0

ASME Conference Presenter Attendance Policy and Archival Proceedings

2014;():V01DT00A001. doi:10.1115/FEDSM2014-NS1D.
FREE TO VIEW

This online compilation of papers from the ASME 2014 4th Joint US-European Fluids Engineering Division Summer Meeting (FEDSM2014) represents the archival version of the Conference Proceedings. According to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

7th Symposium on Transport Phenomena in Mixing

2014;():V01DT26A001. doi:10.1115/FEDSM2014-21048.

Granular mixing processes are commonly used to increase product homogeneity in many industrial applications involving pharmaceuticals, food processing, and energy conversion. Determining the appropriate granular mixing length is necessary to avoid over/under mixing and unnecessary power consumption. The goal of this study is to experimentally characterize the granular mixing process and determine, under various operating conditions, the needed mixing length to achieve adequate mixing in a laboratory-scale double screw mixer. Nine different combinations of screw rotation speeds and dimensionless screw pitches are used to investigate the rate of mixing at dimensionless mixing lengths of L/D = 2, 5, and 10. Composition and statistical analysis methods are employed to assess mixing effectiveness, and it is determined that the dimensionless mixing length is the most influential parameter in terms increasing granular homogeneity. For all the conditions tested, the granular mixture approaches an acceptable level of mixing for all testing conditions when the dimensionless mixing length is L/D = 10. However, the segregation rate throughout the screw mixer is vastly different for various combinations of screw rotation speed and dimensionless screw pitch, and is partly attributed to the influence of entrance effects caused by the material injection process.

Topics: Screws
Commentary by Dr. Valentin Fuster
2014;():V01DT26A002. doi:10.1115/FEDSM2014-21354.

Numerical results depicting the effects of diffusers on confined isothermal high-swirl jets are presented. The aim is analyze the mixing between a non-swirling inner jet (natural gas) and a swirling annular jet (primary air). This simple setup is widely used in burners to promote stabilized flames of lean mixtures.

Flow patterns for sudden-expansion and diffusers are contrasted for swirl number of 1 and expansion area ratio of 4 in a transitional turbulence regime. Diffusers have important influence on size and location of recirculation zones. Hence knowledge of flow characteristics is a prerequisite in the design process. The criteria to establish the optimum diffuser would be better mixing as well as minimum residence time in the recirculation zones to prevent the formation of NOx in the future burner. The dissipated mechanical energy is not important in burner applications.

Commentary by Dr. Valentin Fuster
2014;():V01DT26A003. doi:10.1115/FEDSM2014-21871.

The mixing efficiency, flux Richardson number Rif, is investigated in a horizontally injected turbulent stratified jet with a co-existence of stable and unstable stratifications. The high resolution experimental data from a developed laser-based technique show the statistical relationship between Rif and the gradient Richardson number Rig. In addition, the data are used to study the development of entrainment by two approaches, and compared with theoretical predictions.

Topics: Turbulence
Commentary by Dr. Valentin Fuster
2014;():V01DT26A004. doi:10.1115/FEDSM2014-21945.

In this study, an Eulerian-Lagrangian computational methodology is utilized for large eddy simulation (LES) of mixing phenomena in jet in cross-flows. A high-order multi-block algorithm is used to solve Eulerian equations in a generalized coordinate system. The composition is formulated based on the filtered mass density function (FMDF) and its equivalent stochastic Lagrangian equations, which is solved by Lagrangian Monte-Carlo method. Parameters influencing mixing enhancement including jet velocity profile, and jet pulsation are investigated. A good consistency between Eulerian and Lagrangian components of the numerical scheme is established. In jet in cross-flow (JICF) simulations, the vortical structures and flow features are predicted with the current numerical scheme. The results also show that the jet velocity profile affects both trajectory and mixing condition and the jet pulsation can enhance mixing depending on the Strouhal numbers. The obtained results including concentration distributions are in good agreement with available experimental data ensuring the performance and reliability of LES/FMDF methodology to study mixing in relatively complex flow configurations such as JICF.

Commentary by Dr. Valentin Fuster
2014;():V01DT26A005. doi:10.1115/FEDSM2014-21967.

The flow in a New Brunswick Scientific (NBS (now Eppendorf)) 5 L stirred-tank bioreactor (STR) partially filled with 2.2 L of water and agitated at 60 rpm using a pitched-blade impeller is studied in this work, to determine the suitability of the configuration for expanding stem cell lines. Computational Fluid Dynamics (CFD) model development and testing in this work has found Large Eddy Simulation (LES) to be essential for model fidelity and for capturing spatiotemporal stress fluctuations. Stresses were at levels similar to or even higher than those known to damage stem cells or modulate their cellular function to favour differentiation instead of phenotype maintenance. The results raise questions as to the appropriateness of such STRs for stem cell expansion, and motivate better experimental studies to properly quantify the spatiotemporal variability in fluid shear stresses and its effect on stem cell expansion and stem cell fate.

Commentary by Dr. Valentin Fuster
2014;():V01DT26A006. doi:10.1115/FEDSM2014-21976.

Flash Nanoprecipitation (FNP) is a technique to produce monodisperse functional nanoparticles through rapidly mixing a saturated solution and a non-solvent. Multi-inlet vortex reactors (MIVR) have been effectively applied to FNP due to their ability to provide both rapid mixing and the flexibility of inlet flow conditions. Until recently, only micro-scale MIVRs have been demonstrated to be effective in FNP. A scaled-up MIVR could potentially generate large quantities of functional nanoparticles, giving FNP wider applicability in the industry. In the present research, turbulent mixing inside a scaled-up, macro-scale MIVR was measured by stereoscopic particle image velocimetry (SPIV). Reynolds number of this reactor is defined based on the bulk inlet velocity, ranging from 3290 to 8225. It is the first time that the three-dimensional velocity field of a MIVR was experimentally measured. The influence of Reynolds number on mean velocity becomes more linear as Reynolds number increases. An analytical vortex model was proposed to well describe the mean velocity profile. The turbulent characteristics such as turbulent kinematic energy and Reynolds stress are also presented. The wandering motion of vortex center was found to have a significant contribution to the turbulent kinetic energy of flow near the center area.

Topics: Turbulence , Vortices
Commentary by Dr. Valentin Fuster
2014;():V01DT26A007. doi:10.1115/FEDSM2014-22034.

In contrast to mixing in vertical tanks, jet mixing in long horizontal tanks is scarcely investigated in the literature. It is known that jet mixing in long horizontal tanks is more difficult when compared to short tanks, as more liquid volume must be recirculated through the jets.

In this study, computational fluid dynamics (CFD) simulations are conducted for the flow in a horizontal cylindrical tank with a length-to-diameter ratio of 3:1 and a nominal volume of 112,560 L (liquid volume of 75,708 L, i.e. 20,000 gallons). A pair of back-to-back Coldrey nozzles is placed near the center of the tank bottom, and each nozzle directs its jet towards the corresponding vessel end. An intriguing phenomenon is observed in the transient simulations, where the turbulent jets oscillate in both horizontal and vertical directions with a low frequency. In order to determine the source of such oscillation, a number of simulations are conducted to explore the effects of mesh type and size, boundary condition on the free surface, turbulence model, and time step. Oscillation persists in all cases, indicating that it is unlikely the result of some numerical instability. The oscillation also appears to be insensitive to the Reynolds number or symmetry in the nozzle or tank geometry. Another simulation with a single jet also shows the oscillatory flow behavior, and thus the oscillation is more likely to be caused by interaction between the jet and the recirculating flow in the tank, rather than interaction between the two jets. Further analysis of the jet velocity profile suggests that the secondary flow on the cross section of the jets might also contribute to the oscillation.

While similar confined jet oscillations due to Coanda effect and blind cavity effect have been previously observed in small cavities by both 2D numerical simulations and laboratory scale experiments, this study shows that such oscillation also exists in industrial scale horizontal tanks. The oscillatory motion of the liquid may lead to improved mixing in the tank.

Topics: Oscillations , Jets
Commentary by Dr. Valentin Fuster
2014;():V01DT26A008. doi:10.1115/FEDSM2014-22038.

This paper presents a numerical study of a turbulent acetone spray flow, where the gas phase model includes a transported joint probability density function (PDF) of the gas phase velocity and the mixture fraction. This approach greatly benefits from the fact that the spray evaporation rate appears in closed form, and no additional modeling is required, whereas the molecular mixing requires closure. This is achieved through use of the extended interaction-by-exchange-with-the-mean (IEM) model with an additional term to account for spray evaporation. The dispersed liquid phase is described through a Lagrangian discrete parcel method with a point-source approximation. For droplet evaporation, an equilibrium model is compared with a more advanced non-equilibrium model. Numerical results of droplet size as well as mean and fluctuating velocities are presented and discussed in comparison with experimental results from the literature, and good agreement is observed. The non-equilibrium model predicts somewhat slower spray evaporation compared with the equilibrium model.

Commentary by Dr. Valentin Fuster

5th International Symposium on Turbulent Flows: Issues and Perspectives

2014;():V01DT27A001. doi:10.1115/FEDSM2014-21026.

Experiments were carried out to study the flow and mixing behavior in a smooth lid-driven cavity. In contrast to a simple lid-driven cavity configuration, narrow gaps with finite thickness at inlet and outlet and an additional jet into the cavity were considered, too. It was found that a thin shear layer close to the moving wall occurred whereas the larger part of the flow domain was characterized by large fluctuating eddies. Due to the smooth cavity shape, a pressure gradient in flow direction resulted. The combined effects of that pressure gradient and the moving wall were investigated by measurements of the velocity and fluctuation profiles using Laser-Doppler-Anemometry (LDA). Special attention was spent to the interaction of the inflowing jet with the main flow within the cavity, and different regimes of jet-cavity-interaction were visualized. In addition to the experiments, an extensive Large-Eddy-Simulation (LES) study was conducted, and a reasonable agreement between the experimental data and the LES results were found.

Topics: Turbulence , Cavities
Commentary by Dr. Valentin Fuster
2014;():V01DT27A002. doi:10.1115/FEDSM2014-21054.

A square duct with a 90-degree streamwise curvature is representative of complex flow domains. Such flow domains are encountered in the designs of fluids engineering systems, especially in the aerospace turbo-machinery components. Examples include the gas turbine engine axial compressor interstage spaces, where the rise in air pressure (and hence compressor efficiency) is dependent on suppression of turbulence. In the case of the centrifugal compressor, pressure rise in the U-shaped diffuser assembly where the suppression of turbulence is critical to the attainable pressure ratio. The results obtained from numerical calculations are analysed and discussed along with the corresponding hot-wire measurements and flow visualization result from a wind-tunnel of identical configuration. Calculations are implemented in four turbulent models, i.e. Standard k-e Module, Algebraic Stress Model (ASM), Non-linear Renormalization Group (RNG) - k-e Model and Differential Stress Model (DSM). The discretization up-winding scheme is the Quadratic Up-winding with Interpolation Kinematics (QUICK). Two high Reynolds number turbulent flows are investigated, with mainstream velocities of 12.3 m/s and 20.4 m/s, representing Re=3.56×105 and Re=6.43×105 respectively. Generally strong correlation between theory and experimental data are recorded. Further, as reported in similar studies, the turbulence modules that are formulated to account for turbulence anisotropy return results that more closely match experimental measurements. Uniquely for this configuration, a massive flow detachment is predicted along the convex wall at about the 90° position. Also, the core of the fluid flow is observed to shift from the outer to the inner areas of the bend in proportion to the secondary (recirculating) flow generated by the bend.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A003. doi:10.1115/FEDSM2014-21139.

In the present paper, we are presenting a new mathematical vortex model, which is capable of simulating single and multi-celled steady, incompressible, intense vortices. The solution is obtained using the MATLAB and Maple 14 software. The new methodology is shown to fairly correlate the actual data of naturally occurring and industrial vortices.

Topics: Vortices
Commentary by Dr. Valentin Fuster
2014;():V01DT27A004. doi:10.1115/FEDSM2014-21236.

