3rd Joint US-European Fluids Engineering Summer Meeting

2010;():1-7. doi:10.1115/FEDSM-ICNMM2010-30262.

While there are several best practice standards available for minimizing the energy requirement for compressed air use in an industrial context, moving to best practice often requires investment and operational change. In production facilities, there is often a reluctance to commit to this type of change without a clear view of the benefit. Furthermore, there is very little detailed information available in the open literature that allows even a qualitative assessment of priorities. In order to address this shortcoming, analyses of two industrial compressed air systems which are already installed in manufacturing plants have been conducted in the context of energy usage. The installations are quite different in compressed air needs: one is focused on actuation and drying; while the other uses compressed air primarily for material handling. In both sites, the energy of the compressed air is evaluated at each key element of the system and the typical end use application profile is assessed. Simple models of the consumption rates are used to relate duty cycle and device count with actual total consumption. A new way of assessing the leak rate from the entire system has been developed, based on the pressure decay time, and has been implemented at one site. In this way, the energy balance of the system entire has been analyzed quantitatively, with the effect of distribution leaks accounted for directly. It is found that in both sites, open blowing operations (e.g. drying) are the largest, consumers which are amenable to optimization. It is also found that the measured leak rate at one site represented 23% of the compressed air generated, with an energy input of 455kWh per day. It is concluded that this approach can help to identify priorities for optimizing CA use at an industrial site.

Commentary by Dr. Valentin Fuster
2010;():9-22. doi:10.1115/FEDSM-ICNMM2010-30869.

A new concept of hydrokinetic turbine using oscillating hydrofoils to extract energy from water currents (tidal or gravitational) is presented, tested and analyzed in the present investigation. Due to its rectangular extraction plane, this technology is particularly well suited for river beds and shallow waters near the coasts. The present turbine is a 2 kW prototype, composed of two rectangular oscillating hydrofoils of aspect ratio 7 in a tandem spatial configuration. The pitching motion of each hydrofoil is coupled to their cyclic heaving motion through four-link mechanisms which effectively yield a one-degree-of-freedom system driving a speed-controlled electric generator. The turbine has been mounted on a custom-made pontoon boat and dragged on a lake at different velocities. Instantaneous extracted power has been measured and cycle-averaged for several water flow velocities and hydrofoil oscillation frequencies. Results are demonstrated to be self-consistent and validate our extensive 2D flow simulation database. The present data show optimal performances of the oscillating hydrofoils concept at a reduced frequency of about 0.12, at which condition the measured power extraction efficiency reaches 40% once the overall losses in the mechanical system are taken into account. Further measurements of power extraction with a single oscillating hydrofoil have also been performed by taking out the downstream hydrofoil of the tandem pair. Those measurements favorably compare, quantitatively, with available 3D CFD predictions. The 40% hydrodynamic efficiency of this first prototype exceeds expectation and reaches levels comparable to the best performances achievable with modern rotor-blades turbines. It thus demonstrates the promising potential of the oscillating hydrofoils technology to efficiently extract power from an incoming water flow.

Topics: Testing , Turbines , Hydrofoil
Commentary by Dr. Valentin Fuster
2010;():23-31. doi:10.1115/FEDSM-ICNMM2010-31229.

Every Francis turbine has a thin gap between rotating and non-rotating parts, which prevents contact between the two units. Although necessary, hydraulic seals create energetic losses: some fluid does not flow through the runner (leakage loss) and exerts a torque on the rotor (friction loss). Only analytical and empirical prediction methods of a seal efficiency had been developed before 1980. Numerical methods are now used to predict seals performance. However, most of the studies known to the authors deal with gas labyrinth seals and use the k–ε turbulence model. In hydraulic seals, since the viscous losses in the boundary layer influence the leakage loss, low Reynolds turbulence models appear more appropriate. Our study aims to implement an accurate model to predict losses in labyrinth seals using a low Reynolds model, and validate it using experimental results. The issues of the mesh and boundary conditions are addressed. The commercial code ANSYS CFX 12 is used.

Commentary by Dr. Valentin Fuster
2010;():33-41. doi:10.1115/FEDSM-ICNMM2010-30068.

