Computational Fluid Dynamics Discrete Element Method Simulation Research Articles

Simulating aerosol droplet dynamics and coalescence in tracheal intubation insights from a CFD-DEM analysis of PSAR tube interactions

This study employed the Computational Fluid Dynamics-Discrete Element Method (CFD-DEM) to investigate targeted droplet delivery in tracheal intubation. Our findings reveal that droplet behavior and deposition patterns are significantly influenced by droplet size. Smaller droplets (3 μm and 5 μm) closely follow airflow, showing similar deposition patterns in both CFD-DEM and CFD-DPM simulations. In contrast, larger droplets (7.5 μm and 10 μm) demonstrate increased deposition, particularly in the bent section of the Point Source Release tube (PSAR tube), due to enhanced inertia following collisions and coalescence. The study highlights that larger droplets experience a higher frequency of coalescence, especially in curved tube sections. This leads to altered droplet size distribution at the outlet, affecting deposition in the lower respiratory tract. These insights are crucial for optimizing inhalation therapy devices, emphasizing the importance of considering droplet interactions in CFD simulations to accurately predict aerosol behavior in pulmonary drug delivery systems.

A modified CFD-DEM method for accurate prediction of the minimum fluidization velocity

Monodisperse fluidized bed experiments have been carried out on 13 kinds of particles with different particle sizes in a cylindrical fluidized bed. The accuracy of nine commonly used monodisperse drag models, including the Gidsapow, BVK, and Zhou–Fan models is verified. It is found that the Zhou–Fan model is the most accurate at low and moderate values of the particle Reynolds number. It is found that with the traditional computational fluid dynamics–discrete element method (CFD-DEM) method it is difficult to obtain an accurate minimum fluidization velocity owing to difficulty in obtaining an accurate value of the minimum fluidization solid volume fraction. A modified method is proposed to increase the accuracy of prediction of the minimum fluidized solid volume fraction by using a virtual particle size in the calculation of the particle collision process. The effectiveness of the modified method is verified by comparing the CFD-DEM simulation results with experimental results for Geldart-D particles. The relative error of the minimum fluidization velocity is found to be reduced from 24.6% to 2%.

Hydraulic conveying characteristics of particles in bend based on numerical simulation and explainable stacking machine learning model

As a hydraulic lifting pipeline structure widely used in deep-sea oil, gas transportation, and sediment dredging projects, the pipeline configuration is related to the improvement of transportation efficiency and pipeline safety. Particularly, the bending section consumes the most energy and withstands severe erosion. Understanding and predicting the conveying characteristics of two-phase flow in bends is therefore crucial. In this study, CFD-DEM (computational fluid dynamics-discrete element method) simulation method is employed to calculate various cases, considering five parameters: pipeline bending radius and angle, conveying velocity, particle diameter, and concentration, to explore the influence of these parameters on pressure drop and erosion rate of pipeline and result in a data set of hundreds of cases. Based on this data set, seven machine learning models are trained to predict pressure drop and erosion rate, respectively. To enhance model accuracy, the stacking method in ensemble learning is employed to combine multiple models with good performance. Additionally, the Optuna and SHAP (SHapley Additive exPlanation) methods are utilized to optimize hyperparameters and explain the degree to which parameters impact the predictions. The result demonstrates that pressure drop is almost unaffected by bending radius, while erosion rate initially decreases and then increases with bending angle, and both increase with other parameters. Among the evaluated models, artificial neural network, XGBoost, and random forest all demonstrate high prediction accuracy. The stacking model further improves the accuracy, with mean absolute error improving by 21.7% and 32.2%, and the SHAP method demonstrated good interpretability, which is basically consistent with CFD-DEM results.