Flow structures downstream of a finned-tube are compared to those of an identical pipe; with the same diameter and length, covered with a foam layer. The standard case of cross-flow over a bare tube, i.e. no surface extension, is also tested as a benchmark. Experiments are conducted in a wind tunnel at Reynolds numbers of 4000 and 16000. Particle image velocimetry (PIV) was used for flow visualization on two different perpendicular planes. To characterize the size of the flow structures downstream of the tube, for each of the aforementioned case, two-point correlation, as a statistical analysis tool, has been used. It has been observed that by decreasing the Reynolds number, the flow structures are further stretched in streamwise direction for both bare and finned-tube cases. This is, however, more pronounced with the former. Interestingly, with a foam-wrapped tube the sizes of the flow structures are found to be independent of the Reynolds number. Finally, the structure sizes are smaller in the case of the foam-wrapped tube compared to those of finned-tube.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A005. doi:10.1115/FEDSM2014-21266.

This paper presents an experimental investigation of Reynolds number effects on the characteristics of separated and reattached flow over a smooth forward facing step. Particle image velocimetry technique was used to conduct detailed velocity measurement for a wide range of Reynolds numbers based on the step height and freestream velocity, 2040≤Reh≤8750. For each test case, the aspect ratio, AR = 21, ratio of boundary layer to step height, δ/h = 2.6 ± 0.2 and freestream turbulence level of 4% were kept constant. The results showed that the reattachment length increased monotonically with increasing Reynolds number for Reh < 6000, beyond which the reattachment length was independent of Reynolds number. In the recirculation region on top of the step, the Reynolds normal stresses were independent of Reynolds number but a higher Reynolds number increased the Reynolds shear stress in the region adjacent to the top of the step.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A006. doi:10.1115/FEDSM2014-21276.

Three-dimensional turbulent offset jets were investigated with a particle image velocimetry (PIV) technique. Detailed velocity measurements for the flow were performed at an exit Reynolds number ranging from 8080–12080 for three offset height ratios of 0, 2 and 4. Profiles of the maximum mean velocity decay and wall-normal spread rates were observed to be sensitive to offset height ratio. Contour plots of mean velocity and turbulence kinetic energy exhibited dependence on offset height ratio. The reattachment lengths of the turbulent three-dimensional offset jets were observed to increase with offset height ratio. The results within the symmetry plane revealed that the production of Reynolds shear stress was not significantly enhanced by offset height ratio further downstream.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A007. doi:10.1115/FEDSM2014-21280.

This paper presents results of an experimental research conducted to study roughness effects downstream of a forward facing step (FFS). A rough surface and a hydraulically smooth surface were used as a rough-FFS and a smooth-FFS, respectively. The upstream condition was kept smooth. Particle image velocimetry (PIV) technique was used for the velocity measurements. The Reynolds number based on the step height (h) and freestream velocity of the approach flow was kept constant at 8685. The results show that the mean reattachment length for the smooth-FFS (SM-SM) is 1.9h. Roughness reduced the peak values of the streamwise mean velocity, Reynolds shear stress and turbulent kinetic energy by 3%, 45% and 16.7% respectively in the recirculation region. In the early redevelopment region, roughness also reduced the peak values of turbulent kinetic energy and the Reynolds shear stress by 41% and 22% respectively.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A008. doi:10.1115/FEDSM2014-21300.

Convection heat transfer in upward flows of supercritical water in triangular tight fuel rod bundles is numerically investigated by using the commercial CFD code, ANSYS Fluent© 14.5.

The fuel rod with an inner diameter of 7.6 mm and the pitch-to-diameter ratio (P/D) of 1.14 is studied for mass flux ranging between 550 and 1050 kg/m2s and heat flux of 560 kW/m2 at pressures of 25 MPa.

V2F eddy viscosity turbulence model is used and, to isolate the effect of buoyancy, constant values are used for thermo-physical properties with Boussinesq approximation for the density variation with temperature in the momentum equations. The computed Nusselt number normalized by that of the same Reynolds number with no buoyancy against the buoyancy parameter proposed by Jackson and Hall’s criterion. Mentioned results are compared with V2F turbulence model whereas strong nonmonotonic variation of the thermo-physical properties as function of temperature have been applied to the commercial CFD code using user defined function (UDF) technique.

A significant decrease in Nusselt number was observed in the range of Display Formula10-6<Grq/Reb3.425Prb0.8<5×10-6 before entering a serious heat transfer deterioration regime. Based on an analysis of the shear-stress distribution in the turbulent boundary layer and the significant variation of the specific heat across the turbulent boundary layer, it is found that the same mechanism that leads to impairment of turbulence production in concentric annular pipes is present in triangular lattice fuel rod bundles at supercritical pressure.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A009. doi:10.1115/FEDSM2014-21328.

The drag-reducing phenomenon in forced homogeneous isotropic turbulence (FHIT) with polymer additives was realized in large eddy simulation (LES) results, which causes the variation of turbulent characteristics. Study on intermittency of turbulence is of great importance in investigating turbulent drag-reduction mechanism, because the intermittency has close relationship with coherent structures (CSs) and the transfer of energy in turbulent flows. In the present work, the influences of polymers on intermittency in FHIT were analyzed in detail by extracting CSs, researching the flatness factor based on high-order correlation function of velocity derivative and wavelet transform, surveying local intermittence measure, and discussing four rotational invariants consisting of velocity-strain tensor and vorticity tensor. From the viewpoint of the results, it can be perceived that the intermittency occurs in both the Newtonian fluid and polymer solution flows; moreover polymer additives behave inhibitive effect on the intermittency in turbulent drag-reducing flows.

Topics: Turbulence , Polymers
Commentary by Dr. Valentin Fuster
2014;():V01DT27A010. doi:10.1115/FEDSM2014-21545.

An experimental study is undertaken to investigate the features of separated and reattached flow over surface mounted traverse ribs of varying aspect ratio (1:1, 1:2, and 1:4) in a recirculating open channel turbulent flow. A particle image velocimetry system was used to conduct the velocity measurements. Upstream conditions were kept consistent among all three test cases. The reattachment length of the separated flow was found to decrease as rib aspect ratio increased, primarily as a result of a secondary separation reattachment formation on the ribs of increased aspect ratio. Contour plots of mean velocities, turbulence intensities, turbulent kinetic energy and Reynolds shear stresses, as well as one-dimensional profiles of streamwise mean velocity, turbulent kinetic energy and Reynolds shear stress in the recirculation and reattachment region are presented and discussed. The results show that maximum wall-normal mean velocities are approximately 40% of the approach freestream velocity. The results also indicate that the turbulence levels downstream of the block tend to decrease as the rib aspect ratio increases.

Topics: Turbulence
Commentary by Dr. Valentin Fuster
2014;():V01DT27A011. doi:10.1115/FEDSM2014-21677.

The objective of this study is to characterize flow parameters for two-dimensional turbulent jets impinging on a flat surface. An integral form of the momentum equation has been used to obtain a hydrodynamic solution. The boundary layer was divided into three regions, stagnation zone, developing zone and fully developed zone for free-surface and free shear, and into two regions, stagnation and wall jet zone for submerged jet configurations. A nonlinear ordinary differential equation has been obtained for frictional velocity at each zone using a logarithmic velocity profile with Coles’s law of the wake and solved numerically to predict wall shear stress as well as boundary layer and momentum thicknesses. The proposed method is more straightforward and computationally less expensive in calculating the main flow parameters as compared to turbulent flow models such as RANS and LES. Predicted wall shear stresses for a submerged jet were compared to experimental data for different cases and showed agreement with experimental data.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A012. doi:10.1115/FEDSM2014-21700.

A numerical analysis is performed to study the pre-stall and post-stall aerodynamic characteristics over a group of six airfoils using commercially available transition-sensitive and fully turbulent eddy-viscosity models. The study is focused on a range of Reynolds numbers from 6 × 104 to 2 × 106, wherein the flow around the airfoil is characterized by complex phenomena such as boundary layer transition, flow separation and reattachment, and formation of laminar separation bubbles on either the suction, pressure or both surfaces of airfoil. The predictive capability of the transition-sensitive k-kL model versus the fully turbulent SST k-ω model is investigated for all airfoils. The transition-sensitive k-kL model used in this study is capable of predicting both attached and separated turbulent flows over the surface of an airfoil without the need for an external linear stability solver to predict transition. The comparison between experimental data and results obtained from the numerical simulations is presented, which shows that the boundary layer transition and laminar separation bubbles that appear on the suction and pressure surfaces of the airfoil can be captured accurately by the use of a transition-sensitive model. The fully turbulent SST k-ω model predicts a turbulent boundary layer on both surfaces of the airfoil for all angles of attack and fails to predict boundary layer transition or separation bubbles. Discrepancies are observed in the predictions of airfoil stall by both the models. Reasons for the discrepancies between computational and experimental results, and also possible improvements in eddy-viscosity models, are discussed.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A013. doi:10.1115/FEDSM2014-21896.

The feasibility of reduced order modeling for turbulent flows using Proper Orthogonal Decomposition (POD) based Surrogate modeling and Galerkin Projection is demonstrated for use in the hydrodynamic modeling of the Very High Temperature Reactor (VHTR) lower plenum. The lower plenum of the Helium-cooled VHTR consists of vertical cylinder arrays subjected to turbulent jetting and cross-flow. Unsteady Reynolds-Averaged Navier-Stokes (RANS) Computational Fluid Dynamics (CFD) simulations are used to acquire an ensemble of possible solution fields for flow around a circular cylinder in an open domain. Numerical results are validated to prior published literature. From the resultant data ensemble are extracted the coherent structures to create an optimal basis. POD is used to extract the coherent structures as this technique has been demonstrated to provide a basis of a chosen dimension such that the average L2-error is minimized for the best approximation of the basis to the data ensemble. The resultant optimal basis is used to construct accurate reduced order models. The computational effectiveness and insights revealed by this reduced order modeling approach are discussed for both the Surrogate modeling approach and Galerkin Projection.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A014. doi:10.1115/FEDSM2014-21919.

Characteristics of coherent structures generated in channel flows during low Reynolds numbers mixed convection have been investigated in a square channel. The Gr/Re2 ranged between 21 and 206 which indicates that natural convection was dominant over forced convection. Two-dimensional velocity fields were measured using particle image velocimetry (PIV) technique in different planes to obtain a three-dimensional perspective of the flow field in the channel. The coherent structures were detected from the turbulent velocity fields using an algorithm based on the velocity gradient tensor second invariant (Q). The location of each detected coherent structure was recorded and its turbulent kinetic energy was computed. It was found that the strength of coherent structures increased with an increase in the bottom wall temperature. The results also indicate that the coherent structures present in the region away from the bottom heated wall were more energetic compared to the coherent structures present within the thermal boundary layer.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A015. doi:10.1115/FEDSM2014-21926.

Unsteady Reynolds-averaged Navier-Stokes (uRANS) models can provide good engineering estimates of wall shear and heat flux at a significantly lower computational cost compared with LES simulations. In this paper, we discuss the implementation of two novel variants of the k-ω turbulence model, the regularized k-ω standard and the regularized k-ω SST model, in a spectral element code, Nek5000. We present formulation for the specific dissipation rate (ω) in the standard k-ω model, which would obviate the need for ad hoc boundary conditions of ω on the wall. The regularized approach is designed to lead to grid-independent solutions as resolution is increased. We present a detailed comparison of these novel methods for various standard problems including the T-junction benchmark problem. The two approaches presented in this work compare very well with the standard k-ω model and experimental data for all the cases studied.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A016. doi:10.1115/FEDSM2014-22018.

This work reports an experimental investigation on the response of a planar wake generated by a profiled flat plate to various upstream flow conditions. A tripping wire was placed on the upper side of the flat plate just downstream of the leading edge of the plate that resulted in asymmetric separating shear layers at the trailing edge. The near wake asymmetry is compared to the symmetrical case at two different Reynolds numbers. Two asymmetric initial conditions resulted, namely, laminar boundary layer on the lower side and a turbulent boundary layer on the upper side, and a turbulent boundary layer on the lower side and tripped turbulent boundary layer on the upper surface. The near wake dynamics were investigated under the effects of the degree of asymmetry using hot-wire anemometry and flow visualizations. The measurements showed when one of the two boundary layers was tripped, the wake shifted towards the tripped side and wake spreading was found to be larger than in the case of the symmetrical wake with the effect being more pronounced in the asymmetric laminar wake. Self-similarity of the asymmetrical wakes was established by properly selecting appropriate similarity variables however, the similarity of the wake was less evident in the tripped laminar boundary layer case. Convection velocity, Uc, of the Von Karman large eddies, estimated using processed flow visualization images seemed to increase with increased Reynolds number and with increased upstream momentum thickness. In the symmetric laminar wake, Uc/U increases from 0.2 and reached an asymptotic value of about 0.85 further downstream. In the presence of perturbation, Uc/U attained a constant value of about 0.83 further downstream compared to the symmetric case. For the turbulent wake, however, asymmetry of the turbulence levels was found to increase the convection speed compared to both the laminar wake and the symmetric turbulent wake reaching a constant value nearly at the same downstream position for both the symmetric and asymmetric turbulent wake.

