ASME Conference Presenter Attendance Policy and Archival Proceedings

2016;():V01AT00A001. doi:10.1115/FEDSM2016-NS1A.

This online compilation of papers from the ASME 2016 Fluids Engineering Division Summer Meeting (FEDSM2016) 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presenter attendance policy, if a paper is not presented at the Conference by an author of the paper, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.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

17th Symposium on Turbomachinery Flow Simulation and Optimization

2016;():V01AT02A001. doi:10.1115/FEDSM2016-7535.

This paper presents an automated tool chain for simulating Francis turbine behavior during the transient processes induced by a load rejection event. The proposed methodology combines a commercial CFD solver and a user function and scripts to address the simulation challenges caused by the wicket gate motion and runner speed variation during emergency shutdown. Mesh deformation and re-meshing techniques are used to simulate the large displacement of the wicket gates. The runner speed variation is computed using an angular momentum equation implemented in a user defined function. The proposed methodology was developed and validated by performing 2D unsteady simulations on a high head model Francis turbine used in the Francis-99 workshop, followed by a 3D unsteady simulations on a medium head Francis turbine. These simulations allow computing the evolution of engineering quantities such as turbine angular speed, flow physics and unsteady load on blades during the process. The validation of CFD results with experiments showed 9% discrepancy in the prediction of runaway speed. The investigation of flow physics reveals the presence of complex flow structures such as reversed flow (pumping flow) near the draft tube cone center and a downward tangential flow near the cone wall of the draft tube. Pressure fluctuations are captured when the Francis turbine operating point moves through conditions of zero and negative torque. The proposed methodology is fast and simple to present a qualitative analysis of the flow physics and the turbine behavior during load rejection.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A002. doi:10.1115/FEDSM2016-7538.

A numerical study is conducted to explore the performance and efficiency of Single Dielectric Barrier Discharge (SDBD) plasma actuators for controlling the turbulent boundary layer separation that occurs on the blades of a centrifugal fan. The numerical approach is based on the computational method developed previously to couple a DBD Electro Hydro-Dynamic (EHD) body force model with a RANS/LES flow model. The EHD body force model is based on solving the electrostatic equations for the electric potential due to applied voltage and the net charge density due to ionized air. The efficiency of the actuator at four different alternative current (AC) waveforms including sine, pulse, square, and pulse-amplitude-modulated sine is investigated in this study. The effect of applied voltage on the performance of the plasma actuator is also examined for all waveforms.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A003. doi:10.1115/FEDSM2016-7563.

Laminar to turbulent transition occurs in a broad range of industrial applications, and in nature. There are many mechanisms (natural or bypass) that lead to transition. Accurately predicting both the onset location and length of transition has been persistently difficult. A new, local, intermittency-function-based transition model for both low (< 1%) and high freestream turbulence intensity flows, over smooth and rough surfaces, is introduced and formulated. It is coupled with the k-ω RANS model.

The intermittency model was validated on the ERCOFTAC experimental zero-pressure-gradient smooth flat plate boundary layer cases T3A-, T3A, T3B with leading-edge freestream turbulence intensity 0.9%, 3.5%, 6%, respectively. Skin friction profiles agree well with the experimental data. The model was then tested on periodic wakes, and flows over Stripf’s turbine blades with a broad range of roughnesses, from hydraulically smooth to fully rough. The predicted skin friction and heat transfer properties by the current model agree well with the published experimental and numerical data.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A004. doi:10.1115/FEDSM2016-7589.

Generally, artificial lifts to pump crude oil having a high viscosity from wellbores using an electrical submersible pump (ESP) are not efficient. The present study consists of a numerical approach to understand the effect of fluid viscosity and surface roughness of the flow passage on the performance of an ESP. A three-dimensional numerical analysis was carried out using Reynolds-averaged Navier-Stokes equations at different off-design conditions. The standard k-ε turbulence model was used for the steady incompressible flow. Water and crude oils having different viscosities were used as working fluids and numerical analyses were made by varying surface roughness of the flow passage. Although there was a sharp drop in the efficiency with the increase in surface roughness, but the combined effect of viscosity and surface roughness showed an increase in efficiency up to a certain fluid viscosity.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A005. doi:10.1115/FEDSM2016-7667.