Three-phase gas-liquid-particle flows under microgravity condition were numerically studied. An Eulerian-Lagrangian computational model was used in the simulations. In this approach, the liquid flow was modeled by a volume-averaged system of governing equations, whereas motions of particles and bubbles were evaluated using the Lagrangian trajectory analysis procedure. It was assumed that bubble shape variations were neglected and the bubbles remained spherical. The bubble-liquid, particle-liquid and bubble-particle interactions were accounted for in the analysis. The discrete phase equations included drag, lift, buoyancy, and virtual mass forces. Particle-particle interactions and bubble-bubble interactions were accounted for by the hard sphere model. Bubble coalescence was also included in the model. The transient flow characteristics of the three-phase flow were studied. The effects of gravity and g-jitter acceleration on variation of flow characteristics were discussed. The low gravity simulations showed that most bubbles are aggregated in the inlet region and the bubble plume exhibits a plug type flow behavior. The particles are mainly located outside the bubble plume, with very few particles being retained in the plume. Compared to the normal gravity condition, the three phases in the column are poorly mixed under microgravity conditions. The velocities of the three phases were also found to be of the same order. The simulation results showed that the effect of g-jitter acceleration on the gas-liquid-particle three phase flows is small.

Commentary by Dr. Valentin Fuster
2010;():43-50. doi:10.1115/FEDSM-ICNMM2010-30069.

Rolling detachment of micro particles in turbulent flows under the presence of electrostatic and capillary forces was studied. The maximum adhesion resistance model and the effective thermodynamic work of adhesion including the effects of electrostatic and capillary forces were used in the analysis. The JKR and DMT models for elastic interface deformations and the Maugis-Pollock model for the plastic deformation were extended to include the effect of electrostatic and capillary forces. The turbulence burst model was used to evaluate the airflow velocity near the substrate. The critical shear velocities for removal of particles of different sizes were evaluated and the results were compared with those without electrostatic and capillary forces. The relative critical shear velocities as well as the material dependence were also studied. The effect of the direction of the combined Coulomb force was also included. The predictions of the electric detachment fields for particles were compared with the available experimental data and good agreement was observed.

Commentary by Dr. Valentin Fuster
2010;():51-56. doi:10.1115/FEDSM-ICNMM2010-30250.

This paper presented a numerical study that predicts critical mass flow rate, pressure, vapor quality, and void fraction along a very long tube with small diameter or capillary tub under critical condition by the drift flux model. Capillary tubes are simple expansion devices and are necessary to design and optimization of refrigeration systems. Using dimensional analysis by Buckingham’s π theory, some generalized correlations are proposed for prediction of flow parameters as functions of flow properties and tube sizes under various critical conditions. This study is performed under the inlet pressure in the range of 0.8 ≤ pin ≤ 1.5Mpa, subcooling temperature between 0 ≤ ΔTsub ≤ 10 °C. The tube diameter is in the range of 0.5 ≤ D ≤ 1.5mm and tube length between 1 ≤ L ≤ 2m for water, ammonia, refrigerants R-12, R-22 and R-134 as working fluids. Comparison between the results of the present work and some experimental data indicates a good agreement. Cluster of data close to the fitted curves also shows satisfactory results.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2010;():57-62. doi:10.1115/FEDSM-ICNMM2010-30268.

The process of particle-wall collisions is very important in understanding and determining the fluid-particle behavior, especially near walls. Detailed information on particle-wall collisions can provide insight on the formulation of appropriate boundary conditions of the particulate phases in two-fluid models. We have developed a three-dimensional Resolved Discrete Particle Method (RDPM) that is capable of meaningfully handling particle-wall collisions in a viscous fluid. This numerical method makes use of a Finite-Difference method in combination with the Immersed Boundary (IB) method for treating the particulate phase. A regular Eulerian grid is used to solve the modified momentum equations in the entire flow region. In the region that is occupied by the solid particles, a second particle-based Lagrangian grid is used, and a force density function is introduced to represent the momentum interactions between particle and fluid. We have used this numerical method to study both the central and oblique impact of a spherical particle with a wall in a viscous fluid. The particles are allowed to move in the fluid until they collide with the solid wall. The collision force on the particle is modeled by a soft-sphere collision scheme with a linear spring-dashpot system. The hydrodynamic force on the particle is solved directly from the RDPM. By following the trajectories of a particle, we investigate the effect of the collision model parameters to the dynamics of a particle close to the wall. We report in this paper the rebound velocity of the particle, the coefficient of restitution, and the particle slip velocity at the wall when a variety of different soft-sphere collision parameters are used.