Topics: Wakes
Commentary by Dr. Valentin Fuster
2014;():V01DT27A017. doi:10.1115/FEDSM2014-22022.

We consider initially isotropic homogeneous turbulence which is submitted to an external force, in statistically axisymmetric configurations. First, we study hydrodynamical turbulence in a rotating frame, in which case the Coriolis force modifies the structure and dynamics of the flow, thus creating elongated structures along the axis of rotation, corresponding to an accumulation of energy in the neighbourhood of the equatorial spectral plane. Secondly, a very similar configuration is that of magnetohydrodynamics (MHD) of a conducting fluid within an externally applied space uniform magnetic field, in which case the Lorentz force also concentrates energy to the same spectral equatorial manifold, but creates axially extending current sheets, along the magnetic field. We more specifically consider the quasi-static limit at small magnetic Reynolds number, in which the induction equation is analytically solved. We study the anisotropy of each turbulent flow using progressively refined statistics applied to results of direct numerical simulations, and we show that an accurate characterization of the flow structure requires advanced two-point statistics, which are available easily only in spectral space.

Commentary by Dr. Valentin Fuster
2014;():V01DT27A018. doi:10.1115/FEDSM2014-22086.

Experimental results of a fully pulsed subsonic air jet issuing into the still surrounding air are reported in this paper. The intermittent flow containing a period of no flow between pulses due to the mechanically excitation was gauged by a single wire hot-wire anemometer operated in a constant temperature mode. A range of the Reynolds and Strouhal numbers of 1 × 104 < Re < 4 × 104 and 0.0064 < St < 0.0076 respectively was used to define the jets. Results of the traverse measurement agreed with earlier findings demonstrating strong effects of the excitation on the radial profiles of the mean axial velocity of the jet. Within the parameter ranges investigated, the pulsed jets were found to be significantly more spreading than steady jets. A less dispersive pulsed jet, however, appeared at a higher jet exit velocity. Strikingly, contradictory trends in the jet growth and entrainment at the higher and lower Reynolds number were seen as the lower Reynolds number does not produce a widening radial profile as a result of the increasing Strouhal number. From the axial measurements, the pulsed jets were characterized by the pulsed dominated- and high turbulence steady jet region in which their existences heavily relied on the magnitudes of the controlled parameters. A less fluctuating pulsed jet associated with the reduced magnitudes of aggregate turbulence intensity and relative turbulence energy however, appeared at an increased Strouhal number. Comparative studies with the existing results of non-circular orifice jets i.e cruciform, elliptic, and triangular jets are also reported to display the decay rates of centerline axial velocity and the spreading rates of the jets which benefit for the practical purposes.

Topics: Turbulence , Jets
Commentary by Dr. Valentin Fuster

Symposium on Urban Fluid Mechanics

2014;():V01DT28A001. doi:10.1115/FEDSM2014-21252.

In urban areas, pollutants are emitted from vehicles then disperse from the ground level to the downstream urban canopy layer (UCL) under the effect of the prevailing wind. For a hypothetical urban area in the form of idealized street canyons, the building-height-to-street-width (aspect) ratio (AR) changes the ground roughness which in turn leads to different turbulent airflow features. Turbulence is considered an important factor for the removal of reactive pollutants by means of dispersion/dilution and chemical reactions.

Three values of aspect ratio, covering most flow scenarios of urban street canyons, are employed in this study. The pollutant dispersion and reaction are calculated using large-eddy simulation (LES) with chemical reactions. Turbulence timescale and reaction timescale at every single point of the UCL domain are calculated to examine the pollutant removal. The characteristic mechanism of reactive pollutant dispersion over street canyons will be reported in the conference.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A002. doi:10.1115/FEDSM2014-21292.

Chicago is one of the most populated cites of US. It is located next to a freshwater source, Lake Michigan, and surrounded by productive agricultural land and diverse natural habitats. This study explores the sensitivity of mesoscale urban heat island (UHI) simulations to urban parameterizations, focusing on the Chicago metropolitan area (CMA) and its environs. For this purpose, a series of climate downscaling experiments using the Weather Research and Forecasting (WRF) model at 1 km horizontal resolution. A typical summer hot day in Chicago was considered, which is imitative of a summer day in the late 21st century. This study utilizes National Land Cover Database (NLCD) 2006 classifications to test UHI sensitivity for CMA. Among different urban parameterization schemes, BEP+BEM best reproduces the urban surface temperatures in comparison to other urban schemes. Results show that UHI is more pronounced with BEP and BEP+BEM schemes due to explicit accounting of anthropogenic heat (AH). The study also investigates the effects of urbanization on regional climate by replacing Chicago metropolitan area by agricultural landscape, which yielded increased surface wind speeds due to reduced mechanical and thermal resistance.

Topics: Heat , Cities
Commentary by Dr. Valentin Fuster
2014;():V01DT28A003. doi:10.1115/FEDSM2014-21489.

The prediction of the flow dynamics produced by the interaction between a sheared turbulent flow and a bluff body has important implications in the domain of the wind engineering and for what concerns the simulation of atmospheric dispersion of air-born pollutants.

In this study we present the results of the experimental investigation on the wake of a 2D obstacle, immersed in a neutrally stratified boundary layer flow. Measurements are performed by means of two different techniques, namely Laser Doppler Anemometry and Stereo-Particle Image Velocimetry. These allow us to map the spatial evolution of the velocity statistics up to their third order moments.

The study focuses in particular on the budget of the turbulent kinetic energy (t.k.e.) and the estimate of its mean dissipation rate. The experimental data-set is the basis for a detailed analysis of the reliability and the main limitations of a classical k-ϵ closure model. This has major implication for the numerical simulation of pollutant dispersion in the built environment.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A004. doi:10.1115/FEDSM2014-21556.

In the last decade there has been a growing interest in urban flow CFD simulations. As RANS approaches demonstrated to be not enough accurate to predict urban flows, people focus more and more on LES simulations. Though better results could be obtained with fine grid LES, the complexity of the urban physics seems to vanish the increasing computational resources. A different approach is herein considered, proposing a first uncertainty quantification (UQ) analysis on a single building pollutant dispersion case. A hybrid method merging the anchored-ANOVA and the POD/Kriging-based response surface is proposed to reduce the costs of the UQ analysis. Moreover, simulations are performed by the Lattice Boltzman (LBM) code PowerFLOW. Sensitivity results are presented showing the importance of vortex dynamics and the high sensitivity to the wind angle.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A005. doi:10.1115/FEDSM2014-21566.

This study analyses the aerodynamic effects of trees on local meteorological variables through in situ measurements and Computational Fluid Dynamics (CFD) simulations. Measurements are taken in the inner core of a medium-size Mediterranean city (Lecce, IT) where two adjacent street canyons of aspect ratio H/W∼1 (where H is the average building height and W is the average width of the street) with and without trees are investigated. Building façades and ground temperatures are estimated from infrared (IR) images, while flow and turbulence are measured through three ultrasonic anemometers placed at different heights close to a building façade at half length of the canyon. Tree crown porosity is evaluated through the Leaf Area Index (LAI) measured by a ceptometer. Numerical simulations are made using a CFD code equipped with the Reynolds Stress Model (RSM) for the treatment of turbulence. Overall, the analysis of measurements shows that trees considerably reduce the longitudinal wind speed up to 30%. Trees alter the typical diurnal cycle of surface and air temperature within the canyon, suggesting that in nocturnal hours the trapping of heat is more important than the power of passive cooling through evapo-transpiration. Comparative numerical simulations provide further evidence that flow velocity reduces in presence of trees and although the typical wind channeling observed without trees is still maintained, trees enhance the formation of a corner vortex leading to reverse flow at the openings of the street. The reduction of the exchange of momentum between the canyon and the atmosphere above, shown by the measurements in presence of trees is confirmed by numerical simulations.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A006. doi:10.1115/FEDSM2014-21572.

The aim of this work is to simulate the Urban Heat Island (UHI) in a medium size Mediterranean city (Lecce, IT) and to analyze its consequences for thermal comfort. We use the Weather Research and Forecasting (WRF) model (version 3.2), that accounts for the urban structure with a multilayer urban parameterization (BEP+BEM i.e. the Building Effect Parameterization (BEP) combined with the Building Energy Model (BEM)). Three hot and cloudless summer days have been simulated and results have been compared with field data collected during an experimental campaign performed over the whole summer in the city of Lecce, Italy. In the model, the structure and shape of the city are reproduced using detailed data related to different urban classes, urban fraction and building morphometry. For the residential urban classes, different thermal parameters that are representative of building materials in the oldest and the newer part of the city, are used. Results show that UHI reaches, on average, its maximum intensity (4–5 °C) just before sunrise, and its minimum (2 °C) occurs during the day. Model validation inferred through statistical analysis shows overall a better model performance for the historical city centre than for the suburban area. This suggests that further refinement of the building representation in the outskirts might still be required. Consequences of the increased urban temperature are evaluated in terms of thermal comfort. The maximum thermal stress occurs during the central hours of the day, while, the minimum thermal stress occurs during the twilight hours.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A007. doi:10.1115/FEDSM2014-21581.

Over the past half century, burgeoning urban areas such as Chicago have experienced elevated anthropogenic-induced alteration of local climates within urbanized regions. As a result, urban heat island (UHI) effect in these areas has intensified. Global climate change can further modulate UHI’s negative effects on human welfare and energy conservation. Various numerical models exist to understand, monitor, and predict UHI and its ramifications, but none can resolve all the relevant physical phenomena that span a wide range of scales. To this end, we have applied a comprehensive multi-scale approach to study UHI of Chicago.

The coupling of global, mesoscale, and micro-scale models has allowed for dynamical downscaling from global to regional to city and finally to neighborhood scales. The output of the Community Climate System Model (CCSM5), a general circulation model (GCM), provides future climate scenario, and its coupling with Weather Research and Forecasting (WRF) model enables studies on mesoscale behavior at urban scales. The output from the WRF model at 0.333 km resolution is used to drive a micro-scale model, ENVI-met. Through this coupling the bane of obtaining reliable initial and boundary conditions for the micro-scale model from limited available observational records has been aptly remedied. It was found that the performance of ENVI-met improves when WRF output, rather than observational data, is supplied for initial conditions. The success of the downscaling procedure allowed reasonable application of micro-scale model to future climate scenario provided by CCSM5 and WRF models. The fine (2 m) resolution of ENVI-met enables the study of two key effects of UHI at micro-scale: decreased pedestrian comfort and increased building-scale energy consumption. ENVI-met model’s explicit treatment of key processes that underpin urban microclimate makes it captivating for pedestrian comfort analysis. Building energy, however, is not modeled by ENVI-met so we have developed a simplified building energy model to estimate future cooling needs.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A008. doi:10.1115/FEDSM2014-21682.

The possibility of a major typhoon and its likely effects on Tokyo Bay have been estimated using an atmosphere-ocean-wave coupled model for future global climate conditions, based on the Intergovernmental Panel on Climate Change, Special Report on Emissions Scenarios (IPCC SRES) A1B scenario. In addition, the basin- to channel-scale unstructured grid hurricane storm surge model, Advanced CIRCulation (ADCIRC), has been used to determine the risk of storm surge flood in coastal areas, particularly on the Koto Delta, where inundations would most likely reach maximum levels during a strong typhoon. The system uses a high-resolution (down to 45 m) representation of regional geometry, bathymetry, and topography and emphasizes the seamless modeling of processes including those of storm surge, stormtide inundation, and river flow. The numerical experiment is validated by comparing the temporal and spatial distribution of water elevation and inundation with results obtained using a one-way coupling model of storm surge and wave activity. The simulation results show that the maximum tide level may exceed 4 m on the north side of Tokyo Bay, and surge-induced floods may extend throughout most of the Koto Delta region. And the validation results indicate that the sea-land interaction and river flows may significantly affect the depth and increase of extent of inland inundation.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A009. doi:10.1115/FEDSM2014-21819.

