Two-dimensional Fluidized Bed Research Articles

Model-driven Approach to Improve Sago Drying with a Fluidized Bed Dryer

This study presents a model-driven approach to enhance the efficiency of sago drying utilizing a two-dimensional fluidized bed dryer (FBD). ANSYS® DesignModelerTM 2020 R2 software was employed to simulate the drying profile, considering variations in sago bagasse particle diameter (ranging from 500 to 2000 µm), hot air temperature (ranging from 50 to 90 °C), and inlet air velocity (ranging from 1.5 to 2.1 m/s). The simulation results provided valuable insights into the interplay between these critical drying parameters. The model enabled the prediction of moisture content profiles during the sago drying process under different conditions, thereby facilitating comprehension of the system’s behavior. Using Design Expert® 7.00 (DX7), considering energy efficiency and product quality, an optimal set of conditions for sago drying was determined at 2000 µm, 90 °C and 2.1 m/s. This approach not only streamlined the drying process but also significantly reduced energy consumption while ensuring consistent and high-quality sago. The findings of this research offer a practical and sustainable solution for sago producers, which, when applied, can contribute to improved product quality, reduced production costs, and enhanced food security in the region. Furthermore, the model-driven approach and the integration of specialized software tools demonstrate the potential for broader applications in optimizing various drying processes in the food industry.

Experimental investigation of wet particle flow characteristics in a pseudo two-dimensional fluidized bed

The flow characteristics of wet particles in a fluidized bed are extremely complicated due to the presence of moisture. It is essential to understand the complex behavior and flow characteristics of wet particles in the drying, coating and granulation process for optimizing the process efficiency and product quality. Thus, the flow characteristics of wet particles, such as pressure fluctuation, bubble size and bubble fraction, in a pseudo two-dimensional fluidized bed were investigated through the multi-resolution time-frequency analysis of the pressure signal combined with the high-speed camera images. Results show that the critical fluidization velocity increases after the particles are wetted, and it shows a slight increase with higher moisture content levels. Besides, there are two main characteristic frequencies, which correspond to the gas-solid flow state of the gas channel and the large bubble in the bed, respectively. Notably, higher gas velocities promote the dominance of large bubbles. And increasing particle moisture inhibits bubble expansion and reduces bubble size, while dense particle areas expand vertically due to enhanced bonding. The bed height increases with the increase of moisture, and the fluctuation of bubble fraction decreases. In the drying experiment, two characteristic frequencies also appear and the pressure drop reduce slightly due to the reduction of water. During the drying process, both the average bubble fraction and fluctuation intensity increase.

Study of bubble behavior in a gas–solid dense-phase fluidized bed based on deep learning

Fluidization characteristics and hydrodynamics of gas–solid fluidized beds are controlled by bubble properties. However, the existing conventional image segmentation methods are unable to effectively resolve the problems of overlap and background interference between small bubbles, and the segmentation performance for bubbles must still be enhanced. To study the bubble behavior of a gas–solid dense-phase fluidized bed, a two-dimensional fluidized bed with different air distributors was adopted. The application of deep learning to digital image analysis technology is to improve the precision and automation of bubble recognition. The DeepLab V3+ model incorporates multi-scale information, which is well-compatible with uncertain bubble boundary information while enhancing the segmentation effect. A pixel accuracy of 97.95% and a mean intersection over union (MIoU) of 80.91% for multi-group bubble image segmentation were achieved. In comparison to conventional threshold-based recognition methods, the accuracy of the deep learning method increased by up to 40.31% and the MIoU by up to 53.88%, demonstrating the great potential and application value of semantic segmentation in the study of the complex motion behavior of gas–solid sorting fluidized bed bubbles. The influence of the air distributor on the behavior of bubbles is verified. The results indicate that air distributor aperture directly influences bubble diameter, aspect ratio, and shape factor, with coalescence being the primary cause of bubble enlargement.

