0

ASME Conference Presenter Attendance Policy and Archival Proceedings

2014;():V01AT00A001. doi:10.1115/FEDSM2014-NS1A.
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to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.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

Advances in Fluids Engineering Education

2014;():V01AT01A001. doi:10.1115/FEDSM2014-21027.

A large number of approaches have been made to predict the total pressure loss coefficients and flow deviation angles to the geometry of turbine cascades and the incoming flow. Students feel typically uncomfortable when faced with turbine loss coefficients during their education, and it is challenging to fully understand turbine losses only by means of theory. The integration of a turbine cascade facility into academic courses might be useful but such test facilities are expensive or not available for a large number of engineering schools. To overcome this issue, a cost-efficient test rig for measurements of the flow through a two-dimensional cascade of turbine blades was designed. This test rig enabled the measurement of the flow through a blade cascade and the formation of wakes. The effect of the inlet flow angle on the cascade performance was investigated easily by students. Based on own measurements, the students were able to apply the most prominent approaches for determining loss coefficients. Furthermore, they compared their results with literature data and predictions of available correlations. By doing that, the importance of blade spacing and Reynolds number level on profile loss coefficients became more transparent and invited to further studies.

Topics: Turbines , Education
Commentary by Dr. Valentin Fuster
2014;():V01AT01A002. doi:10.1115/FEDSM2014-21106.

This paper explores modes of instruction for effective student learning and factors that affect student perceptions of information from various sources encountered in undergraduate fluid mechanics. This paper addresses two questions: (i) What source of information do students rely on and have greater confidence in, (ii) What modes of instruction lead to greater understanding of material. These research questions were addressed by considering the conceptual topic of drag on a sphere. In this study, thirty students compared results from experimental lab measurements, CFD (ANSYS-CFX) simulations, and textbook data for drag acting on a sphere. Other concepts covered in the course were done so via lecture and/or lab, but were not examined using CFD. To address the first question, students completed a survey at the end of the experimental portion of lab and a second survey at the end of the CFD portion of lab. To address the second question, student responses to specific final exam questions were analyzed. Our data indicate that students have greater reliance on materials presented via lecture and in the course textbook, than data that originates via hands-on learning methods such as experimental data, and CFD simulations. The results that address the second question indicate that even though there is greater variation in student learning outcome scores, a variety of modes of instruction lead to greater understanding of a topic, even accounting for biases in perceived data authority of various sources of data.

Commentary by Dr. Valentin Fuster
2014;():V01AT01A003. doi:10.1115/FEDSM2014-21718.

This paper presents a dynamic learning framework (DLF) for engineering courses with rich mathematical and geometrical contents. The word “dynamic” implies that there are several moving components in the course contents and assessments. Moving contents are enabled by random-number generators to select text/paragraph from a database or chose a number between two ranges within engineering bounds. Dynamic contents are usually missing in traditional form of instructions such a fixed format book-type problem or static online material. The framework leverages on the computing resources from the recent advancement in touchpad computing devices (such as IPAD and Android based tablets) and web-based technologies (such as WebGL/SVG for virtual-reality and web-based graphics and PHP based server level programming language). All assessments are developed at four increasing levels of difficulty. The levels one through three are designed to assess the lower level learning skills as discussed in the “Bloom’s taxonomy of cognitive skills” whereas level four contents are designed to test the higher level skills. The level-one assessments are designed to be easiest and include guiding materials and solved examples. To lessen the impact of disinterests caused by mathematical abstractions, the assessment and content presentations are strengthened by integrating the mathematical concepts with visual engineering materials from real-world and local important applications. All problems designed to assess the lower level skills are computerized and tested using the Computer Adaptive Testing (CAT) algorithm which enabled the instructor to focus on the higher level skills and offer the course in partially flipped classroom setting.

Commentary by Dr. Valentin Fuster

15th Symposium on Turbomachinery Flow Predictions and Optimization

2014;():V01AT02A001. doi:10.1115/FEDSM2014-21024.

This paper presents the results of modelling of the complete three-dimensional fluid flow through the spiral casing, stay vanes, guide vanes, and then through the Francis turbine runner to the draft tube of the Derbendikhan power station. To investigate the flow in the Francis turbine and also to compute stress distribution in the runner blades, a three-dimensional model was prepared according to specifications provided. The two topics discussed in this study are: (i) the simulation of the 3D fluid flow through the inter blade channels for the Francis turbine runner by using Computational Fluid Dynamics (CFD) and, (ii) the simulation of the stress analysis of the turbine runner by using Finite Element Analysis (FEA). In this study, the water pressure obtained from the CFD analysis for different boundary conditions are incorporated into a Finite Element model to calculate stress distributions in the runner.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A002. doi:10.1115/FEDSM2014-21038.

To improve the performance of the centrifugal pump with a vaned diffuser, the influence of impeller geometric parameters on external characteristics of the pump was investigated by Orthogonal Experimental Method (OEM) based on CFD. Blade outlet width b2, blade wrap angle φ, blade outlet angle β2, and blade number Z were selected as the main impeller geometric parameters and the orthogonal experiment of L9 (33*21), which contained 3 levels of the 3 factors and 2 levels of one factor, was done in this study. Three-dimensional steady simulations were conducted by solving the RANS equations in the design procedure with SST k-ω turbulence model, and about 5.3 million structured grids for the whole calculation domains were used. The experimental results were justified by the variance analysis method. The inner flow of the pump was also analyzed in order to obtain the flow behaviors that can affect the pump performance. The results showed that the blade outlet angle β2 had the greatest influence on the efficiency and power. The high efficiency area of the optimal impeller is wider. The final optimized impeller accomplished better pump performance, which meet the design requirements. The velocity distribution in the optimized impeller is more regular and the area of the high turbulence kinetic energy is smaller in the optimal impeller.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A003. doi:10.1115/FEDSM2014-21097.

This present study deals with a new mechanical device consisting of a set of safety membranes, which has been successfully applied in several middle and small hydropower stations in China instead of a surge tank. Safety membranes are installed on the penstock near the powerhouse as controlled weak points. When the pressure caused by load rejection rises to the preset explosive value, one or more membranes rupture, protecting the penstock and the unit from damage. The device is simple, reliable and economical. The method of characteristics is employed to establish numerical model of safety membranes to simulate their rupture behavior, which is then introduced to investigate how to determine the number and diameter of membranes from two aspects, large fluctuation and hydraulic disturbance. The results show that the diameter of the membranes depends on the negative pressure along pipeline under hydraulic disturbance while the number of the membrane depends on the maximum water hammer pressure under large fluctuation during load rejection of the unit. The conclusion of membrane selection can perfect the present theory of safety membranes, and provide the theoretical guidance and practical basis for membrane device design and safety operation.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A004. doi:10.1115/FEDSM2014-21200.

We developed a high-efficiency half-ducted propeller fan to reduce the electric power consumption of the outdoor unit of a packaged air conditioner by using a design tool combining computational fluid dynamics (CFD) with multi-objective optimization techniques based on a genetic algorithm (GA). The baseline fan was a half-ducted propeller fan with three blades of a currently available product. Blade shape was defined using 16 design variables including inlet and outlet blade angles, setting angles, blade length, sweep angles, dihedral angles, and so on. An in-house program was used to automatically generate the grids for CFD calculation. The objective functions were static pressure efficiency and fan noise level for optimization. The fan noise was calculated with an aerodynamic noise prediction model that used the relative inlet and outlet velocities of the fan blades from the CFD results. We found there was a trade-off relationship between the static pressure efficiency and the fan noise. We then selected the optimized fan that had the same noise level as the baseline fan but with an improved static pressure efficiency. The blade tip of the optimized fan was curled toward the suction side direction. Finally, we confirmed through experiments that the static pressure efficiency of the optimized fan was increased by 1.6% compared to the baseline fan.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A005. doi:10.1115/FEDSM2014-21324.

We report the progress made in our recent study to develop an ultra-quiet axial fan for computer cooling applications. By using a commercially-available cooling fan as the baseline, a number of acoustically tailored modifications are implemented in order to reduce the noise level of the cooling fan, which includes optimizing the rotator blades and guide vanes according to axial fan design theory, adding an intake cone in the front of the hub to guide the airflow into the axial fan smoothly, and reducing the tip clearance to lower the noise generation due to tip vortex structures. A comparison study is conducted to measure the sound pressure level (SPL) of the reformed axial fan in an anechoic chamber, in comparison to that of the prototype fan, in order to assess the effects of the modifications on the fan noise reduction. The measurement results of our preliminary study reveal that, at the same flow rate, the SPL of the reformed fan would be up to 5 dB lower than that of the prototype fan. In addition to measuring the sound pressure levels (SPLs) of the fans, a digital particle image velocimetry (PIV) system is also used to conduct detailed flow field measurements to reveal the changes of the flow characteristics and unsteady vortex structures associated with the modifications. 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 at the exit of the axial fan, “phase-locked” PIV measurements are also performed to elucidate further details about evolution of the unsteady vortex structures in fan exhaust in relation to the position of the rotating fan blades. The detailed flow field measurements are correlated with the SPL measurements in order to elucidate underlying physics associated with the fan noise reduction.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A006. doi:10.1115/FEDSM2014-21479.

In this study, the effect of a row of double-jet film-cooling hole configurations on the thermal-flow characteristics of gas turbine blades was examined. To investigate the effect of the interference of anti-kidney vortices, the ratios of the pitch distance and hole diameter (P/d=5, 6.25, 8.333) were considered with two different compound angles (λ=0°, 4°). The film cooling performance and the generated losses were studied. Then, the relevant mechanisms were identified and explained. A numerical study was performed using ANSYS CFX 14.5 with the shear stress transport (SST) turbulent model. The blowing ratio was kept at a constant value of M=1.5. The film cooling effectiveness and temperature distribution are graphically depicted with various geometrical configurations.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A007. doi:10.1115/FEDSM2014-21595.

Counter-rotating blade rows in single stage machines have been widely investigated in several applications like naval propulsion [1], axial fan [2] and axial pump design [3]. Previous publications have presented this design philosophy as a promising way to improve efficiency and cavitation performance [4]. Differently the goal of the presented design is to achieve lower required rotational speeds and at the same time a higher power density ratio, indirectly improving also the cavitation’s performance, aspiring actually to a compact design. However, a literature survey did not produce any evidence of significant advantage in axial-flow machines from this design philosophy in terms of machine power density. This paper presents a preliminary analysis of the applicability of the counter-rotating impeller concept on mixed-flow and radial-flow pumps. The design example presented in this paper constitutes a mixed-flow first rotor with a radial second rotor that rotates in opposite direction. A 1-D design model demonstrates that power density coefficient has a maximum for optimum speeds of the rotors range. The design example parameters were selected based on the highest power density coefficient and efficiency. Unlike most published literature where the rotor speed ratios are not fixed, instead the speeds can be independently varied through two separate electric drives. Both the front and rear rotor were developed using multi-streamline curvature analysis which combines fluid dynamic loss model [5] with a slip model at the rotor exit [6]. Several hypotheses were considered to determine the most significant and independent parameters that run through a response surface optimizer tool, scripted in MATLAB®. It detects optimum specific speed for the rotors which maximizes the benefits of counter-rotating impeller design compared to rotor-stator machine with same design point. Assuming steady numerical approximation error, the fitness function of the optimizer tool is based on steady state CFD results. The fitness function depends on total head and hydraulic efficiency, which show a maximum as a result. Optimum geometries were identified and one was chosen for manufacturing and testing in a test rig. Transient CFD analysis was also done to determine volute losses and incidence losses between first and second rotor and pressure pulsations generated. At this stage, manufacturing of hydraulic components of test model is in progress, while the mechanical design of the machine is close to completion.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A008. doi:10.1115/FEDSM2014-21647.