We report results from a multi-scale field experiment conducted in Cyprus in July 2010 in order to investigate the Urban Heat Island (UHI) in Nicosia capital city and its interaction with multi-scale meteorological phenomena taking place in the broader region. Specifically, the results are analysed and interpreted in terms of a non-dimensional/scaling parameter dictating the urban heat island circulation reported from laboratory experiments (Fernando et al, 2010). We find that the field measurements obey the same scaling law during the day, in the absence of any other flow phenomena apart from the urban heating. During the night we find that the deduced non-dimensional value reduces to half (compared to that during the day); this is due to the presence of katabatic winds from Troodos mountains into the urban center of Nicosia and their cooling effect superimposed on diurnal urban heating. Based on this deduction, the impact of various proposed heat island mitigation measures in urban planning can be evaluated.

Commentary by Dr. Valentin Fuster
2014;():V01DT28A010. doi:10.1115/FEDSM2014-21820.

The AIRCITY project, partly funded by the European Union, is now successfully achieved. It aimed at developing a 4D innovative numerical simulation tool dedicated to the dispersion of traffic-induced air pollution at local scale on the whole urban area of PARIS. AIRCITY modeling system is based on PMSS (Parallel-Micro-SWIFT-SPRAY) software, which has been developed by ARIA Technologies in close collaboration with CEA and MOKILI. PMSS is a simplified CFD solution which is an alternative to micro-scale simulations usually carried out with full-CFD. Yet, AIRCITY challenge was to model the flow and pollutant dispersion with a 3 m resolution over the whole city of Paris covering a 14 km × 11,5 km domain. Thus, the choice was to run a mass-consistent diagnostic flow model (SWIFT) associated with a Lagrangian Particle Dispersion Model (SPRAY) on a massively parallel architecture. With a 3 m resolution on this huge domain, parallelization was applied to the computation of both the flow (by domain splitting) and the Lagrangian dispersion (management of particles is split over several processors). This MPI parallelization is more complex but gives a large flexibility to optimize the number of CPU, the available RAM and the CPU time. So, it makes possible to process arbitrarily large domains (only limited by the memory of the available nodes). As CEA operates the largest computing center in Europe, with parallel machines ranging from a few hundred to several thousand cores, the modeling system was tested on huge parallel clusters. More usual and affordable computers with a few tens of cores were also utilized during the project by ARIA Technologies and by AIRPARIF, the Regional Air Quality Management Board of Paris region, whose role was also to build the end-users requirements. These computations were performed on a simulation domain restricted to the hypercenter of Paris with dimensions around 2 km × 2 km (at the same resolution of 3 m).

The focus was on the improvements needed to adapt simulation codes initially designed for emergency response to urban air quality applications:

• Coupling with the MM5 / CHIMERE operational photochemical model at AIRPARIF (as the forecast background),

• Turbulence generated by traffic / coupling with traffic model,

• Inclusion of chemical reactions / Interaction with background substances (especially NO / NO2).

Finally, in-depth validation of the modeling system was undertaken using both the routine air quality measurements in Paris (at four stations influenced by the road traffic) and a field experiment specially arranged for the project, with LIDARs provided by LEOSPHERE Inc. Comparison of PMSS and measurements gave excellent results concerning NO / NO2 and PM10 hourly concentrations for a monthly period of time while the LIDAR campaign results were also promising. In the paper, more details are given regarding the modeling system principles and developments and its validation.

Perspectives of the project will also be discussed as AIRCITY system. The TRL must now be elevated from a demonstration to a robust and systematically validated modeling tool that could be used to predict routinely the air quality in Paris and in other large cities around the world.

Commentary by Dr. Valentin Fuster

Fluid Dynamic Behavior of Complex Particles

2014;():V01DT30A001. doi:10.1115/FEDSM2014-21166.

The process of submicron particle movement in laminar boundary layers is present in many practical applications such as the particles depositing on the turbine blade and mist droplets depositing on the surface of aircrafts. Although great progress has been made on this issue during the last decades, many underlying mechanisms still remain unclear. Here, we developed a theoretical model to understand how submicron particles will behave when they enter a supersonic laminar boundary layer above an adiabatic plate along with the main stream. In this model, we used the Lagrangian method to track the particles and calculate their trajectories, and the Eulerian method was used to calculate the flow field. Because of the large velocity and temperature gradient near the wall and the small size of the particle in this question, four forces (e.g., drag force, Saffman lift force, thermophoretic force and Brownian force) acting on the particle are considered. The effects of entering position, Mach number, the size and density of particles are investigated. We discovered that there are three particle movement patterns when they enter the supersonic boundary layer, and that the drag force and Saffman lift force play dominating roles on which pattern will happen in this process. Moreover, the results also reveal that the particle tends to move towards the wall as the diameter and the density of the particle and the Mach number of main flow increases. Finally, we suggested a dimensionless number to describe the three patterns of particle motion. This research provides a better understanding of the particle movement process in the supersonic laminar boundary layer, which can be a useful guidance for the industrial processes involving this phenomenon.

Commentary by Dr. Valentin Fuster
2014;():V01DT30A002. doi:10.1115/FEDSM2014-21558.

The stresses acting on aggregates smaller than the Kolmogorov length scale in homogeneous isotropic turbulence were estimated by a two-scale numerical simulation. The fluid dynamics at the scales larger than the Kolmogorov length scale was calculated by a Direct Numerical Simulation of the turbulent flow, in which the aggregates were modeled as point particles. Then, we adopted Stokesian Dynamics to evaluate the phenomena governed by the smooth velocity field of the smallest scales. At this level the disordered structure of the aggregates was modeled in detail, in order to take into account the role that the primary particles have in generating and transferring the internal stress. From this result, it was possible to evaluate the internal forces acting at intermonomer contacts and determine the occurrence of breakup as a consequence of the failure of intermonomer bonds. The method was applied to disordered aggregates with isostatic and highly hyperstatic structures, respectively.

Commentary by Dr. Valentin Fuster
2014;():V01DT30A003. doi:10.1115/FEDSM2014-21570.

Monodisperse and polydisperse suspension flows form an extensive section of natural and technological flows. These flow structures can be categorized to sedimenting or neutrally buoyant suspensions considering the density ratio between particle phase to dispersion phase. Biological systems, food processing, ceramic injection, dynamic filtration and air conditioning are examples of areas that such flows arise. Various complicated interparticle interactions and their inevitable influence on and from the continuous phase result in some interesting phenomena which are challenging to justify. This research studies axial instabilities of suspension flow in a partially filled Taylor-Couette setup. Previous observations show that when a monodisperse suspension undergoes a rotational shear motion in a partially filled horizontal Couette cell, particles leave their initial uniform distribution and migrate to regions with lower shear rate. This migration helps formation of ring-shape axial concentrated bands. This study examines the noncolloidal neutrally buoyant suspensions of hard spherical particles with average diameters of 150, 360, 850 micron. Using UCON oil (poly ethylene glycol-ran-glycol) as suspending fluid, monodisperse and polydisperse suspensions in partially filled Stokesian Couette-Taylor flow were studied. The results show strong dependence of band number and profile on suspension concentration and filling level. Moreover interesting phenomena in polydisperse suspensions such as different band shape and weak dependence of band formation time on size of constituents were observed.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2014;():V01DT30A004. doi:10.1115/FEDSM2014-21634.

The physics of transport, deposition, detachment and reentrainment re-entrainment of particles suspended in a fluid are of great interests in many practical fluid engineering problems. For spherical particles, analysis of their translational motions is sufficient for a complete description of their transport processes. Prediction of transport and deposition of non-spherical particles, however, is more complicated due to the coupling of particle translational and rotational motions. Most studies related to dispersion of ellipsoidal particles used the traditional creeping flow formulations for hydrodynamic forces and torques. These formulations are valid for very low Reynolds number flows. In this study, dispersion and deposition of ellipsoidal particles in a fully developed laminar pipe flow are analyzed numerically using new correlations for hydrodynamic forces and torques. The deposition efficiency of the ellipsoidal particles in laminar pipe flow are calculated and the results are compared with other theoretical and numerical studies and good agreement is observed.

Commentary by Dr. Valentin Fuster
2014;():V01DT30A005. doi:10.1115/FEDSM2014-21982.

Collisions of small and heavy non-spherical particles settling in a turbulent environment are very important to various fields of physics and engineering. However, in contrast to spherical particles the collision probabilities are virtually unknown. In this study we focus on a very important condition for the numerical determination of collision probabilities: the collision detection. We discuss the need for efficient strategies to narrow down the number of possible collision pairs and compare three collision detection methods for ellipsoidal particles. We derive an analytical formula for the collision probability in the case of gravitational settling and validate the collision detection methods with this. Finally, we present statistics of the accuracy and efficiency of the methods. For the case of ellipsoidal particles in turbulence we find that the continuous collision detection with neglected rotation within a time step is the optimal trade-off between accuracy and efficiency.

Commentary by Dr. Valentin Fuster
2014;():V01DT30A006. doi:10.1115/FEDSM2014-22237.

A mesh-less numerical approach, called the moving particle semi implicit method (MPS), is presented to solve inviscid Navier-Stokes equations in a fully Lagrangian form using a fractional step method. This method consists of splitting each time step in two steps. The fluid is represented with particles and the motion of each particle is calculated through interactions with neighboring particles by means of a kernel function. In this paper, the MPS method is used to simulate a dynamic system consisting of a heavy box sinking vertically into a water tank, known as Scott Russell’s wave generator problem. This problem is an example of a falling rock avalanche into natural or artificial reservoirs. The box sinks into water tank and as a result the water is heaved up to form a solitary wave and a reverse plunging wave which forms a vortex. This vortex follows the solitary wave down the water tank. The good agreement between the numerical simulation and the analytical solution confirms the accuracy of the model. This proves the applicability of the present model in simulating complex free surface problems. The number of particles on free surface is presented as an indicator of stability of the model.

Commentary by Dr. Valentin Fuster

Analysis of Elementary Processes in Dispersed Multiphase Flows

2014;():V01DT31A001. doi:10.1115/FEDSM2014-21610.

The importance of numerical calculations (CFD) for supporting the optimization and lay-out of industrial processes involving multiphase flows is continuously increasing. Numerous processes in powder technology involve wall-bounded gas-solid flows where wall collisions essentially affect the process performance. In modelling the particle wall-collision process in the frame of numerical computations the general assumption is that the particles are spherical. However, in most practical situations one is dealing with irregular non-spherical particles or particles with a certain shape, such as granulates or fibers. In the case of non-spherical particle-wall collisions in confined flows, additional parameters such as roughness, particle shape and orientation play an important role and may strongly affect the transport behavior. The change of linear and angular velocity of the particle depends on these parameters, specifically the orientation and the radius of impact of the particles. In order to improve the non-spherical particle-wall understanding and modeling, in this work regular non-spherical particle-wall collisions in three dimensions are studied experimentally and computationally. For that purpose, cylindrical particles impacting a smooth wall at different angles are used. Single particle motion is tracked in space solving for both the translational and the rotational motion whereby the orientation of the non-spherical particle is obtained through the Euler angles and the Euler parameters. Once the particle touches the wall, the change of translational and angular velocity is determined by the non-spherical particle wall collision model. Experiments are made by shooting cylindrical non-spherical particles against a smooth plane wall at various impact angles and velocities. The collision event is recorded by two perpendicular arranged high-speed cameras. The experimental velocities obtained are used for validating the model.

Commentary by Dr. Valentin Fuster
2014;():V01DT31A002. doi:10.1115/FEDSM2014-21613.