Numerical simulation and multi-process coupling analysis for biomass pyrolysis fluidized bed reactor based on synergistic effects between biomass and nitrogen inlet modes

Biomass fast pyrolysis fluidized bed reactor is regarded as one of the most promising technologies for high value utilization of biomass. In this work, one two-dimensional fluidized bed reactor model was established, coupling with multiphase CFD simulation method and pyrolysis reaction kinetics, and further verified by the experimental data. This work mainly investigated the effects of biomass inlet and nitrogen inlet modes on multi-process characteristics, such as flow behavior, heat transfer and pyrolysis reaction. The synergistic effects between the two inlet modes were further explored and thus would affect the three-phase pyrolysis product distribution. The results showed that the spatial particle distribution and temperature distribution in the dense phase zone were becoming more uniform and the tar yield increased when the biomass inlet mode changed from single nozzle to double nozzles jetting opposite. Besides, when the nitrogen nozzle number was set as one and the auxiliary gas velocity was set as 0.15 m/s, much larger local circulation formed in the dense phase zone, meanwhile the particle volume fraction and temperature became more evenly distributed, which was conducive to the biomass pyrolysis process. The nitrogen inlet mode was found to show a greater impact on the tar yield compared to the biomass inlet mode. For single gas inlet mode, one biomass nozzle should be preferred for lower auxiliary gas velocity and the opposite double nozzle would be better for higher auxiliary gas velocity. The synergistic effects between biomass and nitrogen inlet modes were further summarized, which were found to have a positive impact on the biomass pyrolysis reaction characteristics. Based on the above conclusions, this work could provide a feasible theoretical guidance for designing fluidized bed pyrolysis reactor and optimizing the biomass pyrolysis process.

Experimental Quantification of the Radial Mixing of Binary Mixtures in Annular Fluidized Beds Using the Capacitance Probe Method

A novel measurement system for radial particle mixing in annular fluidized beds is designed on the basis of the capacitance probe method. Mixing parameters at different radial positions are acquired. The effects of the convection/diffusion mechanism on radial mixing are analyzed individually. It is found that the governing mechanism of mixing at the axial line is convection; meanwhile, diffusion is the governing mechanism of mixing near the wall. The effect of convection on radial mixing at the upper part is more important than that at the lower part. The radial dispersion coefficient ranges from 0.006 to 0.072 m2/s. At the upper part, the radial dispersion coefficient at half-radius is between 0.016 and 0.072 m2/s and that near the wall is 0.006 and 0.028 m2/s, which is four times and twice that at the same radial position and at the lower part, respectively. The radial dispersion coefficient is about 1.5 times that in two-dimensional fluidized beds.

Comparison of hydrodynamics in a gas-solids fluidized bed with binary particle systems for dry coal beneficiation

Fluidization behaviour of two binary particle systems, magnetite225-sand225 (M225-S225) particles with the same mean diameter (dp) and magnetite225-sand304 (M225-S304) particles with the same aerodynamic diameter (da) were observed in a two-dimensional fluidized bed. The number, size and rise velocity of gas bubbles in the binary fluidized bed were characterized by image analysis. A decreasing minimum fluidization velocity (Umf) with increasing sand volume fraction was found in the M225-S225 system while a nearly constant Umf was found in the M225-S304 system. Increasing trends of both bubble size and bubble rise velocity along the bed height were detected and a linear correlation between the bubble size and rise velocity was found for binary systems. The higher the fraction for one type of particles in the binary system is, the closer to the corresponding single-component system the hydrodynamic phenomena and bubble properties will be in both binary systems.

Investigation of electrical field effect on the fluidization characteristics in a two-dimensional fluidized bed

Electrostatic charge accumulation in a gas–solid fluidized bed is considered an adverse effect. However, with the presence of an external electrical field, we can take advantage of electrostatic charging to separate charged particles according to their polarity. In this paper, we proposed a bench-scale two-dimensional fluidized bed with an external electrical field. We then systematically investigated the tribo-charging behavior of anthracite-quartz mixture in the fluidized bed at different conditions and the effect of the external electrical field on their fluidization characteristics. Subsequently, an orthogonal design of experiments analysis was implemented to find the optimized parameters of gas velocity, plate voltage, and initial bed height to separate bituminous coal particles. These findings provide a dry method to separate coal particles from their associated minerals using the proposed two-dimensional fluidized bed with an external electrical field and have general implications for the triboelectric separation of various materials.

Effect of Wall Heat Transfer on the Fluidization Process

The fluidized bed and the fluidization process and characteristics were studied in this paper numerically using Computational Fluid Dynamics (CFD) Ansys Fluent 15.0. Constant temperature was applied to both sides of the two-dimensional fluidized bed geometry. The superficial velocity of the working fluid ranged amid (0.08 – 0.5 m/s) and the initial height of the solid particles changed amid (0.05, 0.1, 0.2 m). Aluminum particles and water was used as working materials within the fluidized bed. The model used for the investigations was validated using Ngoh and Lim research results. The results showed that the fluidization head increases as the water inlet superficial velocity increases. As well as when the water inlet superficial velocity increases, the average solid phase temperature increases.