In this study, the relationship between the inlet relative humidity (RH) condition, heat transfer, and droplet accumulation/motions on gas turbine’s compressor blades involved in enhanced film cooling was investigated. Wet compression has gained popularity as a highly effective way to increase power output for gas turbine systems due to its simple installation and low cost. This process involves injection of droplets into the continuous phase (air with high temperature) of the compressor to reduce the temperature of the flow leaving the compressor and in turn increase the power output of the whole gas turbine system. This study focused on a single stage rotor-stator compressor model; the simulations are carried out using the commercial CFD tool ANSYS (FLUENT). In particular, the study modeled the interaction between the two phases including mass and heat transfer, given different inlet relative humidity (RH) and temperature conditions. The Reynolds Averaged Navier-Stokes (RANS) equations with kϵ turbulence model were applied as well as the droplet coalescence and droplet breakup model considered in the simulation. The interaction between the blade and droplets was modeled to address all possible interactions, which include: stick, spread, splash, or rebound. The goal of this study is to quantify the relationship between the RH and the overall heat transfer coefficient, and the temperature on the blade surface.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A009. doi:10.1115/FEDSM2014-21752.

Turbomachinery blade design improvement and optimization by CFD is a time-consuming engineering challenge. Such an optimization process, which requires advanced numerical simulations, uses a large amount of computational resources to provide the required solutions. This paper presents a turbine blade optimization process which uses an algorithm based on response surface methodology (RSM) to increase the simulation speed. The main idea of RSM is to start with a lower number of sample points to generate an analytical model that describes the relationship between the pre-defined numbers of design parameters. In this study, the Kriging approximation is used to generate the surface model. The global minimum on the surface is searched by applying a gradient method. The increase in the convergence speed is achieved by using an adaptive scheme, which creates additional points around the previous minimum while reducing the solution space at each iteration step, until convergence is achieved. Each iteration step is composed of several CFD simulation runs where each point represents different designed geometries inside the n-dimensional parameter space. The process combines a Bezier-spline based airfoil-generator with a parametric meshing tool —G3DMESH— and a CFD solver —TRACE—, both developed and provided by DLR, into a MATLAB script function. A particular characteristic of this optimization method is its lower evaluation number requirement to reach convergence, as well as its capability to run multiple simultaneous RANS. The optimizer process was initially tested by using basic functions to analyze its solution behavior and its performance in comparison to a genetic algorithm (GA) type optimizer. It is observed from this comparison that RSM optimization reaches the convergence faster and more stable than the GA method applied on the test case. Preliminary optimization results show an improvement in function evaluation requirements by up to 50%, which depends on the complexity of the respective surface model of the test case. As an application, a 4-stage low pressure turbine for a turboprop engine is designed by multi-streamline analysis. 2D mid-span cross-sections from both rotor and stator are produced by the Bezier-spline based airfoil-generator. The basis tool requires input parameters as leading and trailing edge blade angles and maximum thickness position. The blade generator is further improved by the additional ability to work with high values of deviation angles between the leading and trailing edges, up to 90°. 6 control points are used to define the two curves, for pressure and suction sides, which encompass the cross-section geometry. Optimization process runs to improve these airfoil parameters. The 2D airfoils of the first stage are optimized by an objective function based on total pressure loss coefficient at the engine on-design point. The same geometry is also optimized using the GA method as a comparison case.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A010. doi:10.1115/FEDSM2014-21773.

A direct numerical simulation (DNS) code has been developed to solve the compressible Navier-Stokes equations. This code is capable of handling unstructured grids, often a requirement for simulating flow past complex geometrical shapes like turbine and compressor blade passages. The code is validated for a compressible turbulent channel flow.

Flow past a low pressure (LP) turbine blade passage is computed at Re = 51,831 based on axial chord length and incoming velocity, and angle of incidence = 45.5° with respect to axial chord. Several interesting fluid-dynamical phenomena including multiple separation bubbles and laminar-turbulent-laminar transition cycles are noticed on the suction side of the blade. Also a strong curvature effect is seen in the boundary layer near the leading edge on the suction side, where use of higher order boundary layer theory will be needed.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A011. doi:10.1115/FEDSM2014-21969.

A generalized model for mapping the trend of the performance characteristics of a double-discharge centrifugal fan is developed based on the work by Casey and Robinson (C&R) which formulated compressor performance maps for tip-speed Mach numbers ranging from 0.4 to 2 using test data obtained from turbochargers with vaneless diffusers. The current paper focuses on low-speed applications for Mach number below 0.4. The C&R model uses four non-dimensional parameters at the design condition including the flow coefficient, the work input coefficient, the tip-speed Mach number and the polytropic efficiency, in developing a prediction model that requires limited geometrical knowledge of the centrifugal turbomachine. For the low-speed fan case, the C&R formulas are further extended to a low-speed, incompressible analysis.

The effort described in this paper begins by comparing generalized results using efficiency data obtained from a series of fan measurements to that using the C&R model. For the efficiency map, the C&R model is found to heavily depend on the ratio of the flow coefficient at peak efficiency to that at the choke flow condition. Since choke flow is generally not applicable in the low-speed centrifugal fan operational environment, an alternate, but accurate estimation method based on fan free delivery derived from the fan test data is presented. Using this new estimation procedure, the modified C&R model predicts reasonably well using the double-discharge centrifugal fan data for high flow coefficients, but fails to correlate with the data for low flow coefficients. To address this undesirable characteristic, additional modifications to the C&R model are also presented for the fan application at low flow conditions.

A Reynolds number correction is implemented in the work input prediction of the C&R model to account for low-speed test conditions. The new model provides reasonable prediction with the current fan data in both work input and pressure rise coefficients. Along with the developments for the efficiency and work input coefficient maps, the use of fan shut-off and free delivery conditions are also discussed for low-speed applications.

Commentary by Dr. Valentin Fuster
2014;():V01AT02A012. doi:10.1115/FEDSM2014-22241.

The primary design goal of a compressor is focused on improving efficiency. Secondary objective is to widen the compressor’s operating range. This paper presents a numerical and experimental investigation of the influence of the bleed slot to enlarge operating range for the 1.2MW class centrifugal compressor installed in a turbocharger. The main design parameters of the bleed slot casing are upstream slot position, inlet pipe slope, downstream slot position and width. The DOE (design of experiment) method was carried out to optimize the casing design. Numerical analyses were done by the commercial code ANSYS-CFX based on the three dimensional Reynolds-averaged Navier-Stokes equations. From the analysis, as the downstream slot position and width are smaller and upstream position is located away from impeller inlet, efficiency and pressure ratio are increased.

Experimental works were done with and without the bleed slot casing. The simulation results were in good agreement with the test data. In case without the bleed slot casing, the surge margin value came out to be only 11.8% but with the optimized bleed slot design, the surge margin reached 23%. Therefore, the surge margin increase of 11.2% was achieved.

Topics: Compressors , Design , Surges
Commentary by Dr. Valentin Fuster

Symposium on Applications in CFD

2014;():V01AT03A001. doi:10.1115/FEDSM2014-21125.

This paper presents a 3D Computational Fluid Dynamics (CFD) modeling of flow, combustion and heat transfer processes into an internal enclosure acting as a combustion chamber, confined by the newly patented air cooled “corner ring”, the lower shaft vertical side walls, the vault and the limestone packed bed, located in a vertical twin-shaft regenerative lime kiln. The numerical simulation is restricted only to the kiln first start-up preliminary phase, with the goal to optimize the thermo-fluid dynamics patterns established during the first heat-up of wet gunning refractory concrete lining of the air cooled “corner ring”, to avoid refractory damages.

The present work is performed in the frame of the commercial general-purpose code ANSYS-CFX R14.5. The CFD model is run under transient flow conditions accomplished by the drying burners operated in single-stage “on-off” control mode, to fit at the best the heat-up curve by optimization of the fluid dynamics patterns, with the goal to prevent local hot spots on the refractory lining. The industrial data collected through the supervision system and the local provisional instrumentation on the vertical twin-shaft regenerative lime kiln, model RD15, commissioned in India on September last year, are used to set the test-case and to partially validate the numerical simulation results.

This CFD numerical simulation represents an useful engineering tool, on behalf of refractory designer and commissioning engineer, for the prediction of the refractory lining behavior during the kiln first start-up.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A002. doi:10.1115/FEDSM2014-21131.

Ground effect is a phenomenon caused by the presence of a fixed boundary layer below a wing. This results in an effective increase in lift to drag ratio of the airfoil. The available literature on this phenomenon is very limited; also the types of airfoils used in traditional aircrafts are not suited for ground effect vehicles, so a computational study has been done comparing traditional airfoils (NACA series) with ground effect airfoil (DHMTU).

In this paper, the aerodynamic characteristics of a NACA 6409, NACA 0012, DHMTU 12-35.3-10.2-80.12.2[1] section in ground effect were numerically studied and compared. In 2D simulation, the flow around each of the airfoils has been investigated for different turbulence models viz. Spalart Allmaras turbulence model and k-ε Realizable turbulence models. Lift coefficient, drag coefficient, pitching moment coefficient and lift to drag ratio of each airfoil was determined on several angles of attack from 0 to 10° (0°, 2°, 4°, 6°, 8°, 10°) and different ground clearances (h/c=0.2, 0.4, 0.6, 0.8, 1.0). The results of the CFD simulation indicate a reduction in drag coefficient and an increase in lift coefficient, thus an overall increment in lift to drag ratio of the airfoils, when flying in proximity to the ground. Also DHMTU airfoils have shown a greater consistency in Cm behavior with decreasing height-to-chord (h/c) ratio.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A003. doi:10.1115/FEDSM2014-21156.

An integrated experimental and numerical investigation was carried out to gain insight into the heat transfer phenomena and flow characteristics inside a domestic refrigerator. A refrigerator model was constructed using insulation foam sheets according to the inner dimensions of a household refrigerator. A reversal heat leak analysis was conducted on the constructed model in a temperature-controlled chamber, where the chamber temperature was lower than the inner temperature of the refrigerator. A temperature-controlled heater was mounted where the evaporator was. The heater was enclosed in a heater box to heat the air and to maintain a high temperature in the refrigerator. A variable speed fan was used to force air circulation. Thermocouples were used to measure the temperature at specified positions and to measure the average temperature difference across the refrigerator side walls. The correlation between the status of the heater and the control temperature variation pattern was analyzed. Heat loss rate was calculated using the data from the thermocouples too. The calculated heat loss rate closely matched the generated heat by the heater and the fan. Moreover, according to the results with different input voltages, the variation trend of the heat flux density was analyzed. A conjugate heat transfer analysis was conducted based on the constructed model using Fluent. The heater was modeled as a heat volume source and the fan was modeled using a pressure jump condition based on the experiment result. Comparisons were made between the experimental and numerical results. The predicted heat loss rate and the heat flux density through the walls matched very well with the experimental results. And the variation trend of the heat flux density with different input voltages also showed the same trend as the experimental result. And the airflow pattern and the temperature distribution were also analyzed in detail.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A004. doi:10.1115/FEDSM2014-21171.

This study is a comparison of two techniques for simulation of particulate flows on fixed Cartesian grids: Sharp interface Method (SIM) (Udaykumar et al., 2001, 2002, 2003) and a modified version of Immersed Boundary Method (Peskin, 1977) (IBM) known as Smoothed Profile Method (SPM) (Nakayama and Yamamoto, 2005; Luo et. al, 2009). Different cases were studied includes flow over one or two moving and stationary particles. Predictions of the drag coefficient shows that SPM and SIM are very close to the experiments. SIM slightly under-predicts the value of the drag coefficient while SPM has a small over-estimation. Moreover, SPM is more accurate on coarse grids. However, with refinement of the grid SIM approaches the exact values very fast leading to better results on fine grids. Flow pattern and vortex structures of SPM and SIM are almost the same. Both methods are capable of analyzing the wake flow. Unlike SIM, SPM is able to simulate the flow when two particles are in contact. When two particles are in motion and are very close in a way that the two interfaces overlap, SPM shows a repulsion force between two spheres which reduces the accuracy in comparison with SIM. However, SPM can achieve the collision of two particles without problem.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A005. doi:10.1115/FEDSM2014-21191.

In order to invent a new near-wall treatment for turbulence in Computational Fluid Dynamics (CFD) simulation, an Analytical Wall Function (AWF) has been studied and shown that it is possible to work accurately with Reynolds Averaged Navier-Stokes (RANS) Simulation even for complicated geometry such as impinging jet flow or separation and reattachment flow. One of the most common wall functions is the Standard Wall Function (SWF) which assumes log-law inside the boundary layer. However, there is a problem that SWF has been used for industrial applications even though it is difficult to analyze the turbulence phenomenon in a complicated geometry accurately because log-law is not applicable in that geometry. On the other hand, since AWF derives the boundary condition on the wall by integrating analytically the boundary layer equation in wall adjacent cells, it can analyze the turbulence accurately even in complicated geometry. AWF has an advantage over SWF from this point of view.