A reverse running centrifugal pump is one of the attractive choices in micro-hydropower development and industrial pressure energy recovery. One of the main problems in utilizing pump as turbine (PAT) is that the performance of PAT is usually not ideal due to the impeller with the routine backward curved blades which do not match well with turbine running condition. A cost effective suitable way for solving this problem is to redesign impeller with forward curved blades from turbine working condition while the other components do not undergo any modifications. Blade inlet width is one of the main factors in impeller design. Therefore, research on the influence of blade inlet width on PAT performance is useful. In this paper, based on the constant velocity moment theory, the velocity moment at impeller inlet is acquired, firstly. Next, a relationship expression between blade inlet angle and the design flow rate is deduced. To perform research on blade inlet width influencing PAT’s performance with special impeller, three impellers which inlet widths are 13 mm, 16 mm and 19 mm, respectively, are designed by using ANSYS Bladegen software. Numerical simulation and analysis of the three PATs are performed using a verified computational fluid dynamics (CFD) technique. Comparison of three PATs’ performance curves obtained by CFD, we can find that the blade inlet width has obvious effect on the performance of PAT. The flow rate, required pressure head, generated shaft power, and efficiency at best efficiency point (BEP) increase with the increase of blade inlet width. The flow rates of three PATs at BEP are about 90 m3/h, 100 m3/h and 105 m3/h, respectively, when impeller inlet width varies from 13 mm to 16 mm and 19 mm. The BEP of three PATs shifts towards higher discharge and its high efficiency range becomes wider with the increase of blade inlet width. At above 100 m3/h discharge, the PAT efficiency increases in accordance with the increase of blade inlet width. And the hydraulic loss and turbulence kinetic energy loss within impeller decrease with the increase of blade inlet width. In order to improve efficiency, it is helpful to choose a relatively larger blade inlet width in the design of special impeller using in turbine mode of PAT.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A006. doi:10.1115/FEDSM2016-7763.

An experimental investigation is carried out to evaluate the performance of bi-directional pumping ring for dual mechanical seals. An experimental setup is constructed, and appropriate instrumentation are employed to measure the pressure, temperature, and flow rate of the barrier fluid. The experimental study focused on the influence of the kinematic viscosity of the barrier fluid on the performance of the pumping ring. This is done by monitoring Q-ΔP curves for various barrier fluids and temperature ranges. The tests are conducted for industrial barrier fluid compositions which are recommended by vendors of dual mechanical seals, such as Propylene Glycol-water mixtures and Diesel fuel (Grade D2).

Despite the importance of the measured performance curves for process control purposes, indicated experimental curved are generalized so that they can be employed in prediction of pumping ring performance for different sizes and operational conditions. Moreover, they can be utilized for numerical models validation.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A007. doi:10.1115/FEDSM2016-7799.

It is important in development of turbomachinery to predict their performance precisely. Especially the prediction of multistage pump performance is one of the challenging problems because internal phenomena which relate to the performance are complicated. Therefore, in this research, we verified accuracy of Computational Fluid Dynamics (CFD) in predicting performance of a five-stage high-pressure volute pump by comparing predicted values by CFD with measurement data. We tried two methods to predict the pump performance. One is a computation with a complete pump model which includes all five stages and leakage passages. This method can be expected to represent total internal flow phenomena. The other method is totaling up the performance data from separate computations of 1st–2nd stages and series stages. This method is simpler than the former and involves less computational cost. As a result, it was clarified that all the methods could predict pump head at the best efficiency point to some extent, even by steady computation. However, no prediction can predict positive gradient in Q-H curve which was observed in measurement at low flow rate. Except for the unsteady complete pump model computation, efficiency and shaft power could not be predicted precisely. In addition, at high flow rate, unsteady computation of the complete pump model shows the best agreement in head. In the complete pump model computation at high flow rate, the series stage next to the long crossover has larger head because of the influence of it. Therefore, the separated model has difficulty in representing series stages’ performance. In order to predict performance at high flow rate, unsteady computations also including properly the influence of the long crossover properly are necessary. In addition, to predict performance at low flow rate, unsteady computation is necessary.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A008. doi:10.1115/FEDSM2016-7831.