Commentary by Dr. Valentin Fuster
2010;():63-71. doi:10.1115/FEDSM-ICNMM2010-30573.

Plunging liquid jets are commonly encountered in nature and are widely used in industrial applications (e.g., in waterfalls, waste-water treatment, the oxygenation of chemical liquids, etc.). Despite numerous experimental studies that have been devoted to this interesting problem, there have been very few two-phase flow simulations. The main difficulty is the lack of a quantitative model to simulate the air entrainment process, which plays a critical role in this problem. In this paper, we present a computational multiphase fluid dynamics (CMFD) approach for solving this problem. The main ingredients of this approach are a comprehensive subgrid air entrainment model that predicts the rate and location of the air entrainment and a two-fluid transport model in which bubbles of different sizes are modeled as a continuum fluid. Using this approach, a Reynolds-averaged Navier Stokes (RaNS) two-way coupled two-phase flow simulation of a plunging liquid jet with a diameter of 24mm and a liquid jet velocity around 3.5m/s was performed. We analyzed the simulated void fraction and bubble count rate profiles at three different depths beneath the average free surface, and compared them with experimental data. We observed good agreement at all locations.

Commentary by Dr. Valentin Fuster
2010;():73-80. doi:10.1115/FEDSM-ICNMM2010-30574.

To simulate the initial formation of sedimentary bedforms, constrained to be in hydraulically smooth turbulent flows under bedload conditions, a numerical model based on Large Eddy Simulation (LES) in a doubly periodic domain has been developed. The numerical model comprises three parts. Given the instantaneous bed geometry, the bed shear stress distribution is obtained from a Large-Eddy-Simulation (LES) method coupled with an Immersed-Boundary-Method (IBM). Flux is estimated by the van Rijn’s formula [1]. Finally, evolution of the bed surface is described by the Exner equation. “Two-dimensional bed” [2] and “three-dimensional bed” models employ, respectively, transversely averaged bed shear stress and instantaneous local shear stress to estimate the bedload flux. Based on this model, the evolution of an initial sand wave has been successfully computed. Compared to the “two-dimensional” [2] model, the three-dimensional model leads to a slightly slower propagation and a smaller sand wave. The tendency of the sand wave evolution in three-dimensional model is two-dimensional during the simulated interval.

Topics: Sands , Turbulence
Commentary by Dr. Valentin Fuster
2010;():81-88. doi:10.1115/FEDSM-ICNMM2010-31032.

Understanding the complex interaction of droplet dynamics with mass transfer and chemical reactions is of fundamental importance in liquid-liquid extraction. High-fidelity numerical simulation of droplet dynamics with interfacial mass transfer is particularly challenging because the position of the interface between the fluids and the interface physics need to be predicted as part of the solution of the flow equations. In addition, the discontinuity in fluid density, viscosity and species concentration at the interface present additional numerical challenges. In this work, we extend our balanced-force volume-tracking algorithm for modeling surface tension force (Francois et al., 2006) and we propose a global embedded interface formulation to model the interfacial conditions of an interface in thermodynamic equilibrium. To validate our formulation, we perform simulations of pure diffusion problems in one- and two-dimensions. Then we present two and three-dimensional simulations of a single droplet dynamics rising by buoyancy with mass transfer.

Commentary by Dr. Valentin Fuster
2010;():89-97. doi:10.1115/FEDSM-ICNMM2010-31106.