In this study, AWF was improved and optimized for Large Eddy Simulation (LES) by changing the way of modeling of eddy viscosity inside the boundary layer for steady state simulation to that for unsteady state simulation. This is because RANS is a steady state simulation; on the other hand, LES is unsteady state simulation, which is one of the largest differences between them. The accuracy of the new AWF for LES (LES-AWF) was validated by both of experimental results and CFD simulation results. Both of the experiment and CFD simulation are conducted in the wind tunnel.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A006. doi:10.1115/FEDSM2014-21208.

The wake flow behind two intersecting flat plates forming a cross is studied by means of direct numerical simulation (DNS) at low Reynolds number. The Reynolds number based on the plate width, d, and the inflow velocity, U0, is 100. The flat plate structure is in one plane. Away from the intersecting center part, vortex streets can be observed similar to the wake flow behind a single normal flat plate. On the other hand, the flow is completely three-dimensional in the vicinity of the intersecting region where the wakes from each of the two normal flat plates interact with each other. The mean pressure distribution on the structure is evaluated in order to study the total drag force as well as the local force distribution.

Topics: Wakes , Flat plates
Commentary by Dr. Valentin Fuster
2014;():V01AT03A007. doi:10.1115/FEDSM2014-21264.

A finite volume numerical approach is used to study the steady, laminar, plane wall jet that evolves from a parabolic velocity profile with uniform temperature to its self-similar behavior downstream of the jet exit. A variety of Reynolds numbers ranging between 50 and 250 is considered in this numerical investigation. The working fluids are air and water with constant physical properties corresponding to Prandtl number of 0.712 and 7 at ambient conditions. In these types of flows, a developing region over which the flow converges to its self-similar behavior is observed in the vicinity of the jet exit. The location of the dimensionless virtual origin, which is of main importance in determining the flow field in the self-similar region, is carefully studied and correlated as a function of Reynolds number. The local skin friction coefficient is observed to converge to the analytical self-similar solution at downstream locations. Given that an analytical solution for the thermal behavior of this problem doesn’t exist in either the developing or self-similar regions, the thermal solution of this problem is studied for isothermal and uniform heat flux boundary conditions at the wall. The idea of a dimensionless thermal virtual origin is introduced and correlated as a function of Reynolds number. The Nusselt number dependence on Prandtl number, Reynolds number and the downstream location are obtained for both thermal boundary conditions at the wall.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A008. doi:10.1115/FEDSM2014-21297.

It has long been realized that condensation in a chamber prefilled with condensable vapor leads to chamber depressurization, and the condensation rate can be cooling controlled. While the final state can be reasonably estimated based on the thermodynamic equilibrium, the dynamic process or rate of depressurization has not been satisfactorily modeled, which is due to the complicated coupling mechanisms of heat and mass transfer, the transient nature of non-equilibrium during the process, the complication by the co-existence of non-condensable gas (NCG) within vapor, as well as the complex geometry and material properties of chamber and cooling device involved. In this paper, we have conducted an experimental study on depressurization by steam condensation onto an internal cooling coil in a steam-prefilled closed chamber. To reveal various parametric effects on the depressurization process, a parametric model consisting of a set of coupled ordinary differential equations has been established, with some simplified assumptions including lumped heat capacity sub-models for chamber walls, cooling coils and the gas phase. To further explore the thermal non-equilibrium characteristics during the process, a simplified and transient simulation of computational fluid dynamics (CFD) is also conducted using FLUENT with user-defined function (UDF) on boundary of condensation. Both parametric and CFD models consider the existence of NCG that is pre-mixed with the vapor as impurity. By comparison with the experimental measurements, both models correctly predict the dynamic and asymptotic characteristics of depressurization with time. The CFD simulation indicates an almost instant equilibrium in pressure within the chamber and yet non-equilibrium in temperature with noticeable temperature gradients over the gas phase. The simplified parametric model provides quick and quantitative assessments of some major parametric effects (e.g., vapor purity, coolant flow rate, and vessel volume) on the rate of depressurization. The detailed mechanistic understanding, gained from proposed models, provides insights essential to the optimized design and operation of the depressurization by cooling-controlled condensation.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A009. doi:10.1115/FEDSM2014-21310.

The objective of this research is to develop models that represent the effects of light and algae and incorporate these effects within a computational fluid dynamics (CFD) model of a photobioreactor (PBR). Several factors, including nutrient availability, carbon dioxide concentration, light intensity, and frequency of high and low light intensity periods, affect the efficiency of biomass yield within a photobioreactor. However, even with a general understanding of the affecting factors, scaling up of photobioreactors from a laboratory to a commercial level exist and provide a challenge concerning efficiency. The development and execution of an integrated light, algae, and CFD model can provide insight into more cost and time efficient configurations of PBRs. In depth CFD studies have been used to predict thermal-fluid effects, including bubble-liquid interaction and temperature profiles; however, studies concerning algae-liquid interactions appear sparsely. In order to better understand up-scaling issues, new modifications of previous CFD methods incorporate an algae particle tracking method, as well as light modeling. The particle tracking method considers the individual algae cell as a volume-less and mass-less particle that follows the liquid velocity profiles within the PBR. The light model takes into account algal concentration as well as bubble location and bubble concentration. The integration of the models allows for the average intensity of light experienced by an algae cell to be numerically estimated, alongside the frequency of light and dark periods the particle experiences. The long term goal of this research is to develop an algae growth model that incorporates light intensity and the flashing light effect. The present research is a continuum of previous work aimed at pursuing the optimum design of a column PBR which is commercially viable and effective at producing algal biofuels and bioproducts.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A010. doi:10.1115/FEDSM2014-21327.

A new numerical simulation methodology for turbulent flows of viscoelastic fluid was developed for engineering application purpose based on commercial computational fluid dynamics code FLUENT package. An in-house subroutine was established and embedded into FLUENT code through userdefined function functionalization. In order to benchmark this methodology, numerical simulations on turbulent channel flows of viscoelastic fluid are conducted under different cases with drag reduction rates varied from low level to high level. FENE-P (finitely extensive nonlinear elastic-Peterlin) constitutive model is used to describe the viscoelastic effect of viscoelastic fluid flow. The turbulent model is developed in the framework of Display Formulakεν2¯f model, for which the elliptic relaxation model is modified to account for the Reynolds stress equilibrium established by the presence of elasticity in the fluid. The numerical simulation results, including velocity profiles, turbulent flow characteristics, elastic stress and conformation field, show good agreements with published DNS results, which validates the newly established method on turbulent flows of viscoelastic fluid based on FLUENT software platform for engineering applications.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A011. doi:10.1115/FEDSM2014-21429.

At the crime scene in case of homicide, spattered bloodstains at the incident site are important evidence. The patterns of the bloodstain on the floors and walls are determined by the impact conditions of blood drop such as drop size, impact angle and velocity. By analyzing the bloodstain pattern, one can retrace the origin of blood drops to reconstruct the crime scene. In the present study, motions of blood drop are analyzed to figure out the correlation between impact conditions of blood drop and bloodstain patterns. Two phase interfacial flows of blood drop impacting the wall are numerically simulated with the VOF method. To get the accurate results of bloodstains on the wall, non-Newtonian fluid viscosity and dynamic contact angle are used as the blood properties. By using these methods, patterns of bloodstains created from the blood drop are predicted for various Reynolds and Weber numbers.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A012. doi:10.1115/FEDSM2014-21510.

In this paper, numerical solutions of fluid flow in the vicinity an oscillating square cavity are presented. OpenFOAM (v 2.2.2), an open-source, Finite Volume CFD code was used to solve the problem. The oscillating cavity problem was studied with respect to the following parameters: (1) peak Reynolds numbers (based on the cavity size, Red) of 50, 100, 200, and 300, and (2) the ratio of the Stokes layer thickness to the cavity size (δ/d) of 0.25, 0.5, and 1. An oscillatory source term was provided to the streamwise momentum equation and the problem was solved with a stationary grid. The resultant fluid velocities were then vectorially corrected for the wall velocity. The patterns and the magnitudes of entrainment and ejection of fluid mass to and from the cavity were studied. Lower Red flows were marked by the absence of a secondary recirculation zone inside the cavity in contrast to higher Red flows. Lower Red flows were observed to have fluid contact with the outside shear layer for a larger portion of the oscillatory cycle, and thus, were observed to result in higher effective fluid transport across the cavity. Higher δ/d ratios resulted in decreased peak mass flow across the cavity aperture plane and only over a smaller portion of the cycle time owing to the thickness of the oscillatory boundary layer in relation to the cavity size. As the δ/d was lowered, effective mass flow was observed to increase and over a larger portion of the cycle time.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A013. doi:10.1115/FEDSM2014-21547.

Precipitation crystallization is one possibility to produce nano-scaled solid particles from the liquid phase. High nucleation and growth rates are generated by mixing two well soluble reactants and their subsequent reaction to a sparingly soluble product. These primary processes can be very fast. Therefore experimental access to internal parameters is given insufficiently due to predominantly very short process times. Computational Fluid Dynamics (CFD) based methods are a promising tool to gain insight into those inaccessible processes. Unfortunately, 3D modeling of complex precipitation reactors poses enormous difficulties and computational costs to CFD especially in the production scale under the aspect of macroscopic flowfields down to microscale modeling of mixing, rheology and particle formation. Therefore, a new methodic approach is presented that is able to handle these complex interactions. Due to local and temporal multiscale complexity, it is not advisable to model the complete apparatus. One basic principle of the methodical consideration is the arrangement of cross-linked compartments to reduce the huge unsimulatable control volume in its complexity and dimensions. Thereby, population balance equations (PBE) are solved, using CFD measured, average state variables, with a discrete one-dimensional High Resolution Finite Volume (HRFV) algorithm. Nevertheless appropriate fundamental kinetics for primary and secondary processes have to be implemented. Besides the new methodic approach, this paper deals with the influence of temporal supersaturation buildup on the product particle distribution. It is shown that important conclusions about the mixing behavior of Confined Impinging Jet mixers (CIJMs) can be drawn by coupling CFD and external population balancing even without any micromixing model. The contribution provides an insight into the methodic approach and first derived results.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A014. doi:10.1115/FEDSM2014-21557.

Analysis and control of flow structure in U-bends are crucial since U-bends are used in many different engineering applications. As a flow parameter in U-bends, the ratio of inertial and centrifugal forces to viscous forces is called as Dean number. The increase of Dean number destabilizes the flow and leads to a three-dimensional flow consisting of stream wise parallel counter-rotating vortices (Dean vortices) stacked along the curved wall. Due to the curvature in U-bends, the flow development involves complex flow structures including Dean vortices and high levels of turbulence that are not seen in straight duct flows. These are quite critical in considering noise problems and structural failure of the ducts. In this work, computational fluid dynamic (CFD) models are developed using ANSYS FLUENT to simulate these complex flows patterns in square sectioned U-bend with a radius of curvature Rc/D=0.65. The predictions of mean velocity profiles on different angular positions of the U-bend are compared against the experimental results available in the literature and previous numerical studies. Performance of six different turbulence models are evaluated, namely: the standard k-ε, the k-ε Realizable, the k-ε RNG, the k-ω SST, the Reynolds Stress Model (RSM) and the Scale-Adaptive Simulation Model (SAS), to propose the best numerical approach with increasing the accuracy of the solutions while reducing the computation time. Numerical results show remarkable improvements with respect to previous numerical studies and good agreement with the available experimental data. The best turbulence model for this application is proposed considering both the computation time and the result accuracy. In addition, different flow control techniques are still under investigation to eliminate Dean vortices and to reduce turbulence levels in U-bends.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A015. doi:10.1115/FEDSM2014-21618.