Experimental Investigations Into Bubble Characteristics in a Fluidized Bed Through Electrostatic Imaging

Fluidized beds are widely applied in many industrial processes. In order to control and optimize the operation of a fluidized bed, it is necessary to develop reliable methods for the measurement of bubble characteristics to monitor the status of the bed. Electrostatic sensing methods based on the detection of charges on particles have been applied to characterize the particle motion in a fluidized bed. However, there is limited research on the measurement of bubble characteristics using the electrostatic methods due to complex electrostatic phenomena around the bubbles. In this article, an imaging method using a two-dimensional electrostatic sensor array is employed for the experimental investigations into the bubble behaviors in a two-dimensional fluidized bed. The bubble size, shape, rising velocity, and generation frequency are measured. Moreover, an optical imaging system is employed to obtain reference information to evaluate the performance of the electrostatic imaging method. Experimental results show that the bubble characteristics measured from the electrostatic sensor array have a good agreement with the results from the optical imaging system. The relative root mean square error between the bubble shapes measured from the electrostatic sensor array and from the optical system is 0.239 with a standard deviation within 4.7%.

Segregation behavior of a ternary mixture of titanium dioxide particles in a pseudo two-dimensional fluidized bed

The segregation behavior of a mixture of silica-coated titanium dioxide (TiO2) particles of three different sizes in a pseudo two-dimensional fluidized bed was studied experimentally by the freeze–sieving method and numerically by the multi-fluid model (MFM). Three-dimensional computational fluid dynamics (CFD) simulations were carried out to evaluate the effects of the solid wall boundary conditions on particle segregation in terms of specularity and particle–wall restitution coefficients. The quantitative indexes of segregation tendency and segregation degree were used to determine the axial segregation of the mixture in triangular coordinates. The simulation results revealed that the axial segregation increased with the specularity coefficient, whereas the particle–wall restitution coefficient had a minor effect on axial segregation. Comparison of the simulation results with experimental data showed that the appropriate value of the specularity coefficient used in the CFD model depended on superficial gas velocity. The study of the effects of superficial gas velocity on segregation behavior demonstrated that the greatest segregation was obtained at minimum fluidization velocity and the segregation decreased as the gas velocity gradually increased.

Particle velocity distribution function around a single bubble in gas-solid fluidized beds

Particle image velocimetry (PIV) is employed in this work to measure particle flow field in a two-dimensional fluidized bed to obtain particle velocity distribution function around a single bubble. Discrete element method (DEM) is also used to investigate particle velocity distribution at the individual particle scale. Both experimental and simulation results show that the velocity distribution of particles surrounding a single bubble can be described by tri-peak model which is a linear superposition of three Maxwellian distributions. A tri-peak distribution model based on the fluid and particle control mechanisms is theoretically derived. Three kinds of models such as tri-peak model, bi-peak model and single-peak model are proposed and compared. The error analysis shows that compared with other models, the tri-peak model can profile particle velocity distribution more accurately.

CPFD simulation on effects of louver baffles in a two-dimensional fluidized bed of Geldart A particles

Numerical simulation based on computational particle fluid dynamics (CPFD) model was carried out in a two-dimensional fluidized bed of Geldart A particles which was previously used by Zhang et al. (2009) to study the effects of louver baffles on bed hydrodynamics and gas mixing. With the help of bubble-based energy-minimization multi-scale (EMMS) drag model, the simulation successfully predicted most of the hydrodynamic results which were highly agreeable with experimental results. However, gas tracing simulation was found to depart far from experimental results due to lack of modeling of porous structure of FCC particles. A direct method to characterize solids back-mixing strength based on the simulated internal circulation fluxes was established. The results not only proved the strong suppression of solids back-mixing by louver baffles, but also validated the feasibility of the two indirect measures used in the previous experiment to characterize solids back-mixing strength. Louver baffles’ ability to break up bubbles was also proved. Under low superficial gas velocities, their vanes can split bubbles directly. However, it acts as a distributor under high superficial gas velocities.