In order to improve overall performance of a turbomachinery, it is important to reduce losses of stationary flow passages, such as diffusers and return channels, as well as impellers. For multi-stage pumps, to achieve high uniformity of the inlet flow of the latter impeller can prevent degradation of subsequent performance.

A two stages high pressure pump which was developed by the authors applied vaned diffuser sections located downstream of the first stage impeller and second stage impeller. The vaned diffusers can be replaced to fit the required specification of operating condition. This design is able to help pumps share the casings so that customers can purchase our products at a low cost.

However, the loss of the first stage diffuser section and crossover downstream of the first stage diffuser to the second impeller is large, so that it has a negative impact on overall performance of the pump. It was difficult to reduce loss of the both sections by conventional way such as trial and error approach by modifying geometrical parameters. If we try to reduce the loss of the diffuser section, loss of the crossover passage increases.

We therefore applied a technology called Adjoint method to the design optimization of the diffuser and crossover sections of the pump. The adjoint method has recently been put to practical use. By using this method it is possible to obtain a complex three-dimensional shape for realizing an optimum flow field in a very short time while maintaining a high degree of freedom.

In this study two objectives were specified in the design optimization of the stationery sections; one is to increase the uniformity of the flow field at the inlet of the second stage impeller and the other is to minimize the loss of the first stage diffuser and crossover passage. In the optimization process, the change of sensitivity of the geometry deformation to the weights of objective functions was analyzed by sensitivity maps. The sensitivity maps show directions of the geometry modification depending on the objectives. The geometry in this design was optimized with 50% - 50% weighting ratio between flow uniformity and pressure loss. The loss of the vaned diffuser and crossover passage, velocity uniformity at the outlet of the crossover passage were verified with Computational Fluid Dynamics (CFD). The improvement of pump stage performance was expected by the optimization.

This optimization design procedure applying Adjoint method is effective to improve overall performance of a two stages high pressure pump which has complex three-dimensional flow passages.

Commentary by Dr. Valentin Fuster
2016;():V01AT02A009. doi:10.1115/FEDSM2016-7875.

Due to the limitation in volume and size of vehicles’ integrated starter generators (ISGs), an effective cooling mechanism has to be considered while designing the ISGs. The main purpose of this research is to improve the fins of heat sink design of the ISG to enhance the cooling performance and fan efficiency. To do this, conjugate heat transfer simulations have been performed using the realizable k-ε turbulence model. Also, the MRF method is employed for simulation of the rotating parts. Base on the analysis of the simulation result of the baseline model, the modified models are suggested and investigated. The rotation speed of the fan is 2000 rpm and the number of blades is 11 and 9 for the front and rear fans, respectively. By assuming that the flow in narrow gap between the rotor and stator is negligible, only the rear part which includes the heat sink is simulated. Firstly, conjugate heat transfer simulation is conducted for the baseline model and the heat transfer characteristics are observed. Then, from the observation, modified heat sink model with wider fin spacing is suggested and compared. In order to have more uniformly distributed flow field, a heat sink model with radially arranged fins is also made and simulated. As a result, it is concluded that increasing the air passages between the fins increases the heat transfer. In addition, it is observed that the optimized fan had better performance. Furthermore, radial fin structure is turned out to have advantage for cooling compare to the original heat sink structure. In a radial model, more uniform distribution of the flow is observed, both of flowrate and fan efficiency are higher and temperature of the coil is lower.

Commentary by Dr. Valentin Fuster

Symposium on Applications in CFD

2016;():V01AT03A001. doi:10.1115/FEDSM2016-7542.