Indoor air quality (IAQ) is very important to human health and comfort as increasingly people spent most of their time in indoor environment. Numerical simulation of indoor airflows has become a significant tool for investigation of the indoor air quality. Cost effective computational methods with reasonable accuracy have the advantage of being more accessible to designers compared to more precise but expensive DNS methods. Recently developed Lattice-Boltzmann Method (LBM) has proved to be a powerful numerical technique for simulating fluid flows in various applications. In comparison with the conventional CFD methods, the advantages of LBM are: simple calculation procedure, simple and efficient implementation for parallel computation, and easy and robust handling of complex geometries. The indoor airflow is typically in turbulent flow regimes. Due to the high costs of more accurate direct numerical simulation (DNS) and large eddy simulation (LES), in this study the Reynolds Averaged Navier-Stokes (RANS) method was used for analyzing the turbulent flow conditions. The RANS governing equations, and in particular, the k-ε turbulence model was incorporated into the Lattice-Boltzmann computational method. The simulation results showed that the combined LBM-RANS provide a reasonably accurate description of the airflow behavior in the room at modest computational cost.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A016. doi:10.1115/FEDSM2014-21620.

It is well known that the airflow is instrumental in the transmission of airborne infectious diseases in indoor environments. The airflow pattern in indoor environment is affected by the ventilation airflow, thermal plume around human bodies, human respiration, human motion and other activities. In this study, the CFD approach was used to simulate airflow field and particle transport in a room to provide exposure assessment for a heated breathing manikin with and without rotational motion. The simulation results indicated that the rotation of the manikin significantly impacts the thermal plume of the body and the associated transport of particulates.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A017. doi:10.1115/FEDSM2014-21831.

Thermal powerplants report a reduction in their dry cooling tower performances due to surrounding wind drafts. Therefore, it is very important to consider the influence of wind velocity in cooling tower design; especially in geographical points with high wind conditions. In this regard, we use the computational fluid dynamics (CFD) tool and simulate a dry cooling tower in different wind velocities of 0, 5 and 10 m/s. To extend our calculations; we also consider the temperature variation of circulating water through the tower heat exchanger or deltas one-by-one. We show that some heat exchangers around the tower cannot reduce the circulating water temperature sufficiently. This causes an increase in the mean temperature of those heat exchangers. The worst performances can be attributed to heat exchanger located on side wind places. We will discuss the detail performance of each delta and their assembly in draft wind conditions. This study suggests some effective ways to overcome thermal-performance of cooling tower in wind conditions.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A018. doi:10.1115/FEDSM2014-21851.

The drive to obtain more accurate petrophysical information from deeper wells has led to the demand for operating various downhole tools at higher temperatures for longer time periods. If the borehole temperature reaches values higher than 175°C, it is considered a high temperature (HT) well. In HT wells, reliability of the electronic components of the logging tools is a major concern. One way to address the concern is through using a thermal flask to reduce the heat flow rate from the formation to the tool and evenly distribute the heat generated by the internal electronic components.

Optimizing the design of the aforementioned thermal flask is very important in providing a longer operative time for the tool before the temperature of the sensitive electronic parts reaches a critical threshold. To obtain the sensitive parameters for designing the flask, the thermal transport inside the tool must be accurately modeled.

In this work, high fidelity FEA and CFD-based transient thermal models are developed for thermal transport in a flask for an ultra-high temperature wireline tool. Two models with different levels of complexity are presented. The models are verified by experimental results and the physical insights obtained from them presented. The predictive capability of the models is used to provide recommendations for safe operating time for various environmental conditions which prevail in the formations. The results obtained from the models can also be used for optimizing the performance of the future generation of the tool and reducing the amount of time spent in unnecessary trip outs.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A019. doi:10.1115/FEDSM2014-21868.

In this study, numerical simulations using unsteady Reynolds-Averaged Navier-Stokes (RANS) approach with Shear Stress Transport (SST) k-ω turbulence closure are employed to investigate the wind loads and wind flow field of a ground mounted solar panel array. Atmospheric boundary layer wind profiles for open terrain roughness with Reynolds number of 2.2×106, based on the wind speed at the lower edge and the chord length of a stand-alone system, are employed. Four different wind directions (0°, 45°, 135° and 180°) are considered. The numerical modeling approach employed in this study is validated for a stand-alone solar panel system by comparing the surface pressures with the study by [1] and the velocity field with a Particle Image Velocimetry (PIV) measurement carried out in the Boundary Layer Wind Tunnel I at the Western University, Canada. Analyzing the wind flow field for the array configuration shows that for 0° and 180° wind directions, all trailing rows are in the complete wake of the first windward row. It is also shown that in terms of maximum drag and lift, 0° and 180° wind directions are the critical wind directions with the first windward row being the critical row. On the other hand, in terms of overturning moment, 45° and 135° are the critical wind directions, with similar overturning moment coefficients for each row.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A020. doi:10.1115/FEDSM2014-21986.

An Eulerian-Eulerian approach was used to model adiabatic bubbly flow with CFD techniques. The OpenFOAM® solver twoPhaseEulerFoam was modified to predict upward bubbly flow in vertical pipes. Interfacial force and bubble induced turbulence models are studied and implemented. The population balance equation included in the two-fluid model is solved to simulate a polydisperse flow with the quadrature method of moments approximation. Two-phase flow experiments with different superficial velocities of gas and water at different temperatures are used to validate the solver. Radial distributions of void fraction, air and water velocities, Sauter mean diameter and turbulence intensity are compared with the computational results. The computational results agree well with the experiments showing the capability of the solver to predict two-phase flow characteristics.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A021. doi:10.1115/FEDSM2014-22087.

Nanoparticle production in flames was modeled in an Eulerian-Lagrangean framework, considering droplet evaporation and fuel combustion to predict the flame chemical species concentration and the flame temperature fields by means of Computational Fluid Dynamics (CFD). A mathematical model was carried out considering two-way coupling between the gas phase and the droplets. For the combustion model, the eddy dissipation concept model was applied, taking into account the droplets vaporization, the chemical reaction mechanisms, and the chemistry-turbulence interaction. 2D axisymmetric and 3D approaches were investigated in standard operations conditions. The initial conditions for the droplet sizes and droplet velocities were taken in experiment test facility by means of Laser-Diffraction. The grid independence study was made according to the Grid Convergence Index (GCI) methodology for both approaches. The droplets mass evaporated, temperature and velocities profiles were used to compare the 2D and 3D results. The results show similar behavior for both approaches, however, with some quantitative difference. The 2D approach showed lower temperature resulted by a larger mass fuel not evaporated and unburned.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A022. doi:10.1115/FEDSM2014-22088.

A preliminary numerical study of the fluid dynamic behavior of flat and ribbed square duct is presented. Fluid dynamics of two configurations is analyzed via Reynolds Averaged Navier Stokes (RANS) modeling in order to underline the main characteristics of each configuration and to give some information at global and local level. This kind of modeling is used as base for setting up a more detailed analysis such as Direct Numerical Simulation (DNS).

Flat and ribbed square duct with a Reynolds number based on bulk velocity and hydraulic diameter of 10320 (Reτ=600 for the flat configuration) are analyzed and the results of the flat configuration are compared with available results obtained via DNS approach.

The ribbed square duct is characterized by a two-pass configuration (aligned ribs in the top and bottom walls), with the duct height about six times the height of the obstacles for a blockage ratio of about 30%. Finally a pitch ratio (rib spacing to rib height) of 10 is used in order to obtain a k-type roughness permitting the flow to reattach before the next obstacle.

Advanced U-RANS and RANS models such as Reynolds Stress Models (RSM), Explicit Algebraic Stress Models (EASM), v2f, and two-equation low Reynolds models have been used and compared.

In this direction the validated results obtained for the smooth square duct are used to select the most appropriate RANS models and to evaluate their performance for a ribbed square duct at the same Reynolds number.

A periodic configuration including one rib and considering the flow fully developed is used to reduce computational costs. The results of the ribbed square duct are hence analyzed in order to evaluate characteristics such as domain and mesh size.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A023. doi:10.1115/FEDSM2014-22089.

Indoor air quality is an important issue involved in a wide variety of industrial applications. In an indoor environment, different types of contaminants exist and have an inevitable potential to cause health problems for human beings and animals. In this study, the focus is on the contaminant contained in painting materials. While painting materials being sprayed to solid surfaces, pollutant plumes are formed near the painting area, which may enclose the body parts of the sprayers. Severe health problems are possible to occur if a significant amount of painting materials settles on the face of workers. By applying exhaust conditions (i.e. exhaust fan with outlet velocity), the flow convection in the room can be enhanced, which may alleviate the contaminant level on the human body. In such a case, the choice of exhaust condition becomes crucial. With the aid of computational fluid dynamics, an optimal exhaust condition can be determined. To simulate this kind of fluid/solid-particle multiphase flow, the current study employs a pure Eulerian or Euler-Euler type model. In the Euler-Euler approach, the properties of the contaminant particles are assumed to be continuous as those of fluids and all phases are computed in the Eulerian framework. Since the exhaust speed is moderately low and fully turbulent flow is not guaranteed in the room, the RNG k-e model is used as a low Reynolds number turbulent model. The current paper firstly investigated the scenario of sprayer self-contamination. Then, inter-contaminations among different workers will be studied.

Topics: Exhaust systems
Commentary by Dr. Valentin Fuster
2014;():V01AT03A024. doi:10.1115/FEDSM2014-22158.

Experiments with submarine models in cavitation tunnels are limited by the length scale of the body in relation to the size of the test section. Especially for high angles of attack, the body will experience strong flow interference due to the proximity of the walls. Consequently, the hydrodynamic coefficients will normally require empirical corrections in order to be extrapolated to represent open-water flows. The present work investigates the effect of local hydrodynamic blockage, i.e. the ratio between the body frontal area to the cross area of the test section, on the determination of the hydrodynamic coefficients of a prolate spheroid. Numerical simulations of the flow around the body in various angles of attack were performed solving the Re-averaged Navier-Stokes equations for Reynolds numbers in the range of 60 million. The numerical domain is built to represent the test section of a large cavitation tunnel in full scale. Results for coefficients of lift, drag and pitching moment are compared for several cases with blockage ratio of 0%, 1% and 5% and angle of attack of 0 and 10 degrees. The maximum deviation of around 30% from the reference case was obtained at the angle of attack of 10 degrees for the highest blockage ratio. The topology of the flow showed no significant dependency on blockage for the tested range.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A025. doi:10.1115/FEDSM2014-22166.

The accurate evaluation of fuel and cladding peak temperatures is of prime importance for nuclear reactor design and safety. The Global Threat Reduction Initiative reactor conversion program often encounters exotic flow geometries in its mission to aid in converting reactors from high-enriched to low-enriched fuel. These geometries can pose modeling challenges. Analysis presented here concerns a reactor with twisted fuel pins that are in direct contact with each other in a large, hexagonal-pitch lattice. The Reynolds number for a unit cell is only 7500. Such flow conditions can present difficulties for standard approaches based on Reynolds-Averaged Navier-Stokes (RANS). Moreover there are no available experimental data and a small expected margin to the limiting cladding surface temperature. Given some of the geometric uncertainties, reducing the turbulence model uncertainty is thus important for meaningful calculations. A computational fluid dynamics model of a full-length unit cell was built using the commercial code STAR-CCM+. Multiple RANS models were employed, which gave disparate results. To provide higher-fidelity data for comparison, given the lack of experimental data, a periodic single-helical-pitch simulation with a Large Eddy Simulation (LES) approach was performed using Nek5000, a massively-parallel spectral-element code. This was compared with single-pitch RANS simulations from STAR-CCM+. Stream-wise velocity profile shape was generally well-represented by RANS. Cross-velocities and peak turbulent kinetic energy (TKE) were underestimated for most of the turbulence models with respect to LES, while mean flow TKE was universally underestimated. The overall results suggest that the Realizable k-ε Two-Layer model, which was the best at reproducing the LES TKE distribution, would likely be the most appropriate turbulence model choice for this flow. Future work includes full conjugate heat transfer simulations of 1/6 sectors of fuel assemblies featuring this type of pin lattice.

Commentary by Dr. Valentin Fuster
2014;():V01AT03A026. doi:10.1115/FEDSM2014-22199.

In this study, a computational fluid dynamics (CFD) model is used to numerically characterize the heat transfer from an I-beam support structure of an aluminum reduction pot, during the free convection cooling process. A slice of the I-beam structure is modeled on two different finite element commercial platforms, ANSYS (FLUENT) and StarCCM+, in a suitable domain of air. The K-epsilon Reynolds averaging technique is used to model the turbulence in both platforms. Validation of the modeling technique and parameters adapted is appropriately performed. The structure is segmented and space mean Nusselt numbers (Nu) characterizing the flow are calculated for each section, for Rayleigh number (Ra) ranges typically experienced by the respective section. Expressions correlating the free convection flow over this structure are deduced based on a regression analysis. To conclude, an application of the deduced correlation in modeling the free convection cooling of an aluminum reduction pot is presented.