Modeling of permeability impairment dynamics in porous media: A machine learning approach

The prediction of clogging and permeability impairment dynamics in porous media is crucial for the optimization of various industrial and natural processes. This paper presents a novel machine learning-based approach for predicting the dynamics of throat clogging and permeability impairment due to fine migration within realistic porous media under varying hydro-physical conditions. A Computational Fluid Dynamics-Discrete Element Method (CFD-DEM) numerical framework, employing a four-way coupling scheme, was used to generate the data for training and validation of the Machine Learning Model (MLM). One hundred and twenty distinct CFD-DEM simulations were performed to generate over 190,000 data points, at throat level, for the training of the MLM. Simulation cases encompassing ranges of porous media geometry, fine particle size, flow velocity, fine particle concentration, grains surface roughness, and fines and grains zeta potential. Geometries of porous media were extracted from high-resolution 3D images of natural sand obtained using micro-computed tomography imaging. The developed MLM predicts the temporal evolution of clogged throats and permeability impairment. The MLM was established by connecting three Machine Learning Sub-Models (MLSMs). The first is a throat-classification MLSM; which classifies the throats based on their location and size to identify clogged throats. Subsequently, a pore volume regression MLSM is implemented to identify the pore volume at which each clogged throat becomes clogged. Finally, the permeability impairment regression MLSM predicts the permeability reduction based on the clogged throat's information and pore volumes associated with clogging. The throats classification in the final MLM showed an accuracy of 95% in predicting clogged throats when compared to direct CFD-DEM simulations whereas the prediction of the permeability impairment had an R-squared value of 0.99. The MLM developed in this study stands as a robust framework for precisely quantifying key microscale parameters; where its predictions were used to quantify the significance of altering the hydro-physical parameters on the microscale parameters of the clogging dynamics. The proposed MLM provides an accurate and fast prediction of porous media clogging and permeability impairment dynamics, with potential applications in various industries, including oil and gas, environmental engineering, and material science.

CFD-DEM simulations of a stirred liquid bath containing particles based on a theoretical scaling framework

BackgroundCoupled computational fluid dynamics-discrete element method (CFD-DEM) simulations are valuable tools for studying particle-fluid coupling problems. However, because of the many particles involved, full-scale simulations are not feasible. MethodsA theoretical scaling framework developed in our previous work was introduced for the first time in the CFD-DEM simulations. A 1/2 scaled-down numerical model was established based on the physical model used to observe the dynamic behavior of dross particles in a hot-dip galvanizing bath. The applicability of the theoretical scaling framework for the CFD-DEM simulations was demonstrated by comparing the results of the physical and scaled-down numerical models. Additionally, the effects of the belt width and velocity on the flow field and dynamic behavior of the particles were investigated. Significant findingsThe results show that the evolution of the particle-accumulation morphology at different belt widths in scaled-down simulations agreed with that in physical model experiments, confirming that the theoretical scaling framework can be used in CFD-DEM simulations and is advantageous in significantly reducing number of particles required for the simulations. As the belt speed decreased, the number of suspended particles also decreased. Furthermore, a reduction in the belt width led to a shift of the aggregation region of the suspended particles away from the vicinity of the symmetrical plane of the bath toward the sidewalls. This shift of the aggregation region was beneficial in preventing the adhesion of dross particles to the surface of the steel strip during the hot-dip galvanizing process.

Comparative analysis on gas–solid drag models in MFIX-DEM simulations of bubbling fluidized bed

In this study, the open-source software MFIX-DEM simulations of a bubbling fluidized bed (BFB) are applied to assess nine drag models according to experimental and direct numerical simulation (DNS) results. The influence of superficial gas velocity on gas–solid flow is also examined. The results show that according to the distribution of time-averaged particle axial velocity in y direction, except for Wen–Yu and Tenneti–Garg–Subramaniam (TGS), other drag models are consistent with the experimental and DNS results. For the TGS drag model, the layer-by-layer movement of particles is observed, which indicates the particle velocity is not correctly predicted. The time domain and frequency domain analysis results of pressure drop of each drag model are similar. It is recommended to use the drag model derived from DNS or fine grid computational fluid dynamics–discrete element method (CFD-DEM) data first for CFD-DEM simulations. For the investigated BFB, the superficial gas velocity less than 0.9 m·s−1 should be adopted to obtain normal hydrodynamics.