Fluidized beds are being used in practice to gasify biomass to create producer gas, a flammable gas that can be used for process heating. However, recent literature has identified the need to better understand and characterize biomass fluidization hydrodynamics, and computational fluid dynamics (CFD) is one approach in this effort. Previous work by the authors considered the validity of using two-dimensional versus three-dimensional simulations to model a cold-flow fluidizing biomass bed configured with a single side port air injection. The side port is introduced to inject air and promote mixing within the bed. Comparisons with experiments indicated that three-dimensional simulations were necessary to capture the fluidization behavior for the more complex geometry. This paper considers the effects of increasing fluidization air flow and side port air flow on the homogeneity of the bed material in a 10.2 cm diameter fluidized bed. Two air injection ports diametrically opposed to each other are also considered to determine their effects on fluidization hydrodynamics. Whenever possible, the simulations are compared to experimental data of time-averaged local gas holdup obtained using X-ray computed tomography. This study will show that increasing the fluidization and side port air flows contribute to a more homogeneous bed. Furthermore, the introduction of two side ports results in a more symmetric gas-solid distribution.

Topics: Biomass
Commentary by Dr. Valentin Fuster
2010;():99-106. doi:10.1115/FEDSM-ICNMM2010-31156.

A novel methodology is presented for the numerical treatment of multi-dimensional pdf (probability density function) models used to study particle transport in turbulent boundary layers. A system of coupled Fokker-Planck type equations is constructed to describe the transport of phase-space conditioned moments of particle and fluid velocities, both streamwise and wall-normal. Unlike conventional moment-based transport equations this system allows for an exact treatment of particle deposition at the flow boundary. Moreover, the equations in the system are linear and can be solved in a sequential fashion; there is no closure problem to address. A Hermite-Discontinuous Galerkin scheme is employed to treat the system. The choice of Hermite basis functions in combination with an iterative rescaling approach, allows for efficient discretization of the, effectively, 5-dimensional phase-space domain. Results demonstrate the effectiveness of the methodology in resolving distributions near an absorbing boundary.

Commentary by Dr. Valentin Fuster
2010;():107-115. doi:10.1115/FEDSM-ICNMM2010-31249.

Gas-liquid two-phase flows on the wall like liquid film flows, which are the so-called wetted wall flows, are observed in many industrial processes such as absorption, desorption, distillation and others. For the optimum design of packed columns widely used in those kind of processes, the accurate predictions of the details on the wetted wall flow behavior in packing elements are important, especially in order to enhance the mass transfer between the gas and liquid and to prevent flooding and channeling of the liquid flow. The present study focused on the effects of the change of liquid flow rate and the wall surface texture treatments on the characteristics of wetted wall flows which have the drastic flow transition between the film flow and rivulet flow. In this paper, the three-dimensional gas-liquid two-phase flow simulation by using the volume of fluid (VOF) model is applied into wetted wall flows. Firstly, as one of new interesting findings in this paper, present results showed that the hysteresis of the flow transition between the film flow and rivulet flow arose against the increasing or decreasing stages of the liquid flow rate. It was supposed that this transition phenomenon depends on the history of flow pattern as the change of curvature of interphase surface which leads to the surface tension. Additionally, the applicability and accuracy of the present numerical simulation were validated by using the existing experimental and theoretical studies with smooth wall surface. Secondary, referring to the texture geometry used in an industrial packing element, the present simulations showed that surface texture treatments added on the wall can improve the prevention of liquid channeling and can increase the wetted area.

Commentary by Dr. Valentin Fuster
2010;():117-125. doi:10.1115/FEDSM-ICNMM2010-30002.

An optimization strategy called response surface methodology (RSM) is applied to a centrifugal fan impeller optimization design in this paper. RSM is used to generate an approximated model of objective function, for which a second-order polynomial function is chosen. The Design of experiment (DOE) technique coupled with CFD analysis is then ran to generate the database. The least-squares regression method (LS) is used to determine the coefficient of the RSM function. Finally, the Genetic Algorithms (GA) is applied to the objective function in order to obtain the optimal configuration. This paper also presents a solution to the problem of imprecise fitting of second-order RSM model by dividing the zone into several subzones which is proved to be effective in this paper. The optimization result shows that RSM is an effective and feasible optimization strategy for the centrifugal fan impeller design, and the complexity of the objective function and the overall optimization time could be significantly reduced.