Commentary by Dr. Valentin Fuster

5th Symposium on Bio-Inspired Fluid Mechanics

2014;():V01AT04A001. doi:10.1115/FEDSM2014-21159.

Since the introduction of the Edmonton Protocol in 2000, islet transplantation has been emerging as promising therapy for Type I diabetes mellitus (T1DM) and currently is the only therapy that can achieve glycemic control without the need for exogenous insulin. Transplanting islet cells has several advantages over transplanting a whole pancreas in that it involves only a minor surgical procedure with low morbidity and mortality, and at a significantly lower cost. However, an obstacle to realizing this goal is a lack of an islet potency index as required by the U.S. Food and Drug Administration (FDA) biologics licensing, as well as a more complete understanding of the physiological mechanisms governing islet and β-cell physiology. Recently, the University of Illinois at Chicago (UIC) has developed a microfluidic platform that can mimic in vivo islet microenvironments through precise and dynamic control of perifusing culture media and oxygen culture levels; all while measuring functionally relevant factors including intracellular calcium levels, mitochondrial potentials, and insulin secretion. By developing an understanding of the physiology and pathophysiology of islets we can more effectively develop strategies that reduce metabolic stress and promote optimization in order to achieve improved success of islet transplantation and open new clinical avenues.

The presentation begins by introducing key issues in the field of pancreatic islet transplantation as a clinical therapy for T1DM. This is followed by brief review various technologies that have been developed to study islet cells. The presentation then describes the design, application, and evolution of UIC’s microfluidic-based multimodal islet perifusion and live-cell imaging system for the study of pancreatic islet and β-cell physiology. The article then concludes presenting initial findings from studies seeking to develop an islet potency test.

Commentary by Dr. Valentin Fuster
2014;():V01AT04A002. doi:10.1115/FEDSM2014-21303.

Micro-air-vehicles (MAVs) and micro-flight robots that mimic the flight mechanisms of insects have attracted significant attention. From this reason, the flight mechanism of the butterflies and their flow fields also has attracted attention. A number of studies on the mechanism of butterfly flight have been carried out. Moreover, a number of recent studies have examined the flow field around insect wings. The present authors conducted a particle image velocimetry (PIV) measurement around the flapping wings of Cynthia cardui and Idea leuconoe and investigated the vortex structure and dynamic behavior produced. However, these results are for a flow field under a fixed condition. The vortex flow structure and the dynamic behavior generated by the wings of a butterfly in free flight are expected to be important for generating the aerodynamic forces required for flight. In the present study, we attempt to clarify the three-dimensional vortex structure around a butterfly in free flight by a scanning PIV measurement. The vortex ring formed by the front wings during the flapping downward grows without attenuation toward the wake. Moreover, during the flapping upward of the wings, a vortex rolls up from the wing, eventually forming a single vortex ring. This vortex ring forms in the vertical direction in contrast to vortex ring formed during the flapping downward, and we may anticipate that the two vortex rings interfere with each other as they advance toward the wake.

Topics: Vortices , Flight
Commentary by Dr. Valentin Fuster
2014;():V01AT04A003. doi:10.1115/FEDSM2014-21465.

More than 90% of the thrust generated by thunniform swimmers is known to be produced by the oscillation of their caudal fin, and the rest by their caudal peduncle. We have designed an experiment in which we can mimic, in a simplified manner, the kinematics of swimmers that mainly use their caudal fin for propulsion. The set-up consists of a rectangular foil attached to a shaft that is controlled by a stepper motor, and the whole assembly can be towed in still water at different controllable speeds. With this system we can study the effect of different types of pitching on the hydrodynamic loads and the performance of the propulsion system. By changing the type of foil, the effects of the flexibility in the propulsion can also be analysed. Hydrodynamic loads were measured with a 6-axes balance, and the flow structures were investigated using a Digital Particle Image Velocimetry (DPIV). Loads and DPIV velocity fields were acquired synchronously.

Commentary by Dr. Valentin Fuster
2014;():V01AT04A004. doi:10.1115/FEDSM2014-21636.

The Large Eddy Simulation (LES) technique was used to study the turbulent airflow field in a realistic model of human upper airways. The geometric model includes nasal cavity, pharynx, larynx and trachea. The Lagrangian approach was used to calculate the trajectories and deposition of micro-particles for the breathing rate of 60 l/min. The results are compared with those obtained from the RANS model from an earlier study. For the latter model the effect of airflow turbulent velocity fluctuations on particle trajectories was modeled using a Continuum Random Walk (CRW) stochastic model.

A qualitative comparison of the results obtained by the LES method with the earlier RANS model reveals that the total depositions of micro-particles evaluated be these two methods are similar. The LES and RANS predictions for regional depositions, however, differ significantly.

Commentary by Dr. Valentin Fuster
2014;():V01AT04A005. doi:10.1115/FEDSM2014-22077.

Underwater fish of the class Batoidea, commonly known as rays and skates, use large cartilaginous wings to propel themselves through the water. This motion is of great interest in bioinspired robotics as an alternative propulsion mechanism. Prior research has focused primarily on the oscillating kinematics used by some species which resembles flapping; this study investigates undulatory motion induced by propagating sinusoidal waves along the fin. An analytical model of undulatory kinematics is presented and correlated with biological literature, and the model is then simulated via unsteady computational fluid dynamics and multiparticle collision dynamics. A bioinspired robot, Batoid Underwater Robotics Testbed (BURT), was developed to test the kinematics of the undulating propulsion system proposed. Finally, BURT was utilized as a platform to investigate engineering challenges in undulating Batoid robotics.

Topics: Propulsion , Robotics
Commentary by Dr. Valentin Fuster
2014;():V01AT04A006. doi:10.1115/FEDSM2014-22102.

A current project is underway to create a prototype of an anatomically correct seagull with biologically accurate flight kinematics. The presented work is focused on the computational fluid dynamics (CFD) analysis of bird flight kinematics. A finite volume approach, using Fluent, was used to attempt to model the kinematics of bird flight with varying degrees of freedom to analyze the lift, drag, pressure, and vortices magnitude associated with a range of flight kinematics. Dimensional analysis has been performed to analyze the effects of angle of incidence on the different sections of a seagull wing. Validated CFD analysis has been performed to identify optimal degree of freedom for generating maximum amount of lift while minimizing drag.

The analysis benefitted from dynamic meshing and a user defined function to model the seagull wing, profiles of which were approximated by the S1223 airfoil. The user defined function allowed for variation of degrees of freedom to model the flight in the current bird prototype and to assess the effects of changing angles of incidence and inlet velocity on lift and drag. Difficulties were encountered when trying to accurately analyze unsteady aerodynamics over a flapping motion. The appropriate grid resolution, the user defined function, as well as the appropriate grid and dynamic mesh parameters within Fluent were all possible areas of concern. The grid resolution was determined by analyzing a steady state case and determining the variation in lift and drag values calculated by increasing the grid density. A user defined function was created that accurately represents the kinematics associated with the bird wing. A triangular grid was utilized for the dynamic mesh with re-meshing procedure activated at every iteration during the analysis. The final geometry provided an accurate method for dynamic re-meshing and overcame the problem of negative cell volume associated with re-meshing using a rectangular mesh configuration. It was determined that maximum cell volume, number of time steps, and time step interval were all important criteria when determining parameters for the unsteady flight analysis.

Results indicate that the unsteady dynamics of bird flapping motion can be effectively represented with modified CFD analysis with updated finite volume scheme. Data indicates that values associated with varying angles of attack at a steady state cannot be used to model flapping flight. The paper will report on further validation to analyze the pressure, lift and drag associated with flapping flight in a three-dimensional study.

Topics: Unsteady flow , Flight
Commentary by Dr. Valentin Fuster
2014;():V01AT04A007. doi:10.1115/FEDSM2014-22118.

A methodology to capture and post-process bat flight 3-D Stereo Triangulation data to formulate an approximated rigid body kinematic model was investigated. Bat flight is unique in nature due to the bats inherent agility and many degrees of freedom when compared to other flying animals. This complexity makes capturing accurate aerodynamic data very difficult. Unlike insects, which utilize few degrees of freedom and a high flap frequency for sustained flight and maneuverability, the agility of bats comes in part from the many degrees of freedom present in the bat wing. In order to better understand the aerodynamics present in bat flight, bats Hipposideridae (Old World leaf-nosed bats) were examined. The trajectories of critical points along the bat wings were recorded using 3D stereo triangulation techniques to capture the complexities of the bat flight. Markers were placed at all the joint locations along the bat wing. The resulting trajectories were then translated into a periodic kinematic model for future computational use.

Commentary by Dr. Valentin Fuster
2014;():V01AT04A008. doi:10.1115/FEDSM2014-22240.

Three dimensional blood flow in a truncated vascular system is investigated numerically using a commercially available finite element analysis and simulation software. The vascular system considered in this study has three levels of symmetric bifurcation. Geometric parameters for daughter vessels, such as their diameters and their angles of bifurcation, are specified according to Murray’s law based on the principle of minimum work. The ratio of blood vessel length to diameter is based upon experimental data found in the literature. An experimentally obtained velocity profile, available in the literature, is used as the inlet boundary condition. An outflow boundary model, consisting of a contraction tube to represent the pressure drop of the small arteries, arterioles, and capillaries that would follow the truncated vascular system, is used to specify the boundary condition at the eight outlets. The results show that although the blood flow velocity experiences a sudden decrease after the bifurcation points due to the higher total cross-sectional area of the daughter vessels as compared to the parent vessel, this decrease in velocity is partially recovered due to the tapering of the blood vessels as they approach the next bifurcation point. The results also show that the secondary flow which is typical after the bifurcation of large arteries does not develop after the bifurcation of small arteries due to the presence of laminar blood flow with very low Reynolds number in the small arteries. The numerical model yields pressure distributions and pressure drops along the vascular system that agree quite well with the physiological data found in the literature. Finally, the results show that, immediately following a bifurcation, the blood flow velocity profile is not symmetrical about the longitudinal axes of blood vessel. However, symmetry is recovered as the blood flow proceeds down the vessel.

Commentary by Dr. Valentin Fuster

Droplet-Surface Interactions

2014;():V01AT05A001. doi:10.1115/FEDSM2014-21224.

Solid rocket motors (SRM)s commonly use aluminized composite propellants. The combustion of aluminum composite propellants in SRM chambers lead to high temperature and pressure conditions resulting in the liquid alumina as a combustion product. The presence of liquid alumina in the flow presents problems such as; chemical erosion of propellant, and mechanical erosion of nozzle. One method of solving the problem of liquid alumina in flow is to change the SRM geometry to induce liquid breakup and suspend the alumina in the flow thus avoiding erosive behavior. To validate numerical simulation methods for geometric breakup induction simulation models of alumina flow can be compared to air and liquid water flows, and the air-liquid water flow models then compared to water-air experimental results. This study investigates experimental geometric induced liquid breakup behavior for the implementation of the alumina flow and nozzle geometry simulation in SRM design. A rectangular chamber was considered for experimental and simulation to explore the air-water flow behavior. The suspension of water was induced with a triangular shaped jump. The resulting two phase flow was examined using photography technique. Significant incitement in the air-water behavior was observed due to geometry modification. Replication of experimental results was simulated with some accuracy.

Topics: Fuels , Rockets
Commentary by Dr. Valentin Fuster
2014;():V01AT05A002. doi:10.1115/FEDSM2014-21256.

This paper presents a method of characterizing liquid breakup phenomena using a probability distribution flow pattern and compares CFD results of a straight channel two-phase flow using k-ε, SST k-ω and Reynolds Stress Model (RSM). Examination of liquid breakup level is essential for solving erosion phenomenon of solid fuel rocket motor (SRM), due to their use aluminum based solid propellants. During the propellant combustion, the aluminum oxidizes into alumina (Al2O3), which tends to agglomerate into molten droplets under a certain flow conditions. The molten droplets can then impinge on the combustion chamber walls, and flow along the nozzle wall. Such agglomerated aluminum leads to erosive damages to the geometry of de Laval nozzle and reduces the SRM propulsion performance. The volume fraction (VF) contour of the liquid can be used as the raw data for time average flow VF contour of straight channel. The flow shows the probability distribution of two-phase boundary which is mostly controlled by the features of different turbulence models. Those results will be used for future comparison to two-phase flow experiment as model selection reference of SRM two-phase supersonic flow simulation.