Clogging and permeability reduction dynamics in porous media: A numerical simulation study

The dynamics of fine particle entrainment, transport, and deposition within pore systems, and in particular, the capacity for mobile fines to impair permeability within porous media is critical in many industrial applications. Considerable effort has been expended over the past several decades to identify and parametrize the governing factors that control permeability reduction, with studies employing a combination of physical experimentation and numerical simulation toward this aim. The objective of this work was to numerically investigate the impact of pore space geometry and fine sizes on the clogging dynamics of porous media. The clogging dynamics were characterized by the length of clogging zone (LCZ), the critical throat size of clogging (CTZC), the clogged fraction of throats (CFT), and the permeability reduction (PR). We implemented a computational fluid dynamics-discrete element method (CFD-DEM) numerical framework with a four-way coupling scheme to obtain insights into throat clogging and permeability reduction within heterogeneous porous media utilizing four different monodisperse suspensions. Geometries of porous media were extracted from 3D images of sand packs obtained using micro-computed tomography. The geometries had porosity values ranging from 0.35 to 0.57 and initial permeabilities between 1.35 and 26.32 μm2. CFD-DEM simulations were performed on each geometry four times, varying the injected fine particle size from 5 to 15 μm diameters. Findings indicate that pore systems with tortuosity <1.28 and angles of solid grains orientation larger than 46o tend to have shorter length clogging zones and greater permeability reduction. Furthermore, although systems with porosities higher than 0.48 tend to have a relatively high clogged fraction of throats, they undergo lower permeability reduction. It was also observed that within the studied pore systems, injection of fine particles larger than 7 μm into systems with aspect ratios higher than 2.5, results in a lower fraction of clogged throats. Additionally, for the studied pore networks,< 7 μm fine particles injected into porous media with a large pore body (> 66 μm) and throat (> 22 μm) radii tended to result in larger clogging zones and a greater critical throat size of clogging.

Numerical investigation of subcritical and supercritical carbon dioxide fluidized beds using two fluid model and discrete element method

Fluidization state of ambient, subcritical and supercritical carbon dioxide (CO2) fluid fluidized beds are simulated using Computational Fluid Dynamics-Discrete Element Method (CFD-DEM) and Two-Fluid Model (TFM) combined with Low Density Ratio-Kinetic Theory of Granular Flow (LDR-KTGF). Quantitative and qualitative analysis were executed between TFM and CFD-DEM simulation results. Fluidization process using two methods coincides with predictions of stability function and different criteria for distinguishing from particulate and aggregative fluidizations. A transitional fluidization with the wave-like and chunk-like flows occurs in subcritical CO2 fluid fluidized bed, and the bubble-like and particle aggregative-like flows are observed in ambient and supercritical states, respectively. The discrepancies in volume fractions and velocities obtained using TFM and CFD-DEM approaches were small, while the predicted turbulent kinetic parameters and bubble-like kinetic energy show sensitivity to the turbulence model. Statistical analysis of the turbulent normal Reynolds stresses indicated an anisotropic flow of particles.

CFD-DEM simulation of fluorination reaction in fluidized beds with local grid and time refinement method

The gas-solid reaction process with wide particle size distribution is extensively used in the chemical engineering field, especially the particle reacts with the gas gradually, such as fluorination reactions in fluidized beds. When the computational fluid dynamics-discrete element method (CFD-DEM) is used for the coupling simulation of multiphase and polydisperse particle reaction system, the grid size directly affects the accuracy of flow field information and simulation of chemical reaction. Furthermore, particle calculation time step will directly affect the efficiency of coupling calculation. In this work, a local grid and time step refinement method is proposed to simulate multiphase and polydisperse particle fluidization reaction system. In this method, the refined DEM grids are automatically generated in the computational domain around the fine particles, and the detailed fluid phase information is obtained with the interpolation algorithm. In the two-phase coupling process, particles are divided into different groups based on physical properties, each group has its own independent time step. The multistage conical-cylindrical spouted bed is proposed for the fluorination reaction process; the operating gas velocity range of the polydisperse particle system is extended by the new design while the particle size distribution changes with the gas-solid reaction process. It is demonstrated that the local grid and time step refinement method can improve the accuracy and efficiency of the traditional CFD-DEM method in the reaction process simulation, which describes a polydisperse particle system with wide particle size distribution. Aimed at improving the simulation accuracy and efficiency, this paper will be helpful for simulating the particle reaction process in the gas-solid fluidized bed and beneficial for the development of the CFD-DEM method.

