Numerical simulation of temperature and concentration distributions in fluidized beds with liquid injection

In this paper a numerical simulation study of dynamic behavior of a fluidized bed with liquid injection is presented. A continuum model has been developed taking into account the mass and energy balances of solid, gas as well as liquid to describe the temperature and concentration distributions in gas-solid-fluidized beds. The model considers the deposition efficiency of the liquid droplets as well as the influence of the spray nozzle region. For solving the non-linear partial differential equations with discrete boundary conditions a finite element method is used. Numerical computations have been done with two different schemes of time integration, a fully implicit and a semi implicit scheme. The complex correlations of mass and liquid flow rates, mass and heat transfer, drying, and transient two-dimensional air humidity, air temperature, particle wetting, liquid film temperature and particle temperature were simulated. The model was validated with transient measurements of the air temperature and air humidity at the outlet of a fluidized bed with water injection.     

Parallel and high resolution numerical solution of concentration and temperature distributions in fluidized beds

In this article higher order in time numerical schemes with efficient time stepping for the solution of concentration and temperature distributions in fluidized beds using parallel computers are presented. The mathematical model equations consist of strongly coupled and semi linear convection-diffusion-reaction equations. Invariant regions for the model are derived to check the solution bounds. The numerical discretization for the space using the finite element method is presented and the numerical treatment is enhanced by using adaptive and higher order linearly implicit Runge–Kutta methods for the time discretization. For different time stepping methods and different spatial grid sizes numerical results are obtained and compared. The methods used show a clear improvement for the problem under consideration compared to previously presented results (Nagaiah, Warnecke, Heinrich, & Peglow, 2008). Additionally, the higher order time stepping methods yield a good parallel efficiency, paving the way for the efficient study of more complex phenomena.     

Fluid dynamic similarity of circulating fluidized beds

The effects of scale-up on the fluid dynamics of circulating fluidized beds (CFB) are investigated using a single cold laboratory facility with the ability to recycle fluidization gas mixtures of adjustable density and viscosity. By matching five dimensionless parameters, experiments employing plastic, glass and steel powders achieve fluid dynamic similarity with high-temperature CFB risers of 0.32, 0.46 and 1 m diameter. Comparisons of results obtained with the plastic and glass powders indicate that static pressure and its fluctuations scale with the riser and particle diameters, respectively. Experiments with the steel powder exhibit incipient choking behavior consistent with the greater analogous bed size that they simulate. The onset of choking with plastic and steel powders is well predicted by the correlation of Yang [Powder Technol., 35 (1983) 143]. Experiments with coated glass beads suggest that the magnitude of the particle Coulomb friction coefficient affects the fluid dynamics of the CFB in the limit where this coefficient is small.     

Size segregation in gas-solid fluidized beds with continuous size distributions

Discrete-particle simulations of a gas-solid fluidized bed are used to investigate the species segregation (de-mixing) behavior of systems with continuous particle size distributions. Both Gaussian and lognormal distributions are investigated over a range of distribution widths, restitution and friction coefficients, and gas velocities. The results indicate that: (i) the average particle diameter decreases as the height within the bed increases, (ii) the level of segregation increases with an increase in the width of the particle size distribution, and (iii) segregation is attenuated as bubbling becomes more vigorous. Furthermore, the shape of the local size distribution (i.e., Gaussian or lognormal) is found to mimic that of the overall size distribution in most regions of the fluidized bed.     

流化床内颗粒混合研究

分析了流化床内颗粒混合机理，综合了床内水平方向和垂直方向上颗粒混合，将床内颗粒混合过程分成两部分：一是向上运动的尾迹相和向下运动的乳化相之间的对流交换，二是乳化相内横向扩散。建立了二维的对流扩散模型，数值结果和实验数据吻合。     