Commentary by Dr. Valentin Fuster
2014;():V01AT05A003. doi:10.1115/FEDSM2014-21329.

Experimental study is performed to analyze the shear driven droplet shedding on cold substrates with different shear flow speeds typical of those in the flight conditions. Understanding the mechanism of simultaneous droplet shedding, coalescence and solidification is crucial to devise solutions for mitigating aircraft in-flight icing. To mimic this scenario experimental set up is designed to generate shear flow as high as 90 m/s. The droplet shedding at high speed is investigated on a cold surface (0 and −5 °C) of different wettabilities ranging from hydrophilic to superhydrophobic. Result analyses indicate that on a hydrophilic substrate, the droplets form a rivulet which then freezes on the cold plate. In contrast, on the superhydrophobic surface, there is no rivulet formation. Instead, droplets roll over the substrate and detach from it under the effect of high shear flow.

Topics: Drops , Solidification
Commentary by Dr. Valentin Fuster
2014;():V01AT05A004. doi:10.1115/FEDSM2014-21371.

Nanodrop impact onto a solid substrate is of interest for nano-scale liquid-impingement, phase-change cooling and for material deposition processes. In the present study, classical molecular dynamics (MD) simulation techniques were implemented to study the thermo-mechanical properties of the impact of nanometer scale liquid droplets upon an atomistic substrate at a temperature higher than that of the droplet. The droplets were comprised of approximately 50,000 Lennard-Jones atoms arranged in tetramer finitely extensible non-linear elastic (FENE) chain molecules forming a sphere of 8 nm radius. They were equilibrated and then projected towards a wall, where we observed the response upon collision by changes in shape, temperature, and density gradients, across a variety of impingement velocities, substrate temperatures, and wetting conditions. The baseline cases of equal substrate and nano-drop temperature were validated by comparison with previously reported results. A reaction spectrum ranging from full thermal vaporization of the drop, with respective substrate cooling, to complete kinetic disintegration upon impact and surface heating are analyzed. The variation of thermal and kinetic effects across the parametric environment is used to identify those regimes that optimize heat transfer from the surface, as well as those that best facilitate material deposition processes.

Commentary by Dr. Valentin Fuster
2014;():V01AT05A005. doi:10.1115/FEDSM2014-21521.

Separating oil from solid particles is of great importance in many industrial processes including the extraction of bitumen from oil sands, and the remediation of oil spills. The usual approach is to separate the oil from the solid by introducing another liquid (e.g. water). Separation is often assisted by fluid mixing, and chemical addition. Yet while oil-water-particle separation has been well studied from a chemical standpoint, little research has taken into account the effect of hydrodynamics on separation. In this work, the separation of oil from a single oil-coated spherical particle falling through an aqueous solution was evaluated as a function of viscosity ratio. Solvents were used to modify the viscosity of the oil. The experiments were recorded using a high-speed camera and post-processed using the MATLAB image-processing toolbox. A CFD model has also been developed to study this phenomenon.

The results indicate that when viscous forces are strong enough, the oil film deforms, flows to the back of the sphere, and forms a tail that eventually breaks up into a series of droplets due to a capillary wave instability. When the viscosity ratio is small (i.e. the oil is less viscous than the solution), a thin tail forms quickly, the growth rate of the instability is high, and hence the tail breaks very quickly into smaller droplets. When the viscosity ratio is high (i.e. the oil is more viscous), more time is required for the deformation/separation to initiate, and the tail is thicker and breaks more slowly into larger droplets. It was observed that when the viscosity ratio is close to 1, the rate of separation is increased and maximum separation is achieved.

Commentary by Dr. Valentin Fuster
2014;():V01AT05A006. doi:10.1115/FEDSM2014-21642.

The oil produced from offshore reservoirs normally contains considerable amount of water. The separation of water from oil is very crucial in petroleum industry. Studying the coalescence of two droplets or one droplet and interface can lead to better understanding of oil-water separation process. In this study, the coalescence of two droplets and droplet-interface are simulated using a commercial Computational Fluid Dynamics (CFD) code FLUENT 14. In order to track the interface of two fluids, two approaches, Volume of Fluid (VOF) and Level-Set method were utilized. The results are compared with experimental measurements in literature and good agreement was observed. The effect of different parameters such as droplet velocities, interfacial tension, viscosity of the continuous phase and off-center collision on the coalescence time has been investigated. The results revealed that coalescence time decreases as the droplet velocities increase. Also, continuous phase with higher viscosities and lower water-oil interfacial tension, increase the coalescence time.

Topics: Drops
Commentary by Dr. Valentin Fuster
2014;():V01AT05A007. doi:10.1115/FEDSM2014-21648.

Several analytical models exist to predict droplet impact behavior on superhydrophobic surfaces. However, no previous model has rigorously considered the effect of surface slip on droplet spreading and recoiling that is inherent in many superhydrophobic surfaces. This paper presents an analytical model that takes into account surface slip at the solid-fluid interface during droplet deformation. The effects of slip are captured in terms that model the kinetic energy and viscous dissipation and are compared to a classical energy conservation model given by Attane et al. and experimental data from Pearson et al. A range of slip lengths, Weber numbers, Ohnesorge numbers, and contact angles are investigated to characterize the effects of slip over the entire range of realizable conditions. We find that surface slip does not influence normalized maximum spread diameter for low We but can cause a significant increase for We > 100. Surface slip affects dynamical parameters more profoundly for low Oh numbers (0.002–0.01). Normalized residence time and rebound velocity increase as slip increases for the same range of We and Oh. The influence of slip is more significantly manifested on normalized rebound velocity than normalized maximum spread diameter. Contact angles in the range of 150°–180° do not affect impact dynamics significantly.

Topics: Drops
Commentary by Dr. Valentin Fuster
2014;():V01AT05A008. doi:10.1115/FEDSM2014-21650.

Ice accretion is a major threat to all exposed structures such as wind turbines, overhead power cables, offshore structures and aircrafts. Such deposition starts by an impact of water droplets of different sizes on the surface of the exposed structure. This work aims to shed more light on the difference in the dynamics occurring upon the impact of microdroplets on substrates with various wettabilities, hydrophilic (aluminum) and Superhydrophobic (Aluminum + WX2100) surfaces. Experiments are conducted on a wide range of diameters, between cloud sized droplets with diameters ranging down to 20μm, and 10 times larger droplets with a diameter of 250 μm. A comparison in the impact (through deformation) results is made all through the wide range and explained using the two extremes. This is done experimentally by analyzing the maximum spread diameter on the hydrophillic surface and superhydrophobic surface and maximum height as a function of time on the hydrophillic surface. Both parameters are visualized experimentally, simulated numerically for the same impact velocities and then results are compared for verification.

Topics: Drops
Commentary by Dr. Valentin Fuster
2014;():V01AT05A009. doi:10.1115/FEDSM2014-21681.

Multiphase Smoothed Particle Hydrodynamics (SPH) method has been used to study the jet breakup phenomena. It has been shown that this method is well capable of capturing different jet breakup characteristics. The value obtained for critical Weber number here in transition from dripping to jetting is a very good match to available values in literature. Jet breakup lengths are also agreeing well with several empirical correlations. Successful usage of SPH, as a comparably fast CFD solver, in jet breakup analysis helps in speeding up the numerical study of this phenomenon.

Commentary by Dr. Valentin Fuster
2014;():V01AT05A010. doi:10.1115/FEDSM2014-21760.

Droplet impact on solid surfaces has been extensively reported in the literature, however the effect of accompanying air flow on the outcome of impacting droplet has yet to be addressed and analyzed which is similar to real scenario of impacting water droplet on aircraft’s leading edge at in-flight icing conditions. This study addresses the net effect of airflow (i.e. stagnation and the resultant shear flow) on the impacting water droplet with the same droplet impact velocity which is exposed to different airspeeds. In order to provide stagnation flow, a droplet accelerator was built which can generate different airspeeds up to 20 m/s. Droplet impact behavior accompanied with stagnation flow on a polished aluminum surface with a contact angle of 70° was investigated by high speed photography. 2.5 mm water droplet size with impact velocities of 2, 2.5 and 3 m/s which correspond to non-splashing regime of impacts are exposed to three different regimes of air speeds namely 0 (i.e. still air case), 10, and 20 m/s. It was observed that when droplet reaches to its maximum spreading diameter, some fingered shape at the end of spreading lamella (i.e. Rayleigh-Taylor instability) is appeared. When stagnation flow is present these fingered shape droplets are exposed to the generated shear flow close to the substrate (i.e. Homann flow approach) causes a droplet break up while complete non-splashing regime is observed in still air case. In spite of the fact that maximum spreading diameter is not largely affected by air flow compare to still air case, droplet height variation is significantly reduced by about 70 percent for strong stagnation flow (i.e. 20 m/s) which generates non-recoiling condition resulting in the thin film formation.

Commentary by Dr. Valentin Fuster
2014;():V01AT05A011. doi:10.1115/FEDSM2014-21763.

A compressed air sprayer was used to spray model paint onto two glass substrates at the same time. Afterwards, one glass substrate was placed on a LED light source and still photographs were taken from the top using a DSLR camera with a timer system. The other substrate was put on a balance to record weight. Pictures and weight measurements were taken at 5 second intervals for 15 minutes. The sprayed film thickness was varied. The pictures were analyzed using ImageJ software. Bubble Count vs. Time, Sauter Mean Diameter (SMD) of Bubbles vs. Time as well as Weight vs. Time was plotted. It was seen that the pace of weight loss was faster for thinner films. The rate of bubble escape also depended on film thickness. It took a longer time for thicker films to lose the bubbles entrapped in them. In the first 30 seconds, large bubbles escaped due to buoyancy forces and afterwards surface-tension driven flows became dominant. There was also a lot of bubble movement in thicker films. The effect of gravity was studied as well. Gravity did not affect the bubble escape rate since a downward facing film had the same bubble count as an upward facing film confirming that bubble motion was not due to buoyancy forces alone. However, the SMD of bubbles in a downward facing film was larger than an upward facing film. Buoyancy is not a factor in bubble escape from the downward facing film and only surface-tension driven flows play a role.

Topics: Bubbles
Commentary by Dr. Valentin Fuster
2014;():V01AT05A012. doi:10.1115/FEDSM2014-22108.

The dynamics of viscous droplets near solid surfaces, especially micro-textured surfaces, and the interaction between them are of great importance in industrial applications, biochemical processes, and fundamental materials research on surface wettability. In this work, a three-dimensional spectral boundary element method has been employed to investigate the dynamics of a viscous droplet falling under gravity influence to micro-textured solid surfaces. The droplet size, in this study, is comparable to the size of the surface texture. The influences of the Bond number, relative size of the droplet with respect to the surface features, and the topological characteristics of the substrate on the droplet motion and deformation are investigated. The stress exerted on the substrate due to droplet motion is also explored.

Topics: Drops
Commentary by Dr. Valentin Fuster

6th Symposium on CFD Verification and Validation

2014;():V01AT07A001. doi:10.1115/FEDSM2014-21056.

Analysis and interpretation of daily emission ventilation with fire ventilation systems in indoor parking lots coupling with jet fans have been done by a CFD program. The ventilation of an eight-story parking lot in Istanbul is carried out and the investigation for a simplified one-story of this system is also considered. The placement of jet fans has been identified on the basis of eliminating dead jones in the passage with the help of air flow analysis. Therefore, the present study may be of great importance as both the fire and smoke can be evacuated effectively by using the optimal position of jet fans in the parking lots to maintain a less polluted atmosphere inside.

Commentary by Dr. Valentin Fuster
2014;():V01AT07A002. doi:10.1115/FEDSM2014-21069.

In this paper, the 3D numerical simulations on falling film behaviour on flat plate with and without interfacial gas-liquid shear stress are carried out. The film thickness and velocity distribution of water film flow with different Reynolds numbers are studied. The results agree well with the experimental and theoretical data. The influence of the surface wave on film velocity is revealed. The calculations also investigate the effect of gas-liquid shear stress on solitary waves of falling film.