Oil and gas produced from wells usually contain impurities such as sand particles transported by fluid flowing through pipelines. The particles impinge on the pipe walls and fittings removing material from the wall and causes erosion damage. The effect of viscosity and particle size on the local thickness loss and total erosion ratio was investigated by conducting a comprehensive experimental study on the erosion of stainless steel 316 specimens caused by sand entrained in a submerged liquid jet. Two types of sand with sizes of 150 μm and 300 μm were used and added to liquids with 1, 14 and 55 cP viscosities. The tests were carried out for three different nozzle angles: 90° (normal to target), 75° and 45°. The results show that for the 90° orientation, the total erosion ratio does not change significantly with increase in viscosity. The measurements also show that the erosion ratio for 300 μm sand is approximately two times higher than for 150 μm sand; where the erosion ratio is the mass loss of the target divided by the mass throughput of sand. The local thickness losses on the specimens were measured using a 3D profilometer, and the results show, maximum erosion depth is increasing as viscosity increases. Comparing the Scanning Electron Microscopy (SEM) images of the specimens after the test revealed that the crater sizes do not change very much with increases in viscosity. SEM images for both viscosities of 1 and 55 cP showed that the craters become longer moving radially outward from the the center of the impact zone. Computational Fluid Dynamics (CFD) simulations of erosion patterns are compared with data and they tend to under predict the total erosion ratio and local thickness loss for slurry flows as viscosity increases.

Commentary by Dr. Valentin Fuster
2014;():V01DT31A003. doi:10.1115/FEDSM2014-22143.

In this present study, the VOF (Volume of Fluid) approach is adopted to capture the interface, and CSF (Continuum Surface Force) model to calculate the surface tension, and the governing equations are founded in numerical simulation of evaporating droplets. In this work, a water droplet is assumed to be suspending in high temperature air, and the gravity of a droplet is ignored. During evaporating process of the droplet, the internal circulation flow will be induced due to the gradient of temperature at the droplet surface. The interface flows from high temperature area to low temperature area, which pulls the liquid to produce convective flow inside the droplet called as Marangoni flow. Marangoni flow makes the temperature distribution tend to uniformity, which enhances heat transfer but weakens Marangoni flow in turn. So, during droplet evaporation, the internal flow is not steady.

Commentary by Dr. Valentin Fuster
2014;():V01DT31A004. doi:10.1115/FEDSM2014-22144.

We study the capillary-induced interactions and configuration of spherical and non-spherical Janus particles adsorbed at flat liquid-fluid interfaces. For Janus spheres, the equilibrium orientation results in each hemisphere being exposed to its more favored fluid. However, experimental observations suggest that some of these particles may take a tilted orientation at the interface, giving rise to a deformed interface. On the other hand, Janus ellipsoids with a large aspect ratio or a small difference in the wettability of the two regions tend to tilt even at equilibrium. The overlap of deformed menisci results in energetic interactions between neighboring particles. We numerically calculate the interface shape around the particles by minimizing the total surface energy of the system comprising of the interface and particle-fluid regions. We quantify these interactions through evaluation of capillary energy variation as a function of the orientation and separation distance between the particles. We find that Janus spheres with similar orientations undergo a relative realignment in the interface plane in order to minimize the capillary energy. In case of ellipsoidal particles, the particles assemble in a preferred side-by-side configuration. We evaluate the role of anisotropy and degree of amphiphilicity on the inter-particle force and the capillary torque. The results can be used to predict the migration and oriented assembly of Janus particles with various geometrical and surface properties at liquid-fluid interfaces.

Commentary by Dr. Valentin Fuster

Multiphase Flow With Heat/Mass Transfer in Process Technology

2014;():V01DT32A001. doi:10.1115/FEDSM2014-21078.

We report results of an analytical investigation of linear convection in a nanofluid. We consider a colloidal suspension of solid particles in a Rayleigh-Bénard geometry set-up. The analysis is confined to the mass-dominated convection regime so that results are obtained through analytical means. Our findings depict the dependence of the critical conditions for convection onset as function of several parameters. Thus, the influence of several factors, such as the particle size, the mean volume fraction of particles, the thermophoretic force as well as the sedimentation force, on the instability onset is quantified.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A002. doi:10.1115/FEDSM2014-21103.

Mass transfer from single carbon dioxide bubbles rising through contaminated water in a vertical pipe of 12.5 mm diameter was measured to investigate effects of surfactant. The bubble diameter was widely varied to cover various bubble shapes such as spheroidal, wobbling, cap and Taylor bubbles. The gas and liquid phases were 99.9 % purity carbon dioxide and a surfactant solution made of purified water and Triton X-100. Comparison of mass transfer rates between contaminated and clean bubbles made clear that the surfactant decreases the mass transfer rates of small bubbles. The Sherwood number of small bubbles in the extreme cases, i.e. zero and the highest surfactant concentrations, is well correlated in terms of the bubble Reynolds number, Schmidt number and the ratio, λ, of the bubble diameter to pipe diameter. The Sherwood numbers at intermediate surfactant concentration, however, are not well correlated using available correlations. The mass transfer rates of Taylor bubbles also decrease with increasing the surfactant concentration. They however increase with the diameter ratio and approaches that of clean Taylor bubbles as λ increases. The main cause of this tendency was revealed by interface tracking simulations, i.e. the surfactant adsorbs only in the bubble tail region and the nose-to-side region is almost clean at high λ.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A003. doi:10.1115/FEDSM2014-21135.

A transient three-dimensional volume of fluid (VOF) simulation on condensation of upward flow of wet steam inside a 12 mm i.d. vertical pipe is presented. The effect of gravity and surface tension are taken into account. A uniform wall heat flux have been fixed as boundary conditions. The mass flux is m=130∼6000 kg m−2 s−1 and the turbulence inside the vapor phase and liquid phase have been handled by Reynolds Stress (RS) model. The vapor quality of fluid x=0∼0.4. The numerical simulation results show that in all the simulated conditions only the bubbly flow, slug flow, churn flow and annular flow are observed, in addition the results of flow pattern are in good agreement with the regime map from Hewitt and Roberts. The typical velocity field characteristic of each flow pattern and the effect of velocity field on heat transfer of condensation are analyzed. It can be found that only slug flow has an obvious local eddy around the slug gas in all simulated flow patterns. The trend of heat transfer coefficients rises throughout with the increase of vapor quality for all simulated conditions, which is good agreement with the correlation from Boyko and Kruzhilin.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A004. doi:10.1115/FEDSM2014-21248.

The entrainment by pulsed or rather so-called synthetic wall jets can be used to sustain a pressure-less transport of thin liquid layers along ducts. These jets exhibit zero-net-mass-flow conditions and lead to a break of symmetry in the flow pattern during one oscillation cycle. Therefore, only a net impulse is transferred in jet direction and induces a directed movement of the ambient liquid by entrainment according to the jet direction. The aim is to investigate the applicability of the principle to counter-current contactors as an alternative to pressure drop or gravity as driving forces typically applied e.g. in counter-current liquid-liquid contactors. Experiments are performed with an apparatus containing wall-jet flow drives with a multitude of narrow slit-openings (slit-width s = 190 μm) along a channel for synthetic wall jet generation. Firstly, one single wall-jet flow drive is investigated regarding its conveying performance at different related oscillation amplitudes (eslit/s = 7–25) and frequencies (f = 1–5 Hz). Subsequently, the apparatus is extended by a second identical device, arranged in parallel but above and oriented in the opposite conveying direction. This is to demonstrate the applicability of synthetic wall jets for counter-current operations.

Topics: Pressure , Jets
Commentary by Dr. Valentin Fuster
2014;():V01DT32A005. doi:10.1115/FEDSM2014-21368.

The fluid catalytic cracking (FCC) riser reactor consists of a bottom section of liquid feed injection and vaporization and an upward straight riser of vapor-catalysts transport and reaction. The product yield, obtained at the top of riser, is an accumulative result of liquid feed injection, vaporization by liquid contacting with hot catalysts, and subsequent catalytic cracking of feed vapor while being transported concurrently with catalysts through the riser. The FCC process involves not only these sequential sub-processes but also complicated coupling among multiphase fluid hydrodynamics, heat and mass transfer between phases, and catalytic kinetic reactions of vapor components in each sub-process. It is essential to build up a model covering all sub-processes/mechanisms mentioned above through riser reactor and giving prompt results, especially for real-time online optimization of industrial operation. This paper aims to develop a parametric model, integrated from bottom feed nozzle to top exit of riser, that can quickly predict both hydrodynamic and kinetic characteristics throughout the riser as well as various parametric effects on production yield and selectivity. Highlights of modeling contributions in this integrated model include a mechanistic and spatial-structural model of multiple-nozzle feeding with strong interactions not only among sprays themselves but also with cross-flowing steam and catalysts, a heat transfer model between gaseous and catalyst phases, and a more-rigorously derived model of reactant conservation in the multiphase flow transport. The convective nature dominating the nozzle feeding, riser transport and kinetic reactions allows us to simplify the governing equations in this integrated model to a set of coupled first-order ordinary differential equations whose solutions can be obtained quickly via Runge-Kutta algorithm. Compared to the published plant data, the predicted VGO conversion and gasoline yield from the proposed model shows a much better agreement to those from previous parametric models, which suggests the newly-added sub-models of previously overlooked mechanisms can be quite important. Some parametric effects, such as the effect of catalyst-to-oil ratio and catalyst inlet temperature, on production yield and selectivity are further predicted. The results show that a higher CTO or catalyst temperature normally leads to higher cracking conversion, higher gasoline production and lower coke content. However, a very high inlet temperature of catalysts does cause over-cracking and lower the gasoline selectivity.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A006. doi:10.1115/FEDSM2014-21419.

Numerical simulation is performed for a quenching process in liquid jet impingement, which is applicable to thermal control in metal manufacturing and emergency cooling of nuclear reactors. The flow and cooling characteristics of jet impingement are investigated by solving the conservation equations of mass, momentum and energy in the liquid and gas phases. The liquid-vapor and liquid-air interfaces are tracked by a sharp-interface level-set method which is modified to include the effect of phase change at the liquid-vapor interface. The temporal and spatial variation of solid temperature is analyzed by solving a conjugate problem with the conduction in the solid as well as the convection in the liquid and gas phases. The numerical results demonstrate that the temporal variations of the temperature and heat flux near the fluid-solid interface are very steep compared to those inside the solid. The heat flux variations at the fluid-solid interface are observed to be much larger in the convection mode than in the film boiling mode. The solid temperatures and heat fluxes obtained from the present study are compared with the experimental data reported in the literature.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A007. doi:10.1115/FEDSM2014-21498.

Accurate estimation of multiphase turbulence, interphase momentum exchange and bubble dynamics parameters such as bubble departure diameter and frequency is critical for a realistic simulation of flow boiling heat transfer. While there are experimental and mechanistic models available for the estimation of these parameters for rather specific geometric configurations, fluids and operating conditions, there is no specific comprehensive model for jet impingement boiling. Nor is there a consensus on a generalized model, particularly for the ebullition parameters, that could be extended to jet impingement boiling. Hence, a problem-based evaluation of the available models to conform to experimental data is often required. In the present work, a rigorous study is carried out to ascertain the suitability of different bubble departure diameter and departure frequency models for the simulation of confined and submerged, subcooled jet impingement boiling. The choice of ebullition models considered encompass both pool boiling as well as flow boiling based models, developed from both experimental as well as mechanistic approaches. The suitability of the models are evaluated by comparison of the predicted local and surface averaged heat transfer characteristics against experimental boiling data from the present research as well as that available in the literature. The computational simulations are carried out using the finite volume computational solver ANSYS FLUENT 14.5, with necessary customized functions for boiling parameters formulated and integrated into the solver.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A008. doi:10.1115/FEDSM2014-21644.

This paper presents a combined experimental and numerical study of the evaporation and solid layer formation of a single bi-component mannitol-water droplet in air. For spherically symmetric droplets, the problem is described mathematically by the unsteady, one-dimensional conservation equations of mass and energy. The effect of the formation of a solid layer at the droplet surface on the droplet evaporation and thermal diffusion rate is included in the present approach. The simulations are validated by comparison with experiments using acoustically levitated droplets. The study includes initial droplet diameters varying from 350 to 450 μm, gas temperatures ranging from 80 to 120 °C, and the initial mannitol mass fraction inside the droplet varies from 0.05 to 0.15. The numerical results are analyzed to identify the occurrence of solid layer formation, and the temporal evolutions of both the droplet size and mass are presented. A parameter study of the initial gas temperature, the initial droplet size, and the initial mannitol mass fraction inside the droplet on droplet evaporation and solid layer formation is presented. The present model accurately captures the initial stages of droplet drying under all conditions investigated.