Study of coal and coke ignition in fluidized beds

Ignition tests were conducted with delayed and fluid petroleum cokes, a high volatile bituminous coal and their blends in bench- and pilot-scale fluidized bed combustors. In the bench-scale FBC tests, a visual inspection ignition criterion was developed based on the ‘CO–CO₂’ profiles obtained as a function of time and bed temperature. In the pilot-scale unit, the rapid increase in SO₂ levels was used as the indicator of fuel ignition. In addition to the fluidized bed ignition tests, thermogravimetric analysis (TGA) measurements were made on all fuels and their chars. The results showed that, while the fluid coke (which has the lowest volatile content) was the most difficult to ignite, ignition was not a simple function of volatile content. Further, different test methods and ignition criteria demonstrated significantly different ignition temperatures, with the largest variation arising between bench- and pilot-scale equipment. Finally, tests on mixtures of petroleum coke and coal showed that there was no interaction between these two fuels, and that the coke ignited only when it achieved its own ignition temperature.     

Verification and validation of CFD models and dynamic similarity for fluidized beds

Claims and suggestions in the literature that verification or validation of CFD numerical models has been achieved for fluidized beds are shown to be inconsistent with objective criteria and accepted usage of terminology. Verification involves confirming the accuracy of the computational aspect of the model, for example by comparing results against known solutions, something that is virtually impossible in dense multiphase systems, except for trivial cases. Validation requires objective consideration of computational and numerical error, as well as comparison of model predictions and experimental data over broad ranges of conditions. More care is required in applying these terms, and in planning and conducting experiments to test the validity of fluidized bed numerical codes. Similar considerations apply to experimental attempts to confirm the completeness of sets of matched dimensionless groups used for dynamic scaling of multiphase systems.     

宽筛分脱油油砂颗粒流化床中固含量轴向分布

引言 随着全球石油资源的日益短缺和石油需求的不断增加,开发非常规石油资源作为石油替代能源已引起广泛关注.油砂是非常规石油资源的主要来源之一,是一种含有沥青或焦油的砂或砂岩,其主要成分为沥青、砂、水、黏土及矿物质[1-2].目前,油砂制油工艺主要有热水碱洗[3-4]、溶剂萃取[5-6]和热解焦化[7-9]三种.     

Solid concentration and velocity distributions in an annulus turbulent fluidized bed

Solid concentration and particle velocity distributions in the transition section of a?200 mm turbulent fluidized bed (TFB) and a?200 mm annulus turbulent fluidized bed (A-TFB) with a?50 mm central standpipe were mea-sured using a PV6D optical probe. It is concluded that in turbulent regime, the axial distribution of solid concen-tration in A-TFB was similar to that in TFB, but the former had a shorter transition section. The axial solid concentration distribution, probability density, and power spectral distributions revealed that the standpipe hin-dered the turbulence of gas–solid two-phase flow at a low superficial gas velocity. Consequently, the bottom flow of A-TFB approached the bubbling fluidization pattern. By contrast, the standpipe facilitated the turbulence at a high superficial gas velocity, thus making the bottom flow of A-TFB approach the fast fluidization pattern. Both the particle velocity and solid concentration distribution presented a unimodal distribution in A-TFB and TFB. However, the standpipe at a high gas velocity and in the transition or dilute phase section significantly affected the radial distribution of flow parameters, presenting a bimodal distribution with particle concentration higher near the internal and external wal s and in downward flow. Conversely, particle concentration in the middle an-nulus area was lower, and particles flowed upward. This result indicated that the standpipe destroyed the core-annular structure of TFB in the transition and dilute phase sections at a high gas velocity and also improved the particle distribution of TFB. In conclusion, the standpipe improved the fluidization quality and flow homogeneity at high gas velocity and in the transition or dilute phase section, but caused opposite phenomena at low gas ve-locity and in the dense-phase section.     