Topics: Fluids , Flat plates
Commentary by Dr. Valentin Fuster
2014;():V01AT07A003. doi:10.1115/FEDSM2014-21478.

Results of experimental study of mixing of liquids with different temperatures in a T-junction are presented. Experiments were performed with liquids which have significantly different physical properties. The liquid with higher temperature was injected through the branch of the T-junction. The test section was made of thin wall stainless tubes. The distribution of wall temperature over the surface was measured using a high speed infrared camera. Both time averaged and fluctuational characteristics of the temperature field were obtained from infrared image processing.

The structure of temperature field inside the channel was measured by a microthermocouple. It was mounted on the traversing unit which allowed its translation along the channel diameter. Measurements were made over different diameters in the same cross section. This allowed to construct the three-dimensional structure of the temperature field.

Results obtained provide experimental data necessary for validation of thermo hydraulic codes for the design of power equipment.

Topics: Junctions
Commentary by Dr. Valentin Fuster
2014;():V01AT07A004. doi:10.1115/FEDSM2014-21482.

Experimental study of liquid flow and heat transfer in an annular channel is performed. The channel consisted of two coaxial tubes with the diameters of 42.2 and 20 mm. An obstacle covering a quarter of the channel section was placed in the channel to produce a strong three-dimensional disturbance of the flow. Measurements of local wall shear stress are performed using an electrochemical technique. Measurements of heat transfer, time-averaged and fluctuational wall shear stress are performed at various points relative to the obstacle, this allowed to study the field of the hydrodynamic parameters of the flow.

The experimental data obtained can be used for validation of existing and developing computer codes accounting for a 3-D structure of turbulent flows.

Commentary by Dr. Valentin Fuster
2014;():V01AT07A005. doi:10.1115/FEDSM2014-21600.

A coaxial piping system (CPS) that involves a transition from a smaller annulus into a larger annulus is investigated to evaluate the generation of vortices and recirculation zones around the transition area. These areas are of interest for industrial applications where erosion within the piping system is a concern. The focus of this work is to evaluate the capabilities of Computational Fluid Dynamics (CFD) using commercial Reynolds-Averaged Navier Stokes (RANS) models to predict the regions and intensity of vortices and recirculation zones. A trusted grid is developed and used to compare turbulence models. The commercial CFD solver Fluent (Ansys Inc., USA) is used to solve the flow governing equations for different CFD numerical formulations, namely the one equation Spalart-Allmaras model, and steady-state RANS with different turbulence models (standard k-epsilon, k-epsilon realizable, k-epsilon RNG, standard k-omega, k-omega SST, and transition SST) [1].

CFD results are compared to time-averaged particle image velocimetry (PIV) measurements. The PIV provides 3D flow field measurements in the outer annulus of the piping system. Velocities in regions of interest were used to compare each model to the PIV results. An RMS comparison of the numerical results to the measured values is used as a quantitative evaluation of each turbulence model being considered. The results provide a useable CFD model for evaluation of the flow field of this flow field and highlights areas of uncertainty in the CFD results.

Commentary by Dr. Valentin Fuster
2014;():V01AT07A006. doi:10.1115/FEDSM2014-21688.

Computational Fluid Dynamics (CFD) modeling software is increasingly being used as the tool of choice for analyzing the flow details and integrated performance of turbo-machinery products. In fact, the use of CFD is rapidly transitioning from a verification tool to an upfront design-enabling & optimization tool. Experimental validation of computational simulation is essential to ensure an acceptable degree of reliability and relevance of the simulated results to real world performance. While CFD has been rigorously validated for numerous simple physics cases like single-phase flow, more complex physics applications, e.g., those involving multi-phase solid-liquid flows, require more elaborate and thoughtful means of validation. In this context, a study was undertaken to review Particle Image Velocimetry (PIV) as a means of validating more complex CFD cases and to contrast the findings with those obtained from CFD simulation. PIV offers a new possibility for flow visualization in turbo-machinery passages, in contrast to traditional methods like flow probing or hot-wire anemometry, which can be a very challenging proposition in the rotating domain of a turbo-machinery blade system. This paper discusses the first phase of this work, which was limited to single-phase flow studies, with the intent to follow up further with multi-phase flow studies. A specially designed fractional horsepower centrifugal pump is used as a test subject to analyze all possible parameters of the flow field using PIV and the result is then compared with the CFD simulations of the same model. The results show a reasonable match in the flow patterns obtained by the two alternate methods, although significant differences are apparent too. In conclusion, each method has its own place in the context of turbo-machinery flow studies.

Commentary by Dr. Valentin Fuster
2014;():V01AT07A007. doi:10.1115/FEDSM2014-21695.

FLASH is a massively parallel, multi-physics, open source code developed by the University of Chicago [1] for investigating astrophysical phenomena. FLASH was modified [2] to handle detailed chemical kinetics for hydrogen and methane combustion and heat release, in order to enable the code to be used for combustion applications. These capabilities have been tested and validated [2, 3] through an extensive suite of simulations. These modifications include the addition of detailed H2-air and CH4-air chemistry along with temperature dependent thermodynamic and transport properties.

The aim of this work is to apply the modified version of FLASH to three cases of highly compressible supersonic flows, involving chemical reactions. The first problem is a reacting shock-bubble interaction, in which the shock triggers combustion in a fuel bubble. In the second problem, a two-dimensional, Richtmyer-Meshkov Instability leading to combustion of the fuel at a non-premixed, single mode perturbed interface has been studied numerically. In these cases, the relationship between the integral heat release rate, the integral H2O production rate and total circulation is investigated. In the third example, a multimode perturbed interface has been implemented into a 3D simulation. Time evolution of the interface undergoing reacting RMI is studied. By defining a reflecting endwall for RMI simulations, the effect of a second reflected shock on the mixing behavior and combustion on an already shocked interface has been studied.

Numerical simulations of the interaction of a shock (Mach number 2) in air with H2 bubble was performed [2]. The misalignment between the pressure gradient across the shock front and the density gradient at the site of H2-Air interface generates baroclinc vorticity. This phenomenon generates counter-rotating vortices that breakdown the H2 bubble (fuel). The rapid breakdown of the H2 bubble transitions to turbulent mixing, intensifying the heat release rate. In two more general configurations, chemically reacting, single-mode and multimode Richtmyer-Meshkov Instability has been studied. RMI is the driving mechanism for growth of small interfacial perturbations. Initially single-mode perturbations of small amplitude grow linearly due to impulsive acceleration by shock. This is followed by non-linear growth at late times due to the formation of secondary Kelvin-Helmholtz instability. Heat release and product formation in the vicinity of interface will affect perturbation growth rates which will affect the mixing behavior and therefore the combustion efficiency [4].

Commentary by Dr. Valentin Fuster

Symposium on Development and Applications of Immersed Boundary Methods

2014;():V01AT08A001. doi:10.1115/FEDSM2014-21284.

In this study, we intend to develop a high-order numerical approach using the immersed-boundary method to solve problems with complex and moving boundaries such as biological locomotion (animal swimming and flying) and biomedical flow systems. The basic idea is to use the compact finite-difference scheme to resolve the flow field and a higher-order forcing scheme to treat the immersed boundaries. In this work, the one-dimensional formulation of the numerical approach is presented and is tested extensively for its convergence. Such tests are necessary before more complicated tests in 2D/3D. An overall third-order accuracy is achieved as desired. Extension to higher dimensions is ongoing.

Commentary by Dr. Valentin Fuster
2014;():V01AT08A002. doi:10.1115/FEDSM2014-22085.

The immersed boundary methods are well known as an efficient flow solver for engineering problems involving fluid structure interactions. However, in order to obtain better results, higher resolutions near the immersed boundary points are desired. Non-uniform Cartesian mesh can easily fulfill this task without introducing a dramatic increase on the cost of computation and coding. In the current paper, an immersed boundary method with non-uniform Cartesian mesh is demonstrated. The Poisson problem is solved with assistance of a scientific parallel computational library PETSc. The code is validated with a three-dimensional flow over a stationary sphere. Then, a fluid-structure interaction model is coupled and validated with two-dimensional vortex induced vibration problems. Comparisons with previous studies are presented. The ultimate goal is to couple the fluid-structure interaction model with the three-dimensional immersed boundary method.

Commentary by Dr. Valentin Fuster

9th Symposium on DNS, LES, and Hybrid RANS/LES Methods

2014;():V01AT09A001. doi:10.1115/FEDSM2014-21178.

Recent years have seen increased emphasis on mathematical model reduction using modal decomposition techniques of high dimensional flow field data from experiments as well as numerical simulations. These tools decode the complex unsteady flow-field into several modes. Different tools highlight different flow dynamics. In the experimental community, Proper Orthogonal Decomposition (POD) has been the most commonly used technique, ranking modes by their relative energy content, without concern for temporal aspects. However, many dynamics are not highlighted by the most energetic structures. In transitional flows for example, structure growth is a more a more important indicator of the turbulent effects. The Dynamic Mode Decomposition (DMD) technique highlighted by Schmid [1] achieves this by ranking modes by the most dynamically varying flow features. In this work, we use DMD and POD to analyze flow past a NACA0015 airfoil at Reynolds number of 100,000 and AoA=15 degree, without and with control. The specific control technique employed is based on the Nano-second Pulsed Dielectric Barrier Discharge (NS-DBD) actuator. Experimentally validated high fidelity 3-D numerical simulations are employed to generate the required snapshots. From the DMD modes, the dominant time-varying flow structures associated with the two cases are identified, and their stability characteristics are compared. DMD and POD modes are compared to each other. The DMD modes highlight the dynamically varying nature of the flow-field. A Floquet stability analysis of the eigenvalues from DMD for both the no-control and control cases is presented. Further, the original flow field is reconstructed from the DMD modes and their individual modal behavior has been analyzed to show the effect of control authority on the flow.

Topics: Airfoils
Commentary by Dr. Valentin Fuster
2014;():V01AT09A002. doi:10.1115/FEDSM2014-21185.

The capability of Large Eddy Simulations (LES) to accurately model Nano-Second Pulsed Dielectric-Barrier Discharge (NS-DBD) plasma actuators for use as a flow control devise is demonstrated by comparing the newly-developed volumetric heating model to experimental results as well as a previously established surface heating model. The purpose of these models and corresponding experiments is to show that use of NS-DBD actuators can mitigate the presence of stall on a NACA0015 airfoil at a Reynolds number of 100,000 and 15° angle of attack in reversed-flow conditions. Actuators are placed at both the aerodynamic leading and trailing edge — the effects of which are analyzed separately — and forced at several Strouhal numbers Display FormulaStF=fcU. The model validation is carried out by comparing the actuator pulse structure, mean value contours of various parameters, static pressure distribution (Cp) along the airfoil surface, and FFT plots of sound pressure level (SPL). The model results are then compared to the no-control simulations to provide evidence that actuation delays the onset of stall. This process is explored for both unsteady and steady quantities, including FFT plots, intantaneous flow field response, static pressure recovery, and mean quanitites, including a boundary layer analysis. It is concluded that at low Reynolds numbers reattachment occurs through enhanced turbulence of a separated, laminar shear layer; the reattachment processes is shown to take place over approximately 8 characteristic times for both actuator locations, although leading edge actuation only results in reattachment in the mean sense. Under similar situations, the volumetric and surfaec heating models showed similar recovery characteristics; however, since the volumetric model is less empirical than surface heating, it is recommended that volumetric heating be used in the future. Both heating models indicate that actuation at the aerodynamic leading edge has the greatest affect on the flow due to the laminar nature of the corresponding shear layer, as opposed to the turbulent shear layer on the trailing edge. It addition, a change in duty cycle was shown to have little effect on the results whereas an incerase in StF had a large negative effect on reattachment.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A003. doi:10.1115/FEDSM2014-21186.