Numerical study of the hydrodynamics of fluidized beds operated under sub-atmospheric pressure

Fluidized beds operated under partial vacuum offer a promising alternative to the expensive freeze-drying technique for processing thermo-sensitive materials in the food and pharmaceutical industries. However, the hydrodynamics of vacuum fluidized beds has not been extensively investigated. In this paper, we apply a two-fluid Eulerian model to investigate the influence of sub-atmospheric pressure on the hydrodynamic characteristics in two-dimensional fluidized beds. The simulations are compared with experiments performed in a batch vacuum fluidized bed plant. A parametric study of the effect of pressure on bed hydrodynamics is performed for two fluidization conditions: (1) constant excess velocity and (2) three times the minimum fluidization velocity. The pressure is varied in the range 1000 mbar–100 mbar with fluidized particles belonging to Geldart group D. A detailed investigation is carried out on porosity distribution, bed expansion, bypass ratio, and bubble characteristics. At low pressure, the porosity distribution is non-homogeneous with increased bubble activity near walls. A decrease in bed porosity and bubble splitting is observed as the operating pressure is reduced. The simulation results are also compared with semi-empirical correlations developed for atmospheric pressure conditions. The applicability of these correlations at sub-atmospheric pressures is, thus, evaluated.

Dynamic forces on an immersed cylindrical tube and analysis of particle interaction in 2D-gas fluidized beds

The interactions of bubbles and particles with fixed cylindrical tubes in two-dimensional fluidized beds were investigated by experiments and by simulations, based on results for single bubbles impinging on a tube. The experimental results based on PIV analysis support our previous force origin model and indicate that the model is able to successfully model bubble behavior and particle motion around fixed objects. The simulation results give useful predictions, dynamic force induced on a tube consists of the force from pressure gradient, fluid viscous force and particle contact force. The predominant force component is from the pressure gradient. As bubbles directly interact with a tube, the particle contact force contribution briefly becomes predominant.Bubble behavior and particle motion are greatly affected by the state of the emulsion phase as the medium of the fluidized bed into which gas is injected. Hence the dynamic forces on immersed objects are directly affected by the state of the emulsion phase.

Bubbling behavior of cohesive particles in a two-dimensional fluidized bed with immersed tubes

Fluidization hydrodynamics are greatly influenced by inter-particle cohesive forces. This paper studies the fluidization of large cohesive particles in a two-dimensional fluidized bed with immersed tubes using “polymer coating” to introduce cohesive force, to gain better understanding of bubbling behavior when particles become cohesive and its effect on chemical processes. The results show that the cohesive force promotes bubble splitting in the tube bank region, thereby causing an increase in the number and a decline in the aspect ratio of the bubbles. As the cohesive force increases within a low level, the bubble number increases and the bubble diameter decreases, while the aspect ratio exhibits different trends at different fluidization gas velocities. The difference in the evolution of bubble size under various cohesive forces mainly takes place in the region without tubes. When the cohesive force is large enough to generate stable agglomerates on the side walls of the bed, the bubble number and the bed expansion sharply decrease. The tubes serve as a framework that promotes the agglomeration, thus accelerating defluidization. Finally, the bubble profile around tubes was studied and found to greatly depend both on the cohesive forces and the location of tubes.

Qualitative comparison of bubble evolution in a two dimensional gas-solid fluidized bed using image analysis and CFD model

Gas-solid fluidized beds are widely used in industries like chemical, metallurgical, environmental, and pharmaceutical sectors for a variety of heat and mass transfer applications. Bubbling beds are the most prevalent type due to excellent solid-gas interfacial contact. Bubble characteristics such as shape, size, and trajectory control the hydrodynamics and therefore heat and mass transfer in fluidized bed reactors. Thus understanding these characteristics is very important for the design and scale-up of fluidized beds. In this work, a pseudo two-dimensional fluidized bed made of polycarbonate material has been used. Single bubbles are injected separately at the center and near the wall by offsetting the jet position, thus presenting the effect wall has on asymmetrical injection as compared to symmetrical injection. The bubble behavior is characterized using an in-house developed digital image analysis technique. The experimental data is compared with an earlier developed Eulerian-Eulerian two-fluid model for simulating dense gas–solid two-phase flow. The high-speed photography reveals an asymmetric wake formation during bubble detachment indicating an early onset of the mixing process. The wall forces act tangentially on the bubble and has a significant impact on the bubble shape, neck formation during detachment and its trajectory through the bed. Larger bubbles are found to drift away from the center with longer paths while the smaller ones found to oscillate. Whereas, centrally injected bubbles found to be more circular and progressing sooner to the surface. This qualitative behavior is well predicted by CFD modelling. The longer bubble hold up in asymmetric injection creates more particle mixing that may improve the fluidized bed heat and mass transfer characteristics.