Topics: Drying , Drops , Evaporation
Commentary by Dr. Valentin Fuster
2014;():V01DT32A009. doi:10.1115/FEDSM2014-21678.

A wire mesh sensor (WMS) is a device used to investigate multi-phase flows. The WMS measures the instantaneous local electrical conductivity of multiphase flows at different measuring points. There is a significant difference in the electrical conductivity of the employed fluids (in this work air and water, conductivity of water is much higher than that of air). Using the difference in the electrical conductivity, the WMS provides the local void fraction. The WMS utilized in this work includes two identical planes of parallel 16×16 grid of wires. The separation distance between these two planes is 32 mm. The WMS was installed in a 76.2 mm (3-inch) diameter vertical pipe to extract information such as void fraction distribution, structure velocity, and slug/churn flow structure. The superficial gas (air) velocity (VSG) ranged from 10 to 38.4 m/s. Liquid (water) superficial velocities (VSL) of 0.30, 0.46, 0.61 and 0.76 m/s were employed. To study the effects of viscosity on the slug/churn flow structure, Carboxyl Methyl Cellulose (CMC) was added to water to increase the liquid viscosity without altering its density. Each experiment was performed for 60 seconds. An operation frequency for the WMS of 10 kHz (totaling 600,000 frames of void fraction measurement per experiment) was used for all experiments.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A010. doi:10.1115/FEDSM2014-21679.

In this study, a numerical analysis method applicable to estimation of the boiling heat transfer has been developed. Currently, the experimental correlations or the empirical laws have been applied to evaluate the boiling heat transfer. Therefore, it is difficult to predict the effects of the change of the heated surface geometry, thermal-hydraulic conditions, the surface activation or modification, because out of the application range of these correlations. The purpose of this work is to construct the boiling two-phase analysis method for thermo-fluid phenomena, and to realize “Design-by-Analysis” independent on the experiments and empirical laws. For this purpose, it is important to predict steam-water interface structure characteristics of the two-phase flow directly. Until now, for evaluating the boiling phenomena, Diffusive Interface Model for the bubble interface tracking was applied. In this model, the steam-water interface is diffuse with a finite width, and values of the thermodynamic properties change between water and steam smoothly within the interface region. For evaluating the wettability of heated surface, the surface energy is estimated by using the phase-field model. The wetting phenomena during boiling are able to be analyzed directly with this model. We present the numerical results of nucleate pool boiling phenomenon by using the developed analysis method. We succeeded in simulating the boiling process, vapor bubbles nucleation, growth, and departure behavior on a heated surface. By present analysis method, it was confirmed that the boiling heat transfer coefficient could be evaluated quantitatively without the experimental correlations.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A011. doi:10.1115/FEDSM2014-21778.

The effects of wall surface roughness on the rate of heat transfer and temperature profiles in turbulent gas-solid flows in pipes at different inclination angles were studied. The earlier developed computational model for 3D flows including the four-way interactions was extended and used in this study for evaluating the mean flow, turbulence intensity and thermal fields. Interaction of particles with the rough wall was included by introducing the available stochastic wall roughness models (shadow effect model) for the dispersed phase in the computational program. It was found that changes in the particle dispersion and particle concentration altered the Nusselt number and heat transfer rate in different regions of the pipe. The Nusselt number decreased in the lower part of the duct for horizontal and inclined pipes due to the reduction in the settling velocity. The surface roughness also altered the heat transfer coefficient in the periphery of the vertical riser. The simulation results showed that the fluid temperature was reduced in the pipe core and increased near the wall region for inclined pipes. On the other hand, particle temperature increased and flattened in the entire pipe cross section.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A012. doi:10.1115/FEDSM2014-21899.

We present a front-tracking/finite difference method for simulation of drop solidification, where the melt is confined by its own surface tension. The problem includes temporal evolution of three interfaces, i.e. solid–liquid, solid–air, and liquid–air, that are explicitly tracked under the assumption of axisymmetry. The solid–liquid interface is propagated with a normal velocity that is calculated from the normal temperature gradient across the front and the latent heat. The liquid–air front is advected by the velocity interpolated from nearest bulk fluid flow velocities. Method validation is carried out by comparing computational results with exact solutions for two-dimensional Stefan problems, and with related experiments. We then use the method to investigate a drop solidifying on a cold plate in which there exists volume expansion due to density difference between the solid and liquid phases. Effects of the tri-junction in terms of growth angles on the solidification process are also investigated. Computational results show that a decrease in the density ratio of solid to liquid or an increase in the growth angle results in an increase in the height of the solidified drop. In addition, reducing the gravitational effect also increases the drop height after solidification.

Topics: Solidification
Commentary by Dr. Valentin Fuster
2014;():V01DT32A013. doi:10.1115/FEDSM2014-21924.

Flashing of high temperature, pressurized slurry within hydrometallurgical processing circuits is a commonly encountered multiphase flow scenario, which can lead to catastrophic equipment failure and serious operational problems if not designed correctly. The current work aims to shed light on the state of the art modelling and experimentation, important physical phenomena, and recent operational experience surrounding this problem. In addition, recommendations will be provided for future modelling and experimental efforts in order to direct research into avenues that provide valuable information for engineers designing piping and vessels where the flashing of high pressure slurry occurs.

Topics: Flashing , Slurries
Commentary by Dr. Valentin Fuster
2014;():V01DT32A014. doi:10.1115/FEDSM2014-21998.

An advanced atomizer concept to obtain larger production rates of nano-particles by the Flame Spray Pyrolysis process (FSP) is investigated. In the conventional FSP process (external mixing gas/liquid nozzle) production rates may be varied by increasing the precursor feed rate and/or the precursor concentration. However, both measures typically result in the formation of larger nanoparticles. These effects may be avoided by the development and integration of advanced atomizer concepts. The aim is to address the spray structure in a way that keeps the flame height constant and modifies the flame width. Therefore, the time scales and the residence time-temperature histories of droplets and nanoparticles are expected to be similar while the production rate is increased. The atomizer concept for creation of a modified spray and flame combines a swirling liquid film generation that is atomized with an external swirling gas flow.

In the first step a hollow cone of liquid ligaments and primary droplets is generated through a conventional pressure-swirl nozzle. The liquid phase is atomized in the second step, by the expanding gas of a circular ring nozzle. To study the main characteristics of the combined atomizer in model experiments, water and water/glycerol mixtures are used as the liquid phase and air as the gas phase. For investigation of the atomizer and spray properties, the relation between liquid outlet angle, inlet angle of the gas, the gas/liquid flow rates, the spray cone geometry and droplet size distribution are investigated. The spray structure and the breakup of the film are analyzed by high speed images. Laser diffraction is used to measure the droplet size distribution in the spray.

A numerical model is developed and used to simulate the cold gas flow and spray distribution as in the adapted atomizer concept. The Eulerian-Lagranian approach is solved by means of a computational fluid dynamics (CFD) code. The process parameters such as liquid composition, liquid and gas flow rates are varied to meet the specific requirements of the nanoparticle production in the FSP process. The experimental and numerical investigation showed that an enlarged and steady spray resulted from an increased outlet angle of the liquid and gas swirl. Increased tangential velocities increase the entrainment of surrounding gas, widening and providing a more uniform velocity profile to the spray. Spray droplet mean diameters resulted in the desired range of ≤ 20 μm.

Commentary by Dr. Valentin Fuster
2014;():V01DT32A015. doi:10.1115/FEDSM2014-22136.

In this study, we experimentally investigated the distribution of liquid and gas flow in the head of plate-fin heat exchanger. The scale-downed model of proto-type heat exchanger was used and the air and water replaced the natural gas and liquefied natural gas. Two CCD cameras were synchronized with laser to capture the image of liquid-gas flow in the head. Advanced imaging techniques such as LIF, PIV method were used to measure the velocity field of liquid and gas flow simultaneously. The characteristics of momentum transfer process between liquid phase and gas phase are presented in the terms of ensemble averaged result. The effect of liquid mass fraction on the flow distribution was investigated. Then, we quantified the flow distribution of liquid and gas phase for different inlet nozzle configurations.

Commentary by Dr. Valentin Fuster

Fluid Mechanics of Aircraft and Rocket Emissions and Their Environmental Impacts

2014;():V01DT33A001. doi:10.1115/FEDSM2014-21311.

This study uses a Fickian-Distribution parameterization [Chen & Lamb, 1994] to model the effects of ice habits on contrail formation within a large eddy simulation (LES). Box model cases were first performed at various ambient temperatures and relative humidities over ice (RHi) and results compared with available laboratory data of ice crystal growth and habit distribution [Bailey & Hallett, 2004]. The model was then used in a full 3-D LES of contrails and results were compared with in-situ observations [Febvre et. al., 2009]. Comparisons are also made with results from simulations that used a probabilistic ice habit model [Inamdar et. al., 2013].

Topics: Ice
Commentary by Dr. Valentin Fuster
2014;():V01DT33A002. doi:10.1115/FEDSM2014-21349.

This paper presents a feasibility study of a hybrid RANS–LES approach to numerical simulation of aircraft wing-tip vortices. A NACA 0012 wing is considered for which earlier published experimental and numerical data are available. Mesh sensitivity tests of our RANS solver and comparisons between two different turbulence models indicate that the RANS approach adequately describes the flow upstream from the trailing edge, but overestimates the rate of decay of the wing-tip vortex. A hybrid RANS–LES method is presented that results in a better agreement with the wind tunnel experiment, hence this approach is suggested for numerical simulation of the wake of an airliner.

Commentary by Dr. Valentin Fuster
2014;():V01DT33A003. doi:10.1115/FEDSM2014-21536.

This paper focuses on two decisive steps towards Large Eddy Simulation of a solid rocket booster jet. First, three-dimensional Large Eddy Simulations of a non-reactive booster jet including the nozzle were obtained at flight conditions of 20 km of altitude. A particularly long computational domain (400 nozzle exit diameters in the jet axial direction) was simulated, thanks to an innovative local time-stepping method via coupling multi instances of a fluid solver. The dynamics of the jet is analysed and comparison of the results with previous knowledge validates the simulations and confirms that this computational setup can be applied for Large Eddy Simulations of a reactive booster jet. The second part of this paper details the implementation of a simple method to study the hot plume chemistry. Despite its limitations, it is accurate enough to observe the various steps of the chemical mechanism and assess the effect of uncertainties of the rate parameters on chlorine reactions. It was also used to reduce the set of chemical reactions into a short scheme involving a minimum of species and having a limited impact on the physical time step of the Large Eddy Simulations.

Commentary by Dr. Valentin Fuster

1st Symposium on Algorithms and Applications for High Performance CFD Computation

2014;():V01DT35A001. doi:10.1115/FEDSM2014-21669.

GPU computation in recent years has seen extensive growth due to advancement in both hardware and software stack. This has led to increase in the use of GPUs as accelerators across a broad spectrum of applications. This work deals with the use of general purpose GPUs for performing CFD computations.

The paper discusses strategies and findings on porting a large multi-functional CFD code to the GPU architecture. Within this framework, the most compute intensive segment of the software, the BiCGSTAB linear solver using additive Schwarz block pre-conditioners with point Jacobi iterative smoothing is optimized for the GPU platform using various techniques in CUDA Fortran. Representative turbulent channel and pipe flow are investigated for validation and benchmarking purposes. Both single and double precision calculations are highlighted.

It is found that the precision has a negligible effect on the accuracy of predicted turbulent statistics. However, it was found that single precision calculations led to instabilities in the initial convergence of the pressure equation if the convergence criterion was set at too low a value. This was remedied by limiting the number of iterations during the initial stages of the calculation. For a modest single block grid of 64×64×64, the turbulent channel flow computations showed a speedup of about 8 fold in double precision whereas it was more than 13 fold for single precision on the NVIDIA Tesla GPU. For the pipe flow consisting of 1.78 million grid cells distributed over 36 blocks, the gains were more modest at 4.5 and 6.5 for double and single precision respectively.

Commentary by Dr. Valentin Fuster
2014;():V01DT35A002. doi:10.1115/FEDSM2014-22057.