Large Eddy Simulations (LES) were performed for Mach 0.9 and 1.3 cold jets to associate the structures of the shear layer with near field pressure fluctuations. The jets were excited by Localized Arc Filament Plasma Actuators (LAFPAs) arranged around the periphery of the nozzle with the axisymmetric (m = 0) mode. Excitation frequencies of St = fD/Uj = 0.05 to 0.25 (close to the column mode frequency) were computed for each Mach number. The St = 0.05 produces one pulse that propagates downstream without interacting with previously emitted pulses. This is referred to as the the impulse response. The St = 0.25 frequency exhibits subsequent pulse interactions. Simulation data for both Mach numbers was collected along three arrays at different radial locations. Strong agreement was found for the near field response to excitation and the mean center-line axial velocity between the subsonic simulations and the experiments. The experiment and simulations depict a large hydrodynamic wave downstream of the exit moving at the speed of convection near the shear layer consisting of a large peak followed by a large trough after the actuator pulse. For the highest excitation frequency, the interaction between structures yields an almost sinusoidal wave in the near field. These hydrodynamic waves are associated to the phase-averaged flow structure which includes a series of rollers and ribs and the associated dilatation field. The structure interactions from subsequent pulses results in a quasi-linear superposition of the impulse jet response (St = 0.05) to actuation. Auto-correlations and two-point correlations describe the development and interaction between adjacent structures in time and space.

Topics: Pressure , Jets
Commentary by Dr. Valentin Fuster
2014;():V01AT09A004. doi:10.1115/FEDSM2014-21332.

Multiphase flow with particles covers a wide spectrum of flow conditions in natural world and industrial applications. The experiments and the direct numerical simulation have become the most popular means to study the dilute particle-laden flow in the last two decades. In the experimental study, the mean Reynolds number is often adjusted to the value of single-phase flow for each set of particle conditions. However, the friction Reynolds number usually keeps invariable in the direct numerical simulation of the particle-laden flows for convenience. In this study the effect of the difference between given mean Reynolds number and friction Reynolds number was investigated. Two simulations were performed for each set of particle parameters, and the mean Reynolds number and friction Reynolds number were kept invariant respectively. From the results it can be found that the turbulence intensity and the dimensionless velocities are larger when keeping the friction Reynolds constant. And the results calculated from the cases of keeping the mean Reynolds number invariable agree with the experiment results better. In addition, the particle distribution along the wall-normal coordinate was found to be unchanged between two simulation conditions. As a suggestion, keeping the same mean Reynolds number in the direct numerical simulation of particle-laden flow is more appropriate.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A005. doi:10.1115/FEDSM2014-21335.

The present paper concerns a numerical benchmark of various turbulence modelings, from RANS to LES, applied to Taylor-Couette-Poiseuille flows in a narrow gap cavity for six different combinations of rotational and axial Reynolds numbers. Two sets of refined Large-Eddy Simulation results, using the WALE and the Dynamic Smagorinsky subgrid scale models available within an in-house code based on high-order compact schemes, hold for reference data. The efficiency of a RANS model, the Elliptic Blending Reynolds Stress Model (EB-RSM) [1], and a hybrid RANS/LES method, the so-called “Equivalent DES” [2], both run with Code Saturne, is then questioned. Thin coherent structures appearing as negative (resp. positive) spiral rolls are obtained by the LES but also the hybrid RANS/LES along the rotor (resp. stator) sides. More quantitatively, the hybrid RANS/LES does not improve the predictions of the EB-RSM for both the mean and turbulent fields, stressing the need for further theoretical development.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A006. doi:10.1115/FEDSM2014-21374.

A fourth-order accurate symmetry-preserving discretization for compressible flow is used to perform simulations of the turbulent flow over a delta wing. A symmetry-preserving discretization eliminates the non-linear convective instability by preserving conservation of kinetic energy at the discrete level. This enhances the stability of a simulation method, so that little artificial dissipation is needed for numerical stability. It is shown that simulations of the flow over a sharp-edge delta wing at Re = 50,000 with the symmetry-preserving discretization are stable without artificial dissipation in a region of interest around the delta wing. To assess the accuracy of the simulation method, results obtained on a fine computational grid are compared with results obtained on a coarser grid. Also results obtained with large-eddy simulation models and with sixth-order artificial dissipation are presented.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A007. doi:10.1115/FEDSM2014-21391.

Equilibrium turbulent flat plate boundary layers with time invariant statistics were obtained at Mach numbers 1.7, 2.3, and 2.9. These are to be used as the initial condition for Large Eddy Simulations (LES) or Direct Numerical Simulations (DNS) of shock wave/turbulent boundary layer interactions utilizing a body force-based method. The results obtained are supplemented by an analysis of the mean and statistical properties of the respective boundary layers. The spanwise extent of the domain required to allow adequate decorrelation between the centerline and the boundaries is investigated by extensively probing the flowfields obtained. This is done to quantify the coherent structures of the turbulent flow. Specifically, two point correlations and integral length scales are used to investigate spanwise decorrelation distances in an attempt to pick a computational domain which is large enough to permit decorrelation downstream but small enough to minimize computational costs. It is shown that by examining the precursor events in the upstream region, namely the generalized stability criterion, it is possible to provide estimates for the force field parameters necessary for transition for a given flow, with only a small portion of the domain in the neighborhood of the trip. The technique is made even more efficient by investigating the possibility of determining these parameters using a two-dimensional simulation. Additionally, the three flow fields obtained are surveyed to confirm that they are suitable for subsequent SBLI simulations. We check that (i)they possess the expected turbulent characteristics and (ii)there is no signature of the tripping mechanism.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A008. doi:10.1115/FEDSM2014-21393.

In an effort to reduce the aircraft jet noise, control of jets has become one of the highly explored areas. In this work, we examine an underexpanded jet subjected to control with Localized Arc Filament Plasma Actuators (LAFPA) to complement prior results on perfectly expanded flow. High fidelity, Large Eddy Simulations (LES) are employed with a simple model for the actuators, eight of which are placed along the periphery of a Mach 1.2 converging nozzle exit. The axisymmetric mode (m=0) is excited at two different Strouhal numbers of 0.3 (corresponding to the most amplified jet-column mode) and 0.9, based on the exit diameter of the nozzle. Baseline (no control) simulations at two different Reynolds numbers (100,000 and 1.2 million) are also performed. Results indicate a good correlation between the numerical and the experimental results. Undulations are observed in the mean flow, which correspond to the increase and decrease of the flow velocity as the jet traverses the complex shock cell structures generated as a result of the under-expansion of the jet. Baseline simulations at the two chosen Reynolds numbers reveal no significant difference between the two cases indicating that the effect of Reynolds number is negligible. Phase-averaged results, for St=0.3, indicate the presence of large vortical structures generated as a result of amplification of the natural structures due to actuation. Two different kinds of structures are generated corresponding to the switching on and switching off of the plasma actuators. These structures are absent when the flow is actuated at St=0.9. Quantitative near field acoustic analysis is conducted using two-point correlation technique. The qualitative effect of forcing on far-field noise propagation is also investigated.

Topics: Jets
Commentary by Dr. Valentin Fuster
2014;():V01AT09A009. doi:10.1115/FEDSM2014-21435.

The fluid dynamics and heat transfer characteristics of a turbulent round jet are modelled numerically using Reynolds-Averaged Navier-Stokes (RANS) and Large Eddy Simulation (LES). Meshes with varying degrees of coarseness, with both radial and axial refinements are investigated. Discretization is carried out using the finite volume method. The jet configurations are chosen to enable validation against well-established experimental jet-impingement heat-transfer studies, particularly that of Cooper et al. [1]. The Reynolds number studied is 23000. The height of discharge from the impingement wall is fixed at twice the jet diameter. The work critically examines the effect of Reynolds number, standoff distance and helps to ascertain the relative merits of various turbulence models, by comparing turbulent statistics and the Nusselt number distributions. The present work is carried out as a preliminary validation, in a wider study intended to determine the thermofluidic behaviour of jets impinging upon an oscillating surface.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A010. doi:10.1115/FEDSM2014-21517.

Large Eddy Simulations (LES) of wall-bounded flows at high Reynolds numbers demand an extremely fine mesh resolution in the wall proximal inner layer. Accurate modeling of near wall turbulence is therefore crucial in reducing the computational cost of LES at practical Reynolds numbers. One approach is the implementation of a two-layer model that solves for a reduced one-dimensional boundary layer equation in the inner wall layer. A wall modeled LES thus allows for a coarser grid resolution than a wall resolved LES. This work evaluated the performance of a wall modeled LES against a wall resolved LES as well as experimental data for the flow over a wall mounted hump at Reynolds number 9.36×105. Results from the wall modeled LES were in good agreement with both wall resolved LES and experimental data in parameters such as surface pressure coefficient, skin friction, mean velocity profiles, Reynolds stresses and flow reattachment. It was observed that the wall modeled LES required only a fifth of the computational resources required for the wall resolved LES.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A011. doi:10.1115/FEDSM2014-21588.

The paper is concerned with the prediction and analysis of dynamic stall of flow past pitching NACA-0012 airfoil at 105 Reynolds number based on the chord length of the airfoil and at reduced frequency of 0.188 in a three dimensional flow field. The turbulence in the flow field is resolved using large eddy simulations with dynamic Smagorinsky model at the sub grid scale. The lift hysteresis plots indicate closer match to experimental results, although discrepancies exist during the downstroke. The development of dynamic stall vortex, vortex shedding and reattachment as predicted by the present study are discussed in detail. This study has shown that the downstroke phase of the pitching motion is strongly three dimensional and is highly complex, whereas the flow is practically two dimensional during the upstroke.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A012. doi:10.1115/FEDSM2014-21719.

Lack of accurate criteria for onset of incipient motion and sediment pickup function remain two of the biggest hurdles in developing better predictive models for sediment transport. To study pickup and transport of sediment, it is necessary to have a detailed knowledge of the small amplitude oscillatory flow over the sediment layer near the sea bed. Fully resolved direct numerical simulations are performed using fictitious domain approach [1] to investigate the effect of a sinusoidally oscillating flow field over a rough wall made of regular hexagonal pack of spherical particles. The flow arrangement is similar to the experimental data of Keiller et al. [2]. Transitional and turbulent flows at Reδ = 95,150,200 and 400 (based on the Stokes layer thickness, Display Formulaδ=2ν/ω) are explored over a range of non-dimensional sphere sizes. Turbulent flow is characterized in terms of coherent vortex structures, Reynolds stress variation, turbulent kinetic energy budget and PDF distributions. The nature of unsteady hydrodynamic lift forces on sediment grains and their correlation to sweep and burst events is also reported. The dynamics of the oscillatory flow over the sediment bed is used to understand the mechanism of sediment pick-up.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A013. doi:10.1115/FEDSM2014-22043.

This work presents Large Eddy Simulations of the flow and heat transfer characteristics in a matrix of surface mounted cubes, studying the effect of the grid resolution and the sub-grid scale modeling. Three sub-grid scale models, implemented in an unstructured, finite volume, commercial solver, are compared on four different grids in terms of first and second order statistical quantities. A classical Dynamic Smagorinsky model is compared with a no-model, Implicit Large Eddy Simulation, approach and a recently developed, two parameters, dynamic-mixed model. A general lack of sensitivity to the sub-grid scale model is evidenced for the flow quantities at all the resolutions, but the grid design emerges as the most determining factor for this kind of flows, showing that accurate results are possible with very coarse resolutions. In contrast, heat transfer characteristics show a strong dependence on both the grid and the sub-grid scale model with a lack of clear convergence in the investigated range of scales. The dynamic-mixed model, which for the first time is tested in a heat-transfer application, is found stable on strongly stretched grids and cheaper than the classical Dynamic Smagorinsky model due to its specific finite volume formulation, showing its suitability for more complex applications.

Commentary by Dr. Valentin Fuster
2014;():V01AT09A014. doi:10.1115/FEDSM2014-22055.

A dynamic hybrid RANS/LES (DHRL) model has been implemented in the spectral-element solver Nek5000 to reduce computational expense for high Reynolds number applications. The model couples a k-ε URANS model and the dynamic Smagorinsky model for LES. The model is validated for plane channel flow at Reτ = 590 using DNS data, and compared with LES predictions. The model is then applied for the ANL-MAX case, which is a test case relevant to nuclear reactor cooling flow simulations. For the channel flow case, DHRL predictions were similar to LES on finer grids, but on coarser grids, the former predicted velocity profiles closer to DNS than the latter in the log-layer region. The improved prediction by the DHRL model was identified to be due to a 30% additional contribution of RANS stresses. For the ANL-MAX case, the URANS simulation predicts quasi-steady flow, with dominant large-scale turbulent structures, whereas LES predicts small-scale turbulent structures comparable with results in rapid mixing of cool and warm flow jets. DHRL simulations predict LES mode in the inlet jet region, and URANS mode elsewhere, as expected.

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