Coupled CFD-DEM Simulation of Seed Flow in Horizontal-Vertical Tube Transition

A series of computational fluid dynamics–discrete element method (CFD-DEM) simulations were applied to seed flow in horizontal-vertical 90-degree elbows. The performance of one-way and two-way CFD-DEM coupling methods was compared. Additionally, simulated seed velocities were compared to the current pneumatic conveying theory for each coupling method. Simulated field peas (Pisum sativum) were pneumatically conveyed to study the effect of air velocity (20, 25, and 30 m/s), seed rate (0.07, 0.21, and 0.42 kg/s), elbow diameter, D, (48.3, 60.3, and 72.4 mm), and elbow bend radius (1.5D, 2.5D, 3.5D, and 4.5D) on seed attributes (trajectory, velocity, and force). Results showed that seed velocity was significantly different between one-way and two-way coupling. Both methods resulted in nearly identical seed trajectory and force. Overall, simulated seed velocities had a strong correlation to values calculated through the current pneumatic conveyance theory. Dimensional analysis revealed that seed contact force was proportional to the elbow diameter to the power of 0.26 and inversely proportional to the elbow bend radius to the power of 0.5. Simulation results indicated that one-way coupling could be suitable to describe seed flow when two-way coupling may not be possible or practical.

A novel coupling method for unresolved CFD-DEM modeling

In CFD-DEM (computational fluid dynamics-discrete element method) simulations particles are considered Lagrangian point particles. The details of the flow near the particle surface are therefore not fully resolved. When the particle scale is larger than the resolved flow scale, the coupling between the CFD model and the DEM model is critical. An effective coupling scheme should minimize the risk of artificial influences on the results from choices of numerical parameters in implementations and consider efficiency and robustness. In this work, a novel coupling method is developed. The method includes both the smoothing of the particle data and the sampling of the gas phase quantities. The smoothing employs the diffusion-based method. The gas sampling method can reconstruct the filtered fluid quantities at the particle center. The sampling method is developed based on the diffusion-based method with higher efficiency. The new method avoids mesh searching and it can be easily implemented in parallel computing. The developed method is validated by the simulation of a forced convection experiment for a fixed bed with steel spheres. With the well-posed grid-independent coupling scheme, the simulation results are in good agreement with the experimental measurements. The coupling effects and the computational cost are discussed in detail.

A detailed gas-solid fluidized bed comparison study on CFD-DEM coarse-graining techniques

Computational Fluid Dynamics - Discrete Element Method (CFD-DEM) is a numerical tool used for detailed fluidized bed studies. However, CFD-DEM is computationally expensive, leading to a restriction regarding the number of simulated particles. Therefore, coarse-graining techniques have been developed to increase the simulation scale or reduce computational requirements for CFD-DEM simulations in fluidized beds. In this work, we critically compared the coarse-graining scaling laws of Mu et al. [2020, Chemical Engineering Science: X 6.] and Sakai and Koshizuka [2009, Chemical Engineering Science: 64, 533–539.] for their effectiveness in characterizing the original system. The first mentioned approach was not able to accurately characterize the original system, while the latter methodology showed good correspondence. We also demonstrated that applying a continuous two-way smoothing function to the coarse-graining process can yield grid-independent solutions, particularly when the particle diameter is much larger than the grid cell size, which is often the case in CFD-DEM simulations.

CFD–DEM Simulation of Dust Deposition on Solar Panels for Desert Railways

With the greening of the railway energy supply chain, large-scale photovoltaic power stations will be the best choice to integrate with the railways. Understanding the deposition mechanisms and rules of dust grains on photovoltaic panels is of great guiding significance for the operation of photovoltaic (PV) power stations. In this paper, based on computational fluid dynamics (CFD) combined with the discrete element method (DEM), the dynamic dust deposition process on solar panels was simulated, and the flow field around solar panels and the movement of dust particles in the wind were calculated. The simulation results of clay particles (d = 10 mm) and fine sand particles (d = 100 mm) under different wind speeds showed that the clay particles could follow the air flow properly, and their deposition rate was only 4.6%, while the deposition rate of the fine sand particles was up to 32%, which was determined by the inflow wind speed and cohesion parameters. An image of the non-uniform distribution of particles on the panels was given in this paper for the first time. This will provide a basis for a more accurate assessment of the impact of dust accumulation on PV output in real-world environments. These results provide a critical reference for railway photovoltaic power supply development in desertification areas.