Cavitating bubbly flows are encountered in many engineering problems involving propellers, pumps, valves, ultrasonic biomedical applications, … etc. In this contribution an OpenMP parallelized Euler-Lagrange model of two-phase flow problems and cavitation is presented. The two-phase medium is treated as a continuum and solved on an Eulerian grid, while the discrete bubbles are tracked in a Lagrangian fashion with their dynamics computed. The intimate coupling between the two description levels is realized through the local void fraction, which is computed from the instantaneous bubble volumes and locations, and provides the continuum properties. Since, in practice, any such flows will involve large numbers of bubbles, schemes for significant speedup are needed to reduce computation times. We present here a shared-memory parallelization scheme combining domain decomposition for the continuum domain and number decomposition for the bubbles; both selected to realize maximum speed up and good load balance. The Eulerian computational domain is subdivided based on geometry into several subdomains, while for the Lagrangian computations, the bubbles are subdivided based on their indices into several subsets. The number of fluid subdomains and bubble subsets are matched with the number of CPU cores available in a share-memory system. Computation of the continuum solution and the bubble dynamics proceeds sequentially. During each computation time step, all selected OpenMP threads are first used to evolve the fluid solution, with each handling one subdomain. Upon completion, the OpenMP threads selected for the Lagrangian solution are then used to execute the bubble computations. All data exchanges are executed through the shared memory. Extra steps are taken to localize the memory access pattern to minimize non-local data fetch latency, since severe performance penalty may occur on a Non-Uniform Memory Architecture multiprocessing system where thread access to non-local memory is much slower than to local memory.

This parallelization scheme is illustrated on a typical non-uniform bubbly flow problem, cloud bubble dynamics near a rigid wall driven by an imposed pressure function.

Topics: Bubbly flow , Modeling
Commentary by Dr. Valentin Fuster
2014;():V01DT35A003. doi:10.1115/FEDSM2014-22257.

Virtual Cylinder Model (VCM) was used to simulate flows over vegetation plants (cylinders) in coastal wetlands. VCM is capable of characterizing the flow field with a few plants as well as numerous plants with high efficiency and accuracy. Numerical results of flow over cylinders at a regular pattern are compared with Direct Numerical Simulations and at irregular patterns are presented with varied resolutions. VCM provides decent accuracy and efficiency without high resolution in tiny mesh. Results demonstrate that it is suitable for large-scale simulation of vegetation resilience to protect coastal wetlands from waves.

Topics: Waves , Modeling
Commentary by Dr. Valentin Fuster

Performance of Multiphase Flow Systems

2014;():V01DT38A001. doi:10.1115/FEDSM2014-21058.

The heat transfer performance of various thermal devices can be augmented by active and passive techniques. One of the passive techniques is the addition of nanoparticles had the size of 1 and 100 nanometers to the common heat transfer so that the thermal transport properties of the prepared suspension called nanofluid will be enhanced compared to the base fluid. Nanorefrigerants as a special type of nanofluids which are mixtures of nanoparticles and refrigerants have a wide range of applications in diverse fields such as refrigeration, air conditioning systems and heat pumps. In this study, the missing points on this new method are also indicated regarding the lack of studies on the determination of physical properties of nanorefrigerants and the flow of nanoparticles.

Commentary by Dr. Valentin Fuster
2014;():V01DT38A002. doi:10.1115/FEDSM2014-21126.

The pressure drop is considerably significant for the differential pressure meter to measure the flow rate of the two-phase flow. Little is known about the pressure drop characteristics of the V-Cone meter when it is used to measure the wet gas flow. The objective of this paper is to investigate the two-phase pressure drop of the V-Cone meter and develop a correlation for predicting its pressure drop. A V-Cone meter with the equivalent diameter ratio of 0.55 was investigated experimentally. The experimental fluid was air and water. The test pressure ranged from 0.1 MPa to 0.4 MPa, and the gas and liquid mass flow rate ranged from 100 Nm3/h to 500 Nm3/h and from 0.05 m3/h to 2.2 m3/h, respectively. The experimental results showed that the existing correlations, which are developed for the orifice plate meter and the Venturi meter, are not applicable for the V-Cone meter to predict the pressure drop. The two-phase mass flow coefficient, K, was used to develop the two-phase pressure drop correlation. The influences of the Lockhart-Martinelli parameter, the gas densiometric Froude number and the operating pressure on K were investigated. The new pressure drop correlation can accurately predict the pressure drop of the V-Cone meter for the wet gas. The relative error of the pressure drop is less than ± 9.0% at the 95.1% confidence level and the average relative error is 3.88%. The pressure drop prediction correlation provides a reference for developing the correlation of the wet gas measurement.

Commentary by Dr. Valentin Fuster
2014;():V01DT38A003. doi:10.1115/FEDSM2014-21144.

Swirling gas-liquid two-phase flow patterns and pressure drop in vertical pipes of a large diameter are widely present in practical applications but not well documented in experimental studies. This paper presented an experimental study on gas-liquid two phase flow patterns and pressure drop inside a vertical pipe of 62mm in inner diameter (ID) containing a helical tape insert. Experimental results were obtained in a vertical visualization test section with a length of 7m, liquid mass velocities ranging from 0.3 to 1000 kg/(m2·s), and gas mass velocities from 3.2 to 900kg/(m2·s). Considering the decay of the swirl flow, the swirling flow regime map at different cross sections (z/D = 16, 32 and 64) were concluded, and their effects on the pressure drop were investigated.

Commentary by Dr. Valentin Fuster
2014;():V01DT38A004. doi:10.1115/FEDSM2014-21314.

Two phase flow in vertical risers are common place in oil and gas industry and many other process industries. Depending on the flow rates of the phases, there could be several flow patterns could exist inside the riser. These could vary from bubbly flow to annular flow with slug and churn flow in between. When the liquid phase flow rate is higher the bubbly flow exists while the annular flow is dominated by higher gas flow rate that forms a distinct gas core in the middle of the vertical riser. Of these flow regimes, churn flow is of particular interest as it is not well understood. The paper will report findings of an experimental campaign investigating the development of churn flow.

Experiments were carried out in a closed loop flow facility with a 127 mm ID, 11 m long vertical test section. The maximum flow rates achievable in the system were 17 and 1.2 m/s for gas and liquid phases respectively. Compressed air was used as the gas phase while water and water/glycerol mixtures were used as the liquid phase. The mixtures of water and glycerol were used to investigate the influence of the viscosity on the flow regime investigated. The flow was investigated using a Wire Mesh Sensor (WMS), an intrusive measurement device that can map the cross sectional distribution of phases. Void fraction measurements were made at several axial locations for a number of flow rate combinations from onset of churn flow until it turns into annular flow.

A region in flow rates where large liquid ligaments (wisps) suspended in the gas core was found and the breakup mechanism has been observed. Furthermore, huge waves were observed in this region. Analysis of results shows that the frequency of both huge waves and the wisps entrained in the gas core increase along the axial distance. The changes to the flow behaviour with the increase of viscosity and other findings will be presented in detail in the paper.

Commentary by Dr. Valentin Fuster
2014;():V01DT38A005. doi:10.1115/FEDSM2014-21776.

The effect of presence of solid particles on stratified wavy gas-liquid flows has been studied. The height of liquid phase in the natural gas pipeline is a key parameter in designing and can affect the corrosion/erosion rate. In present paper, the numerical four-way simulation of solid particles in gas-liquid wavy stratified flow has been used. The computational model is shown to be able to evaluate the effect of the particles on liquid holdup which is critical for gas pipeline design. The particles cause the liquid phase height in horizontal pipe decreases by increasing the solid phase concentration.

Commentary by Dr. Valentin Fuster
2014;():V01DT38A006. doi:10.1115/FEDSM2014-22017.

It is known that in large vessels (whole) blood behaves as a Navier-Stokes (Newtonian) fluid; however, in a vessel whose characteristic dimension (e.g., a diameter in the range of 20 to 500 microns) is about the same size as the characteristic size of the blood cells, blood behaves as a non-Newtonian fluid, exhibiting complex phenomena, such as shear-thinning, stress relaxation, the Fahraeus effect, the plasma-skimming, etc.. Using the framework of mixture theory an Eulerian-Eulerian two phase model is applied to model blood flow, where the plasma is treated as Newtonian fluid and the RBCs are treated as shear thinning fluid.[5]

Commentary by Dr. Valentin Fuster
2014;():V01DT38A007. doi:10.1115/FEDSM2014-22263.

Multiphase flow measurement is a very challenging issue in process industry. One of the promising approaches for multiphase flow analysis is image processing. Image segmentation is very important step in multiphase flow analysis. Determination of appropriate threshold value is very critical step for correct identification of the liquid and gas phases. There are two main thresholding techniques: Global and Adaptive. Adaptive thresholding is more suitable for multiphase flow case due to it’s adaptability to image conditions such non-uniform illumination and noise. In this work, six adaptive thresholding techniques are examined for the case of wavy flow regime. These algorithms are used to estimate the wave shape and mix region between liquid and gas. In general, the adaptive algorithms are able to compensate for non-uniform illumination and they are able to estimate wave shape and mix region correctly. The execution time for the adaptive techniques is higher than global thresholding technique, but with the availability of new powerful PCs, it will become a minor issue.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster

Symposium on the Fluid Dynamics of Wind Energy

2014;():V01DT39A001. doi:10.1115/FEDSM2014-21283.

Wind turbine icing represents the most significant threat to the integrity of wind turbines in cold weather. Advancing the technology for safe and efficient wind turbine operation in atmospheric icing conditions requires the development of innovative, effective anti-/de-icing strategies tailored for wind turbine icing mitigation and protection. Doing so requires a keen understanding of the underlying physics of complicated thermal flow phenomena pertinent to wind turbine icing phenomena, both for the icing itself as well as for the water runback along contaminated surfaces of wind turbine blades. In the present study, an experimental investigation was conducted to characterize the surface wind-driven water film/rivulet flows over a NACA 0012 airfoil in order to elucidate the underlying physics of the transient surface water transport behavior pertinent to wind turbine icing phenomena. The experimental study was conducted in an icing research wind tunnel available at Aerospace Engineering Department of Iowa State University. A novel digital image projection (DIP) measurement system was developed and applied to achieve quantitative measurements of the thickness distributions of the surface water film/rivulet flow at different test conditions. The measurement results reveal clearly that, after impinged on the leading edge of the NACA0012 airfoil, the micro-sized water droplets would coalesce to form a thin water film in the region near the leading edge of the airfoil. The formation of rivulets was found to be time-dependent process and relies on the initial water runback flow structure. The width and the spacing of the water rivulets were found to decrease monotonically with the increasing wind speed. The film thickness icing scaling law is evaluated by the time-average measurement film thickness. The measurement results show good consistent with the analytical scaling predictions.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A002. doi:10.1115/FEDSM2014-21285.

An experimental study was carried out to investigate the aeromechanics and wake characteristics of dual-rotor wind turbines (DRWTs) with co- and counter-rotating configurations, in comparison to those of a conventional singlerotor wind turbine (SRWT), in order to elucidate the underlying physics to explore/optimize design of wind turbines for higher power yield and better durability. The experiments were performed in a large-scale Aerodynamic/Atmospheric Boundary Layer (AABL) wind tunnel under neutral stability conditions. In addition to measuring the power output performance of DRWT and SRWT systems, static and dynamic wind loads acting on those systems were also investigated. Furthermore, a high resolution PIV system was used for detailed near wake flow field measurements (free-run and phase-locked) so as to quantify the near wake turbulent flow structures and observe the transient behavior of the unsteady vortex structures in the wake of DRWT and SRWT systems. In the light of the promising experimental results on DRWTs, this study can be extended further to investigate the turbulent flow in the far wake of DRWTs and utilize multiple DRWTs in different wind farm operations.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A003. doi:10.1115/FEDSM2014-21286.

