Search

Combining the Discrete Ordinates method with the so-called blocked-off approach to calculate radiative heat transfer is investigated for a prospective application in DEM/CFD simulations. Therefore, a single cubic DEM object is located within a prismatic enclosure. The temperature of all walls of the enclosure and of the surface of the DEM object are set to fixed values. The three-dimensional radiative transport is calculated and the heat transfer onto the individual surfaces of the DEM object is determined. A conventional CFD simulation employing a body-conformal mesh at the surface of the embedded DEM object, serves as a reference, an "exact" solution. The results obtained are compared to corresponding solutions with a blocked-off approach, where the shape of the DEM object and the required boundary conditions are enforced within the discretized intensity field while suppressing the solution process in the region obstructed by the object. Mesh-resolution and orientation of the cube in the enclosure are varied and two computationally cheap methods to obtain the radiative fluxes onto the embedded cube surfaces are employed. The heat fluxes onto the cube's surfaces are generally overestimated. This is mainly due to the "artificial" increase of the cube's surface area by the blocked-off approach. The introduction of a simple surface area scaling factor improves the results significantly. The deviation in radiative heat from the exact solution depends on the cube's orientation in the prismatic box and on the position of the actual surface relative to the closest mesh line. The approximation deviates about 1.4% to 3.1% in radiative heat to the cube, values which seem to be acceptable. [Display omitted] • 3D CFD simulations of radiative heat transfer onto a cubic DEM object with the DOM. • Using the blocked-off approach for prospective application in DEM/CFD simulations. • Comparison between the blocked-off approach and a body-conformal mesh solution. • Application of a surface area ratio scaling improves the accuracy of the simulation. • The deviations for the blocked-off approach are in the range of 1.4% to 14.0%. [ABSTRACT FROM AUTHOR]

The mixing process greatly influence the mixing efficiency, as well as the quality and the price of the intermediate and/or the final product. Static mixer is used for premixing action before the main mixing process, for significant reduction of mixing time and energy consumption. This type of premixing action is not investigated in detail in the open literature. In this article, the novel numerical approach called Discrete Element Method is used for modelling of granular flow in multiple static mixer applications (1 - 3 Komax or Ross mixing elements were utilized), while the Computational Fluid Dynamic method was chosen for fluid flow modelling, using the Eulerian multiphase model. The main aim of this article is to predict the behaviour of granules being gravitationally transported in different mixer configuration and to choose the best configuration of the mixer taking into account the total particle path, the number of mixing elements and the quality of the obtained mixture. The results of the numerical simulations in the static mixers were compared to experimental results, the mixing quality is examined by RSD (relative standard deviation) criterion, and the effects on the mixer type and the number of mixing elements on mixing process were studied. The effects of the mixer type and the number of mixing elements on mixing process were studied using analysis of variance (ANOVA). Mathematical modelling is used for optimization of number of Ross and Komax segments in mixer in order to gain desirable mixing results. [Projekat Ministarstva nauke Republike Srbije, br. TR31055]

Heat transfer in packed or fluidized beds in the presence of a surrounding fluid is an important phenomenon which is relevant to numerous industrial applications. Here we extend an earlier derived 3D heat transfer model (Oschmann et al. in Powder Technol 291:392-407, 2016) to take into account particle-fluid heat convection in the case of Biot numbers $$Bi\gg 1$$ . The Discrete Element Method (DEM) which is coupled with the commercial Computational Fluid Dynamics (CFD) package ANSYS Fluent is used as the modelling framework. As a first approximation of the flow induced inhomogeneity of the local heat transfer on the particle surface a distribution function is employed. To validate the resolved heat transfer model, we compare DEM/CFD simulations of three different materials (wood, Polyoxymethylene (POM) and aluminum) with performed experiments. This firstly includes cases where particle surface temperatures are compared with measurements of an infrared camera. Secondly, a numerical study of the average bed temperatures of particle core and surface is conducted to show the differences of the used materials. Thirdly, the core temperatures of three selected particles are compared against experiments. The DEM/CFD framework provides an accurate description of the temperature evolution where the wall effects are negligible. Close to the walls a qualitative agreement can only be achieved for materials with low thermal conductivities. As a consequence of this, in the second part of our investigation we provide various CFD simulations for the heating of an aluminum oxide wall which is required for the evaluation of the particle surface temperatures measured by an infrared camera. The simulation results show the same tendencies as the experiments, underline the complexity of the heat transfer at the walls and are a first step for the formulation of a complex particle-wall heat transfer model in the context of a DEM/CFD framework. [ABSTRACT FROM AUTHOR]