流化床中动态行为硬球模拟与软球模拟的比较

用硬球方法和软球方法分别模拟了不等粒径流化床中的动态行为;比较了颗粒平均速度、运动轨迹的差异;研究了恢复系数对流化床中动态行为、颗粒平均速度的影响.研究结果表明：对流化床中动态行为进行模拟的数值结果强烈地依赖于恢复系数,恢复系数较小时,硬球方法可以模拟出鼓泡现象,恢复系数较大时,软球方法可以模拟出鼓泡现象;并且恢复系数越大时,硬球方法中颗粒平均速度随时间的变化越大,而软球方法中颗粒平均速度随时间的变化越小;尽管硬球方法比软球方法更加准确地考虑了颗粒碰撞,但由于硬球方法和软球方法处理颗粒碰撞方式的不同,以及颗粒碰撞时间步长选取原则的差异,使得硬球模拟需要的计算时间远大于软球模拟需要的计算时间.     

Optimal placement of probes for dynamic pressure measurements in large-scale fluidized beds

Pressure data sampled at sufficiently high frequency (typically 20 Hz or higher) can yield much information about the hydrodynamic state of a fluidized bed. Since part of the pressure waves travelling through large (industrial) fluidized beds is only detectable in a limited area of the bed, pressure measurements need to be performed at several positions to cover the whole bed. We examine these local pressure waves (caused by, e.g., passing bubbles or coalescing bubbles) in a 0.80 m i.d. bubbling fluidized bed of Geldart B particles. Experiments and simulations are performed to determine the intensity decrease as local pressure waves propagate from their origin. A new spectral method is applied to determine the degree of coherence for pressure signals measured at two different positions in a fluidized bed. For a superficial gas velocity of 5u_mf, local pressure waves can be detected up to a radial distance of about 0.5 m from their origin; this distance is somewhat lower for lower gas velocities. This means that the radial spacing of pressure probes should not exceed 1 m. For large diameter beds with a bed height below 1.5 m, a set of probes at a single level and at several radial positions is sufficient to observe or monitor the dynamic state of the complete bed; the probes should preferably be placed at a height of 30% to 40% of the total bed height.     

Optical probe measurements of the temperature of burning particles in fluidized beds

The temperatures of freely moving petroleum coke particles burning in the interior of a fluidized bed of sand have been measured. An optical probe technique, employing two-wavelength pyrometry with wavelengths centred at 800 and 1000 nm, was used for the measurements over a wide range of operating conditions. Measurements were conducted in bubbling fluidized beds of 1.82 × 10⁻⁴ m and 1.09 × 10⁻³m sand particles maintained at bed temperatures of 973 and 1200 K. Batches of petroleum coke particles having average diameters in the range 7.5 × 10⁻⁴ to 8.1 × 10⁻³ m were charged into the hot fluidized bed combustor, and upon the passage of a burning particle across the optical probe's field of view, a pulsed signal was generated. Signals from the interior of the bed were continuously monitored by a pair of radiometers interfaced to a high speed data acquisition system and were processed off-line for particle temperature calculations. The average particle temperatures exceeded the bed temperature by 60–360 K for the operating conditions considered. It was found that particle temperatures in the bubble phase were on average, higher than in the emulsion phase. Further, the temperatures of particles were lower closer to the distributor as a consequence of particle circulation and combustion-assisted attrition within the bed.     

Grid leakage in fluidized beds

Solids leakage rates through different grids in two-dimensional and cylindrical beds of cracking catalyst fluidized with air have been measured. Single and multihole perforated plate grids and single nozzle grids were studied. Leakage fluxes were related to the average air velocity through the grid holes. The effects of superficial gas velocity and grid loading were also investigated. The decisive advantage of using nozzles to mini- mize solids leakage in fluidized beds has been observed.     

Dynamics of particle size/conversion distributions in fluidized beds: application to char gasification

Dynamic relations are given which describe the time-dependent behavior of the distribution of reacting particles which are distributed in both size and in their degree of conversion in a fluidized bed. These relations are solve by the method of orthogonal collocation. The adequacy of the orthogonal collocation formulation is determined by comparing steady-state results to steady-state solutions obtained directly by a semi-analytical method. Transient responses of bed weight and particle size/conversion distributions to a step increase in the feed rate are given to exemplify results of the dynamic solution algorithm.     