Study of fluid cell coarsening for CFD-DEM simulations of polydisperse gas–solid flows

Particle polydispersity is ubiquitous in industrial fluidized beds, which possesses a significant impact on hydrodynamics of gas–solid flow. Computational fluid dynamics-discrete element method (CFD-DEM) is promising to adequately simulate gas–solid flows with continuous particle size distribution (PSD) while it still suffers from high computational cost. Corresponding coarsening models are thereby desired. This work extends the coarse-grid model to polydisperse systems. Well-resolved simulations with different PSDs are processed through a filtering procedure to modify the gas–particle drag force in coarse-grid simulations. We reveal that the drag correction of individual particle exhibits a dependence on filtered solid volume fraction and filtered slip velocity for both monodisperse and polydisperse systems. Subsequently, the effect of particle size and surrounding PSD is quantified by the ratio of particle size to Sauter mean diameter. Drag correction models for systems with monodisperse and continuous PSD are developed. A priori analysis demonstrates that the developed models exhibit reliable prediction accuracy.

CFD–DEM simulation of particle deposition characteristics of pleated air filter media based on porous media model

In this study, the three-dimensional physical model of pleated air filtration media was simplified to porous media model, and the calculation parameters of porous media were obtained based on experimental data. The model of V-shaped pleated air filter media is constructed, the height of the media pleat is 50 mm and the pleat thickness is 4 mm, the pleat angle is 3.7°. The Hertz-Mindlin contact model was modified by Johnson Kendall Roberts (JKR) adhesion contact model. The deposition process of particles in media was simulated based on computational fluid dynamics (CFD) theory and discrete element method (DEM). Results show that the CFD–DEM coupling method can be effectively applied to the macro research of pleated air filter media. The particles will form dust layer and dendrite structure on the fiber surface, and the dust layer will affect the subsequent air flow organization, and the dendrite structure will eventually form a “particle wall”. The formation of the “particle wall” will prevent the particles from moving further in the fluid domain, which makes area of pleated angle become the “low efficiency” part about the particle deposition. Compared with area of pleated angle, the particles are concentrated in the opening area and the middle area of the pleated to agglomerate and deposit.

Prediction of particle saltation velocity for horizontal dilute phase negative pressure pneumatic conveying

The evaluation of particle saltation velocity is vital for the design of dilute negative phase pneumatic conveying systems. Nonetheless, the particle saltation velocity in dilute negative phase pneumatic conveying is largely understudied, it also lacks a consistent correlation over operating conditions and particle properties. As such, this paper investigates the effects of solid-gas ratio, particle size, and drill pipe rotational speed on particle saltation velocity in the drill pipe via CFD-DEM (computational fluid dynamics-discrete element method) simulation. Consequently, the particle saltation velocity increased gradually seemingly leaning toward a constant value with the increase of solid-gas ratio. The particle saltation velocity with larger particle size under similar solid-gas ratio was higher, while the speed of the drill pipe and the particle saltation velocity showed a negative correlation. The correlation for particle saltation velocity was established by dimension analysis method geared toward accurately predicting the particle saltation velocity in dilute negative phase pneumatic conveying. A comparison of the new correlation and the experimental data found that the error between the predicted value of the correlation and the measured particle saltation velocity is within 20% and the correlation derived by the dimensional analysis method give the well agreement.