DEM/CFD (Discrete Element Method/Computational Fluid Dynamic) approach was used to develop a three dimensional model of fluid flow and mixing process of solid particles within the revolving static mixer, type Komax. Static mixers are widely used in process industry, food industry, pharmaceutical and chemical industry. Mixing process adds significant value to the final product and it can be regarded as a key process. The quality and the price of the product often depend on mixing efficiency. Both design and operation of the mixing unit itself have a strong influence on the quality produced. In this paper, DEM is used for modeling granular flow of zeolite spheres in the revolving Komax static mixer and CFD method was used for modeling fluid flow through the Eulerian multiphase model. Coupling of these two methods provides reliable analysis of particle motion and flow pattern of the working fluid. The objective of this paper is to predict the behavior of particles after several rotations of the revolving static mixer. This type of device is used for premixing action before the main mixing process, for significant reduction of mixing time and energy consumption. This type of premixing action is not investigated in detail in the open literature. The results of the numerical simulation are compared with appropriate experimental results. The special design revolving static mixer, made of transparent Plexiglas was used for this experiment. Mixing quality was examined by RSD (relative standard deviation) criterion. Application of this model provides the optimization of the number of revolutions and geometrical parameters of mixing process taking into account the quality of the mixing process. [ABSTRACT FROM AUTHOR]

Mixing of powders, particles and/or granules obtains an economic importance in various technologies, and in the food, pharmaceutical and chemical industries. In many cases, the mixing process greatly influences the quality of the final product. The parameters of the mixing process and the design of equipment have a strong impact on the mixing efficiency, and the quality and the price of the intermediate or final product. In this study, the discrete element method (DEM) is used for modeling of granular flow in various multiple Komax and Ross mixing applications. Computational fluid dynamic (CFD) method was used for modeling of fluid flow through the Eulerian multiphase model. Coupling of these two methods provides reliable, sufficiently correct and adequate results of the proposed model compared to experimental measurements. The objective of this paper is to predict the behavior of particles in different static mixer configurations. The results of the numerical simulation are compared with appropriate experimental results. Application of this model provides the optimization of the geometry and parameters of mixing systems taking into account the quality of the mixing process and the cost of the final product. [ABSTRACT FROM AUTHOR]