温度对循环流化床双床气化中热解炉产物的影响

在循环流化床双床煤高温热解气化试验台上,以神木烟煤为燃料在不同温度下进行了热解气化试验,该试验台利用上下返料器将热解炉和气化炉耦合在一起,其中热解炉为N2气氛,气化炉通入空气作为气化剂,试验主要研究热解炉底部温度对热解煤气及热解炉底渣的影响。试验结果表明:随热解炉底部温度升高,热解煤气中H2体积分数升高,CH4,CO2体积分数降低,CO体积分数先降低后升高,热解煤气主要组分气体的收率增加。试验所取热解炉底渣样品的孔比表面积分布和孔比体积分布主要集中于中小孔(0—50 nm),其总比表面积和总孔体积大小顺序为在817℃最大,844℃次之,766℃最小,在N2气氛、1 200℃条件下,CO2反应的活性大小顺序为817℃最大,766℃次之,844℃最小。     

Link between bubbling and segregation patterns in gas‐fluidized beds with continuous size distributions

Experiments involving a bubbling, gas‐fluidized bed with Gaussian and lognormal particle‐size distributions (PSDs) of Geldart Group B particles have been carried out, with a focus on bubble measurements. Previous work in the same systems indicated the degree of axial species segregation varies non‐monotonically with respect to the width of lognormal distributions. Given the widely accepted view of bubbles as “mixing agents,” the initial expectation was that bubble characteristics would be similarly non‐monotonic. Surprisingly, results show that measured bubble parameters (frequency, velocity, and chord length) increase monotonically with increasing width for all PSDs investigated. Closer inspection reveals a bubble‐less bottom region for the segregated systems, despite the bed being fully fluidized. More specifically, results indicate that, the larger the bubble‐less layer is, the more segregated the system becomes. The direct comparison between bubbling and segregation patterns performed provides a more complete physical picture of the link between the two phenomena. © 2011 American Institute of Chemical Engineers AIChE J, 2011     

Measurement of radiative heat transfer coefficient in a high temperature circulating fluidized beds

Experimental measurements of the radiative heat flux were made, and radiative heat transfer coefficients were determined for a circulating fluidized bed of sand particles of mean diameters of 137 and 264 microns. The bed used in this study measured 0.05 m in diameter. The heat transfer test section was 0.9 m long and located in the middle of CFB riser. Operating temperature was varied from 200–600 °C, and the gas velocity in the CFB riser varied from 6 m/s to 11 m/s. The suspension densities covered a range from 3 to 35 kg/m³. Time-averaged radiative heat flux was directly measured with a radiometer. Radiative heat flux and suspension emissivity showed strong dependence on the suspension density. Particle size effect on suspension emissivity was observed. Experimentally determined suspension emissivities, which ranged from 0.3 to 0.85, were in good agreement with the predicted suspension emissivity based on independent scattering theory. The radiative heat transfer coefficients were determined from the measured radiative heat fluxes and were found to be well predicted by the Stefan-Boltzmann law. It was also found that for a dilute system, the prediction of suspension emissivity by Hottel and Sarofim, in conjunction with independent scattering theory of Brewster and Tien, showed good agreement with experimentally determined suspension emissivity.     

Fluid dynamic simulation in a chemical looping combustion with two interconnected fluidized beds

Flow behavior of gas and particles is simulated in a 2-D chemical-looping combustion (CLC) process with two interconnected fluidized beds. A Eulerian continuum two-fluid model is applied for both the gas phase and the solid phase. Gas turbulence is modeled by using a k-ε turbulent model. The kinetic stress is modeled using the kinetic theory of granular flow, while the friction stress is from the combination of the normal frictional stress model proposed by Johnson and Jackson (1987) and the frictional shear viscosity model proposed by Schaeffer (1987) to account for strain rate fluctuations and slow relaxation of the assembly to the yield surface. Instantaneous and local velocity, concentration of particles and granular temperature are obtained. Predicted time-averaged particle concentrations and velocities reflect the classical core-annular flow structure in the air reactor. Flow behavior of bubbles is predicted in the fuel reactor and pot-seal. Computed leakage qualitatively agrees with experimental data in the fuel reactor and pot-seal.