Super-quadric CFD-DEM simulation of chip-like particles flow in a fluidized bed

Chip-like particles may be practised in fluidized beds in some emerging industries, but its fluidization is not well understood. In this work, the computational fluid dynamics-discrete element method (CFD-DEM) is extended to integrate with a super-quadric model to study the fluidization behaviours of chip-like particles in a three-dimensional bubbling fluidized bed (BFB). After model validation, the typical fluidization behaviours of chip-like particles in the BFB are described in terms of bubble dynamics, pressure signals, particle orientation, particle mixing, and particle dispersion. Particularly, the ordered arrangement of chip-like particles appears at the near-wall region while random arrangement occurs in other regions. The effects of superficial gas velocity and aspect ratio on the mixing and dispersion characteristics of chip-like particles are discussed. The results show that increasing aspect ratio induces a higher mean pressure drop and promotes the initial mixing state. Particle dispersion coefficients range from 10-3 m2/s to 10-2 m2/s, whose vertical component is one order of magnitude larger than the horizontal components. Larger aspect ratios and superficial gas velocities lead to larger particle dispersion coefficients. The fundamental study sheds light on the design and optimization of fluidized beds of chip-like particles including solar panel waste recycling.

POD-based identification approach for powder mixing mechanism in Eulerian–Lagrangian simulations

Numerous products are manufactured through powder mixing. Understanding the mixing mechanism is essential to improve product quality. Convection, diffusion, and shear are well-known classifications in the powder mixing mechanism. During powder mixing, plural mixing mechanisms may occur simultaneously. In this study, to identify the main mixing mechanisms and investigate the transition of the main mixing mechanisms, an advanced identification technique is developed by incorporating the proper orthogonal decomposition (POD) method into numerical modeling for powder mixing. The discrete element method (DEM) coupled with computational fluid dynamics (CFD) is employed to simulate powder mixing. Several investigations are performed to show the adequacy of the developed technique. First, numerous CFD–DEM simulations for solid–liquid flows are performed in a rotating paddle mixer. Next, an efficient Lanczos-based POD (LPOD) method is proposed to characterize the main features of powder mixing via the POD analysis. The results show that the mixing mechanism is dominated by convection in the early stage and by diffusion in the late stage. Besides, a novel mixing identification technique is established by giving the relation between POD modes and mixing mechanisms, namely, clumped and random spatial distributions of the POD modes appear in convective and diffusive mixing, respectively. Consequently, it is shown that combining the CFD–DEM simulation with the LPOD method is effective to identify the main mixing mechanism and to explain the time transition of mixing mechanisms between convective and diffusive mixing.

Effects of dose loading conditions and device geometry on the transport and aerosolization in dry powder inhalers: A simulation study

The transport and aerosolization of particles are studied in several different dry powder inhaler geometries via Computational Fluid Dynamics-Discrete Element Method (CFD-DEM) simulations. These simulations combine Large Eddy Simulation of gas with Discrete Element Model simulation of all the carrier particles and a representative subset of the active pharmaceutical ingredient (API) particles. The purpose of the study is to probe the dominant mechanism leading to the release of the API particles and to demonstrate the value of the CFD-DEM simulations where one tracks the motion of all the carrier and API particles. Simulations are performed at different inhalation rates and initial dose loading conditions for the screen-haler geometry, a simple cylindrical tube inhaler, and five different geometry modifications that took the form of bumpy walls and baffles. These geometry modifications alter the residence time of the powder sample in the inhaler, pressure drop across the inhaler, the severity of gas-carrier interactions, and the number of collisions experienced by the carrier particles, all of which are quantified. The quality of aerosolization is found to correlate with the average air-carrier slip velocity, while collisions played only a secondary role. Some geometry modifications improved aerosolization quality with very little increase in the pressure drop across the device.

Resolved CFD-DEM simulations of the hydraulic conveying of coarse grains through a very-narrow elbow

This paper investigates numerically the hydraulic conveying of solids through a 90° elbow that changes the flow direction from horizontal to vertical, in the very-narrow case where the ratio of pipe to particle diameters is less than 5. We performed resolved CFD-DEM (computational fluid dynamics - discrete element method) computations, in which we made use of the IB (immersed boundary) method of the open-source code CFDEM. We investigate the effects of the water flow and particle injection rate on the transport rate and sedimentation by tracking the granular structures appearing in the pipe, the motion of individual particles, and the contact network of settled particles. We found the saturated transport rate for each water velocity and that a large number of particles settle in the elbow region for smaller velocities, forming a crystal-like lattice that persists in time, and we propose a procedure to mitigate the problem.