The combined "Blocked-Off & Discrete Ordinates" (BO/DO) approach is used within a DEM/CFD frame to model the radiative heating of stationary and moving (rotating with respect to the mesh) cubic particles. A single cube, placed inside a larger cubic box, is heated by radiation and its temperature evolution over time is investigated. Two different models for the particle heating are considered: the first one assumes a uniform cube temperature while the second model calculates the intra-particle heat transfer leading to a non-uniform temperature distribution as it occurs in thermally thick particles. Three different particle materials, wood, styrofoam and steel, are considered. Initially, the radiative heating of stationary cubes, approximated by the BO approach in a CFD mesh, is compared to the heating of cubes enclosed by a corresponding body conformal mesh as a reference. The results show that deviations in cube temperature mainly result from the artificial increase of the particle surface in the BO. When applying a particle surface area ratio to compensate for the artificial surface area increase deviations in temperature can be reduced to 5.2% compared to body conformal mesh. In a further step, the cube is constantly rotating which changes the orientation of the cube's surfaces relative to the radiation source. The evolution of the cube's surface temperatures, which is dependent on the orientation, is modelled by the BO/DO approach. As expected, materials with low thermal conductivity show larger temperature gradients over time than materials with high thermal conductivity. The temperature difference between different surfaces turns out to be in the ranges 50/14/0 K (for styrofoam/wood/steel) and 134/65/0 K between surface and core, thus reflecting the correct physical behavior. The evaluation of the BO/DO approach concerning its capability to account for radiative heat transfer in DEM simulations for particles with a regular (cubic) geometry, as examined in the current paper, forms the basis to extend the description to arbitrarily shaped particles, a great potential of the BO/DO approach as will be shown in an outlook example. [Display omitted] • Transient simulations of radiative heating of moving, complex particles in DEM/CFD. • Using the Blocked-Off approach in combination with the DO model. • Due to the BO approach, temperature deviates by up to 5.2%. • Particle rotation has a large impact on the heating of the particle. • Low thermal conductivity materials have large intra-particle temperature gradients. [ABSTRACT FROM AUTHOR]

Abstract: The flow and segregation behaviours of the binary mixtures with different particle sizes during die filling in air and in a vacuum were investigated using a coupled discrete element method (DEM) and computational fluid dynamics (CFD) in three dimensions (3D). Two types of computational setups were considered: i) a fully 3D model with real wall boundaries and ii) a thin sliced model with a slice of the full domain using two parallel periodic boundaries. Die filling with a stationary shoe and a moving shoe were simulated. It was found that the flow and segregation results obtained using the fully 3D and thin sliced models were essentially identical, implying that the thin sliced model can be used to simulate the die filling with a reduced computational time and good accuracy. It was also observed that, for die filling with a stationary shoe, the powder flow rate was reduced by the entrapped air in the die. For the die filling in a vacuum, vertical segregation occurred with a high concentration of fine particles at the bottom of the die as the fine particles can sift through the voids between coarse particles. The presence of air reduced the extent of this segregation by suppressing the percolation of fines through the voids. For die filling with a moving shoe, the effect of air on powder flow and segregation behaviours is negligible when the process is dominated by nose flow as considered in this study, since the air can readily escape from the die cavity. Segregation was also induced during die filling with a moving shoe, due to the free surface segregation and the filtration of fines through the voids between the coarse particles. Consequently, a high overall concentration of fines in the die was obtained, while the concentration of fines at the far end of the die was much lower than that at the near end. [ABSTRACT FROM AUTHOR]

Abstract: In this study, segregation behaviour of binary granular mixtures with the same particle size but different densities during die filling in the presence of air was investigated using a combined discrete element method (DEM) and computational fluid dynamics (CFD) approach, in which the kinematics of particles was modelled using DEM, the motion of air was analysed using CFD and a two-way coupling of the particles and the air was incorporated. The depositions of powder from stationary and moving shoes into the die cavities of different geometries were simulated and the corresponding segregation behaviours were analysed. It has been found that, for die filling from a stationary shoe, the concentration distributions of the heavy and light particles along the die width mainly depend on the initial spatial distribution of the granular mixture in the shoe. For die filling from a moving shoe, a low concentration of light particles on the leading side of the die (referring to the direction of the shoe motion) is observed for die filling with a square die, in which the process is dominated by nose flow. The density difference can cause segregation along the die depth with a low concentration of light particles at the bottom. The presence of air enhances this segregation tendency by resisting the flow of light particles into the bottom of the die and causes a higher concentration of the light particles at the top. Finally, the segregation index, defined as the volume weighted root-mean-square deviation in the content of light particles, was introduced to quantify the degree of segregation in the horizontal and vertical directions. It has been found that the degree of segregation is determined by the presence of air and also the powder flow pattern. [Copyright &y& Elsevier]