and has proven to be very useful to generate closure information required in more coarse-grained models. In this paper, a basic DPM, based on both the hard- and soft-sphere approaches is described. The importance of the closures for particle–particle and gas–particle interaction is demonstrated with several illustrative examples. Finally, an outlook for the use of DPMs for the investigation of various chemical engineering problems in the area of fluidization is given.
A three-dimensional, incompressible, multiphase particle-in-cell method is presented for dense particle flows. The numerical technique solves the governing equations of the fluid phase using a continuum model and those of the particle phase using a Lagrangian model. Difficulties associated with calculating interparticle interactions for dense particle flows with volume fractions above 5% have been eliminated by mapping particle properties to an Eulerian grid and then mapping back computed stress tensors to particle positions. A subgrid particle, normal stress model for discrete particles which is robust and eliminates the need for an implicit calculation of the particle normal stress on the grid is presented. Interpolation operators and their properties are defined which provide compact support, are conservative, and provide fast solution for a large particle population. The solution scheme allows for distributions of types, sizes, and density of particles, with no numerical diffusion from the Lagrangian particle calculations. Particles are implicitly coupled to the fluid phase, and the fluid momentum and pressure equations are implicitly solved, which gives a robust solution.
A critical review of the literature on fluidization using the kinetic theory of granular flow is presented. An equation of state for the particles relating solids pressure to the granular temperature and the solids volume fraction, similar to the van der Waals equation for gases, has been verified experimentally to be reasonably correct. Experiments have also shown that the particulate viscosity expression obtained from the kinetic theory gives the same values as that measured by classical methods.We demonstrated using a kinetic theory based particle image velocity (PIV) meter that there are two kinds of turbulence in fluidization:1.random oscillations of individual particles, measured by the classical granular temperature and2.turbulence caused by the motion of clusters of particles, measured by the average particle normal Reynolds stress.These two kinds of turbulence give rise to two kinds of mixing, mixing on the level of a particle and mixing on the level of cluster or bubble. To compute the granular temperature, it must be programmed into the computational fluid dynamics (CFD) codes. The code itself computes the Reynolds stresses, similar to the calculation of single-phase turbulence by direct numerical computation.CFD simulations by several groups throughout the world have shown that the multiphase flow models correctly predict transient and time-averaged behavior of fluidized beds: bubbles, clusters and flow regimes. Two challenge problems in the last decade show the capability of the hydrodynamic models to predict, at least qualitatively, radial and axial profiles before their publication.
The role of medium size facilities in the HPC ecosystem: the case of the new CRESCO4 cluster integrated in the ENEAGRID infrastructure
- G Ponti
- F Palombi
- D Abate
- F Ambrosino
- G Aprea
- T Bastianelli
- F Beone
- R Bertini
- G Bracco
- M Caporicci
- B Calosso
- M Chinnici
- A Colavincenzo
- A Cucurullo
- P Dangelo
- M De Rosa
- P De Michele
- A Funel
- G Furini
- D Giammattei
- S Giusepponi
- R Guadagni
- G Guarnieri
- A Italiano
- S Magagnino
- A Mariano
- G Mencuccini
- C Mercuri
- S Migliori
- P Ornelli
- S Pecoraro
- A Perozziello
- S Pierattini
- S Podda
- F Poggi
- A Quintiliani
- A Rocchi
- C Scio
- F Simoni
- A Vita
Ponti, G., F., Palombi, D., Abate, F., Ambrosino,
G., Aprea, T., Bastianelli, F., Beone, R.,
Bertini, Bracco, G., Caporicci, M., Calosso, B.,
Chinnici, M., Colavincenzo, A., Cucurullo, A.,
Dangelo, P., De Rosa, M., De Michele, P.,
Funel, A., Furini, G., Giammattei, D.,
Giusepponi, S., Guadagni, R., Guarnieri, G.,
Italiano, A., Magagnino, S., Mariano, A.,
Mencuccini, G., Mercuri, C., Migliori, S.,
Ornelli, P., Pecoraro, S., Perozziello, A.,
Pierattini, S., Podda, S., Poggi, F., Quintiliani,
A., Rocchi, A., Scio, C., Simoni, F., Vita, A.
(2014). The role of medium size facilities in
the HPC ecosystem: the case of the new
CRESCO4 cluster integrated in the
ENEAGRID infrastructure. In Proceedings of
the 2014 International Conference on High
Performance Computing and Simulation,
HPCS 2014, art. no. 6903807, 1030-1033.