Experimental study of cluster properties in a two-dimensional fluidized bed of Geldart B particles

The paper presents an analysis of measurement results on cluster properties in a two-dimensional lab-scale cold fluidized bed operating in turbulent and circulating fluidized bed modes. Experiments were carried out with Geldart B particles of (Sauter mean) diameters varying from 0.256mm to 0.440mm. Different particle size distributions and bed mass were used along with different superficial gas velocities in the range of 2.75m/s to 4.00m/s. High-speed images of the bed were taken in a controlled lighting environment. Analysis of the experimental images was conducted to determine the horizontal length scale distribution of the solid clusters along the 3m high CFB. Additionally, the solid hold-up inside the clusters was evaluated. The effects of particle size, bed mass and superficial gas velocity on the measured cluster properties were determined. The solid hold-up of the clusters was correlated to the horizontal length scales of the clusters to estimate the importance of each cluster size in terms of the mass fraction of solids residing in clusters of a specific width range. In the bottom region up to 0.2m from the air distributor, about 5% of the solid mass in clusters was found to reside in structures with horizontal length scales less than 1cm while at 1.2m height the corresponding share was about 10%. This information is relevant for determination of the mesh resolution required to describe all the significant cluster sizes in CFD simulations. The findings of the study on the width distributions of the clusters and the solid concentrations inside the clusters may also be used as a point of comparison with corresponding measures observed in simulations, by which a quantitative validation of CFD results could be made.

A novel dual-material probe for in situ measurement of particle charge densities in gas–solid fluidized beds

Particle charge density is vitally important for monitoring electrostatic charges and understanding particle charging behavior in fluidized beds. In this paper, a dual-material probe was tested in a gas–solid fluidized bed for measuring the charge density of fluidized particles. The experiments were conducted in a two-dimensional fluidized bed with both single bubble injection and freely bubbling, at various particle charge densities and superficial gas velocities. Uniformly sized glass beads were used to eliminate complicating factors at this early stage of probe development. Peak currents, extracted from dynamic signals, were decoupled to determine charge densities of bed particles, which were found to be qualitatively and quantitatively consistent with charge densities directly measured by Faraday cup from the freely bubbling fluidized bed. The current signals were also decoupled to estimate bubble rise velocities, which were found to be in reasonable agreement with those obtained directly by analyzing video images.

Simultaneous measurements of particle charge density and bubble properties in gas-solid fluidized beds by dual-tip electrostatic probes

The aim of this work was to develop a new dual-tip electrostatic probe for in-situ measurements of particle charge density and bubble properties in bubbling fluidized beds. Probes containing two retractable, vertically-aligned tips were tested in a lab-scale two-dimensional fluidized bed operated in both single bubble injection and freely bubbling modes, with glass beads and polyethylene particles of narrow size distributions as bed materials. Different decoupling methods were proposed and employed to analyze the electrostatic signals from the probes. The estimated particle charge density and bubble rise velocity were found to follow the same trends as those measured by a Faraday cup sampling system and obtained from video images respectively, with relative errors depending on the decoupling methods and probe configurations, especially for polyethylene particles.

Devolatilization of a single fuel particle in a fluidized bed under oxy-combustion conditions. Part A: Experimental results

Devolatilization of a single fuel particle and the related flame combustion were studied in a two-dimensional fluidized bed with a quartz wall, allowing visual observation of the conversion process. The aim was to evaluate the devolatilization behavior (ignition, flame temperature, flame life-time, devolatilization time) of different fuels (4 ranks of coal from lignite to anthracite and wood) when replacing O2/N2 by O2/CO2 in O2 volume concentrations from 0% to 40%, at a fixed bed temperature of 1088K using 6mm spherical fuel particles. The volatiles’ flame was recorded by a color video camera to analyze ignition and extinction. The flame temperature was estimated by two-color pyrometry. Two thermocouples were inserted in the fuel particle to measure the temperature at the center and near the surface. Homogeneous and heterogeneous ignition modes, times of devolatilization, and flame duration (flame-life) under different gas atmospheres were analyzed. Results indicate that the mode of ignition of bituminous coal, lignite coal and wood changes when N2 is replaced by CO2. The ignition-delay time is much longer, and the flame temperature is lower in the O2/CO2 atmosphere than in an O2/N2 atmosphere for all the tested fuels. The devolatilization time of the anthracite particle is almost unaffected by the surrounding atmosphere, while for the other fuels this time is generally longer in O2/CO2 than in O2/N2 at the same O2 concentration. The presence of a flame during the volatiles combustion did not accelerate the particle heating, not even at the highest O2 concentration tested (40vol%), however, after the extinction of the flame, the rate of particle heating is significantly affected by the oxygen concentration.