An experimental study was conducted to compare the characteristics of the dynamic wind loads and evolution of the unsteady vortex and turbulent flow structures in the wake of a wind turbine sited in onshore and offshore wind farms. A scaled three-blade Horizontal Axial Wind Turbine (HAWT) model was placed in Atmospheric Boundary Layer (ABL) winds with different mean and turbulence characteristics to simulate the wind conditions in onshore and offshore wind farms. In addition to measuring dynamic wind loads acting on the wind turbine model by using a high-sensitive force-moment sensor unit, a high-resolution digital Particle Image Velocimetry (PIV) system was used to achieve flow field measurements to quantify the characteristics of the turbulent flow in the wake of the wind turbine model. Besides conducting “free-run” PIV measurements to determine the ensemble-averaged statistics of the flow quantities such as mean velocity, Reynolds stress, and Turbulence Kinetic Energy (TKE) distributions in the wake, “phase-locked” PIV measurements were also performed to elucidate further details about evolution of the unsteady vortex structures in the wake flow in relation to the position of the rotating turbine blades. The detailed flow field measurements are correlated with the dynamic wind loads measurements to elucidate underlying physics in order to gain further insight into changes of the dynamic wind loads and wake characteristics behind the wind turbine operating in either onshore or offshore wind farms.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A004. doi:10.1115/FEDSM2014-21356.

This paper aims to predict the performance of Vertical Axis Wind Turbine (VAWT), hence the modeling of kinetic energy extraction from wind and its conversion to mechanical energy at the rotor axis, is carried out. The H-type Darrieus turbine consists of three straight blades with shape of aerofoil attached to a rotating vertical shaft. The criterion on the selection of this kind of turbines, despite its reduced efficiency, is the easy manufacture in workshops.

A parametric study has been carried out to analyze the camber effect on the non dimensional curves of power coefficient so that the self starting features as well as the range of tip speed ratio of operation could be predicted.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A005. doi:10.1115/FEDSM2014-21377.

Dynamic Line Rating (DLR) is a smart grid technology that allows the rating of power line conductor to be based on its real-time temperature. Currently, conductors are generally given a conservative static rating based on near worst case weather conditions. Using historical weather data collected over a test bed area in Idaho, we demonstrate there is often additional transmission capacity not being utilized under the current static rating practice. We investigate a DLR method that employs computational fluid dynamics (CFD) to determine wind conditions along transmission lines in dense intervals. Simulated wind conditions are then used to calculate real-time conductor temperature under changing weather conditions. In calculating the conductor temperature and then inferring the ampacity, we use both a steady-state and transient calculation procedure. Under low wind conditions, the steady-state assumption predicts higher conductor temperatures which could lead to unnecessary curtailments, whereas the transient calculations produce temperatures that can be significantly lower, implying the availability of additional transmission capacity. Equally important, we demonstrate that capturing the wind direction variability in the simulations is critical in estimating conductor temperatures accurately.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A006. doi:10.1115/FEDSM2014-21783.

A novel ducted wind turbine, referred to as the Wind Tower technology, for capturing wind power in either residential or commercial scale applications is developed. A mathematical model of the fluid flow inside the tower is derived and the experimental tests have been conducted on a 1/8th prototype. Proposing an optimum design under the consideration of a maximum output power generation and a minimum cost of energy, which leads to a maximum return on investment, is a big challenge in developing the Wind Tower technology. Numerical simulation of the fluid flow inside the tower helps to make a more precise and accurate estimation of the flow characteristics at different sections of the tower. Hence, a two-dimensional single-nozzle Wind Tower is modeled to perform the computational fluid dynamics calculation. This provides the effect of major components of the Wind Tower, including the inlet and the outlet dimensions, and the Wind Tower structure configurations on the flow characteristics going through the tower at laminar steady-state condition. The average velocity values at the outlet due to changing the Wind Tower component configurations are measured. The results provide the optimum dimension ranges of the major Wind Tower components; these values would be further refined at a specific site with its certain wind characteristics.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A007. doi:10.1115/FEDSM2014-21784.

The statistical data of five years wind speed measurements at University of Maryland, Baltimore County are used to find out the availability of wind energy resource for power generation. Wind speeds are measured at an approximately 30 meters above the ground; the monthly and yearly mean wind speeds are calculated and evaluated by using the Weibull distribution function. The annual values of k (dimensionless Weibull shape parameter) ranged from 1.78 to 1.99 with a five-year mean value of 1.87. The annual values of c (Weibull scale parameter) ranged from 3.15 to 3.60 with a five-year mean value of 3.28. The results show the highest and lowest wind power potential occurs in February and July, respectively. While this site is not appropriate for large-scale power generation, this study shows the availability of enough wind potential for non-grid connected electrical and mechanical applications. Different residential wind harvesting technologies in urban areas have been studied and more promising ones are introduced as solutions to provide larger-scale power generation at this site with a low annual mean wind speed.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A008. doi:10.1115/FEDSM2014-21953.

This paper presents a numerical simulation of unsteady flow over wind turbine arrays to understand rotor-rotor and rotor-tower wake interaction in wind farms. The computations are carried out by incorporating Actuator Line method into a large eddy simulation. This methodology is validated by comparing the results to predictions of large eddy simulation using exact 3D blade geometries from a two-blade NREL Phase VI turbine. The method is then used to simulate the wake development in a two-turbine case. It is discovered that in the full wake setting the tower has a significant influence on the central part of the turbine wake. It is observed that the tower wake is twisted due to the rotation of the turbine wake. As a result, this tower wake is expected to have impact on the performance of downstream wind turbines, which cannot be overlooked. The present work also demonstrates the potential of combining AL method with LES to predict wake interactions in wind farms.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A009. doi:10.1115/FEDSM2014-22009.

In this paper, the Large Eddy Simulation coupled with the Actuator line (LES-AL) method is employed to analyze the performance of the downstream wind turbine under varying inflow conditions. A direct LES, which solves the flow physics around turbine blades using exact three-dimensional blade geometries, is carried out to predict the aerodynamic loadings and output powers of the downstream turbines by prescribing the wake profiles predicted by LES-AL simulation as the inflow boundary conditions. The upstream tower shadow effect is presented in this study by carrying out two simulations with no tower wake and real tower wake inflow conditions. The LES results show that both the power and aerodynamic loading components fluctuate periodically due to the presence of upstream tower. In additional, an additional force component is exerted on the downstream wind turbine in the vertical direction (z direction). The increase in velocity deficit in wake in behind the downstream turbine is due to a sequence of momentum extraction by the wind turbines. The tower shadow effect accumulates and generates lower velocity regions in wake, and the low velocity regions shift due to the rotational motion of wake vortex. The development of the asymmetric and velocity deficit region has the potential to generate more unstable power output and fatigue loading on turbines in further downstream.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A010. doi:10.1115/FEDSM2014-22072.

First results from an experimental investigation of the fully developed wind turbine array boundary layer are reported, using arrays of up to 100 model wind turbines with a diameter of 0.25 m. The wind turbine array was simulated by a combination of drag-matched porous disks, used in the upstream part of the array, and by a smaller array of realistically scaled 3-bladed wind turbines just upstream of the measurement location. The model array was placed in the 6.0 m × 2.7 m × 72.0 m test section of the UNH Flow Physics Facility. Power, rate of rotation and rotor thrust were measured for select turbines, and hot-wire anemometry was used for flow measurements. Development of a fully developed wind turbine array boundary layer was noted with increase in array size.

Commentary by Dr. Valentin Fuster
2014;():V01DT39A011. doi:10.1115/FEDSM2014-22105.

The turbulent flow modulation on the unsteady behavior of a model VAWT is investigated and compared with a model HAWT of similar size in a laboratory experiment. The turbines operated in low and high freestream turbulence. The research was performed at the Talbot Laboratory wind tunnel at the University of Illinois at Urbana-Champaign (UIUC). High-resolution measurements of the turbine voltage for a small, 12 cm HAWT and a 16 cm VAWT are acquired at high temporal resolution, sufficient to capture the turbulent scales of flow relevant to the problem. Both turbines were built at the UIUC rapid prototyping lab and have realistic airfoil shapes. An understanding of the distinctive physical processes modulating the scale-to-scale fluctuating behavior in a VAWT and a HAWT exposed to the same turbulent flow conditions is discussed. A relation between turbulent motions and fluctuating behavior is extended from the knowledge of HAWTs to VAWTs.

Commentary by Dr. Valentin Fuster

Symposium on Uncertainty Quantification in Flow Measurements and Simulations

2014;():V01DT40A001. doi:10.1115/FEDSM2014-21230.

Pressure sensitive paint (PSP) is useful for measurements of wall pressure in high speed flows, but can be used in an alternative manner in low-speed flows as a gas species concentration detector. Film cooling technology studies have been greatly aided by this use of PSP through use of a mass transfer analogy to determine the adiabatic film cooling effectiveness. The PSP technique allows measurements that have high spatial resolution at high enough sampling rate that a good statistical mean can be determined rapidly. Due to the potential of this technique to deliver high quality adiabatic effectiveness measurements, a detailed analysis of its associated uncertainty is presented herein. In this study, an ambient temperature low speed wind tunnel drives air as the main flow while carbon dioxide (CO2, DR=1.5) is used as the “coolant” gas, though the experiments are done under isothermal conditions. A detailed analysis of the technique is performed here with focus on the measurement uncertainty and process uncertainty for a film cooling study using an array of five cylindrical holes spaced across the span of a flat test plate at a spacing of three diameters center-to-center. The final analysis indicates that the total uncertainty depends strongly on the local behavior of the coolant, with near-field uncertainty as high as 5% at isolated points. In the far-field, the total uncertainty is more uniform throughout the measurement domain and generally lower, at about 3%.

Commentary by Dr. Valentin Fuster
2014;():V01DT40A002. doi:10.1115/FEDSM2014-21577.

Recent concerns over the safety of oil and natural gas extraction, fracking, and carbon sequestration have driven the need to develop methods for uncertainty quantification for coupled subsurface flow & deformation processes. Traditional monte-carlo methods are versatile, but exhibit prohibitively slow convergence. In this work, we develop an intrusive polynomial chaos expansion method for Biot’s Poroelasticity Equations based on galerkin projection with uniform and log-normally distributed material parameters. We analyze accuracy and efficiency of our method and compare it to monte-carlo and anova based probabilistic collocation methods.

Commentary by Dr. Valentin Fuster
2014;():V01DT40A003. doi:10.1115/FEDSM2014-21886.

Both experimental and computational methods applied to the study of porous media flows are challenging due to the complex multi-phase geometry and ability to resolve scales over a reasonably large domain. This study compares experimentally obtained results based on refractive index matching of detailed velocity field vectors with those obtained using DNS to evaluate both methods for consistency. Data were obtained in a randomly packed bed using uniformly sized spherical particles. Experimental challenges including refractive index matching errors, magnification uncertainties, and the identification of the proper geometry as well as, the arduousness, of matching the geometry, grid resolution particularly near solid contact points, and proper boundary conditions DNS are presented. Detailed comparison of the numerical simulation with PIV measurements are presented by attention paid to the statistical distribution of velocities, and their deviation from DNS estimations from the measured values. There is reasonable matching the velocity fields except for some regions of constricted flow. The axial velocity results are within 12 percent and the normal velocity within 9%. Streamline details show that both methods agree well.

Commentary by Dr. Valentin Fuster
2014;():V01DT40A004. doi:10.1115/FEDSM2014-21918.

The current study investigates the flow field near a surface with a micro-PIV system using a square tube to enhance optical access. Measurements of velocity fields and eddy structures near the wall of tubes are important to the design of in-tube surface geometries. In experimental fluid mechanics, particle image velocimetry (PIV) is now a common way to measure velocity. However, PIV measurements near walls require efforts to deal with low particle density, high shear gradient and wall reflection. The current paper discusses a PIV measurement technique utilized to observe flow dynamics in near-wall regions. PIV uncertainty analysis is discussed in this study. The experimental results are compared with previous results for validation.

Commentary by Dr. Valentin Fuster
2014;():V01DT40A005. doi:10.1115/FEDSM2014-22068.

Measurement uncertainty of quantities that are functions of more than one measured variable are usually determined using propagation methods; either the Taylor Series Method or the Monte Carlo Method. Each of these requires critical assumptions that are more sweeping and potentially dangerous than commonly realized. The implications of these assumptions is explained along with some examples of how violation of these assumptions invalidates the uncertainty estimate. A new perspective based on consideration of potential error sources is presented ranging from the physical system being measured, the measurement system, the data reduction system, to the final experimental measurement result. An alternative approach that relies on sampling of error sources, whether known or unknown, is demonstrated through several examples. These examples demonstrate that the alternative approach can provide a more realistic estimate of total experimental uncertainty; one which is commonly much larger than that provided by traditional propagation methods.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In