This work involves a methodology for Mechanical Engineering students to learn Computational Fluid Dynamics playing an active role. Students carry out a fluid mechanics down scaled projects with the steps of sensibility of mesh, convergence of numerical algorithm, validation of turbulence model and description of flow patterns. Furthermore, they develop their critical thought when they identify weak points susceptible for improvement.

The offer of benchmark test cases ranges from head loses, driven cavities, swirling flows, to external aerodynamics. Simplifications to the level of undergraduate courses imply two dimensional simulations and a limited number of grid points. Hence, the assessment is based in coherence of decisions and efficient use of limited resources.

A review of the offer of workshops is supplied, such as the Ahmed car, the Roback and Johnson burner, aerodynamics of different NACA airfoils, and different geometries of driven cavities. These are classical test cases of numerical research and a sample of applications in wind energy, industrial furnaces, and lubrication.

Parametrization based in geometry, Reynolds number, Pitch angle among other, let simulate different flow patterns with similar degree of difficulty.

There is a deeper understanding of the topic when students need to discuss the strategies to accomplish the project, they need to write a technical report and finally they need to justify the evaluation of other works. Also, it is important to link the simplified projects of the workshop with the real world and the industrial applications.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A002. doi:10.1115/FEDSM2016-7543.

Reactive flow simulations involving turbulence-chemistry interactions can be very challenging because of the strong non-linear coupling between chemistry and fluid dynamics. Furthermore, the chemistry is described with hundreds of reactions, which is prohibitive to solve using computational fluid dynamics (CFD). Using a reduced set of mechanisms that contains a subset of the important species is more practical. However, the species modeled must capture the key combustion characteristics of interest, such as ignition, species distributions and major pollutant formation. Previously, the authors used the joint probability density function (PDF) to study the non-premixed turbulent flames, and continue the work here. New CFD simulations were conducted for a non-premixed turbulent syngas flame using four reduced mechanisms models (3-step, 8-step, 9-step and 12-step reactions) to assess the predictive capabilities in the calculation of turbulence-chemistry interactions. The performance of the different reduced mechanism models was assessed and compared with previous PDF model results and the experimental results of Correa and Gulati (1992, “Measurements and Modeling of a Bluff Body Stabilized Flame,” Combustion and Flame, 89(2)). The predictions of temperature and species from the reduced mechanisms of the 3-step and 8-step were found to have differences as large as 20%. It was also found that the reduced 12-step mechanism was able to represent the strong turbulence-chemistry interactions in the syngas flame and demonstrated good ability of predicting species distribution. Therefore, a simplified chemical mechanism model was successfully developed to simulate the non-premixed syngas flame. The 12-step reduced mechanism will guide other reduced mechanism models for syngas fuels. However, the PDF method still gives the best predictions of temperature and requires the smallest computational time.

Topics: Flames
Commentary by Dr. Valentin Fuster
2016;():V01AT03A003. doi:10.1115/FEDSM2016-7581.

Louisiana coast experiences significant erosion due to wave actions. The loss of beaches in some coastal areas in Louisiana is severe. There are wetlands and marshes located in the coastal areas. Wetland loss is a major threat to the coast areas. 3D numerical simulations of wave-levee interactions were conducted, and the results were analyzed to determine the flow characteristics and surface shear distributions. The simulation setup is exactly the same as an experiment conducted in a wave tank facility. The velocity histories on different locations near the test levee surface were compared, and the agreement is very good, therefore the simulation is validated. A test levee system was also constructed on a test Gulf beach site, approximately 4.6 miles west of Holly Beach in Cameron Parish, Louisiana. Long term observation of erosion was conducted, and survey data showing the change of the test levee were produced. From the observations, the loss of this portion of Gulf beach is significant during the 2-year research period. Real-time images were recorded to show this significant change in topography. The losses of the levee materials during the entire project period were quantified based on the survey data. The history of the loss was plotted. It indicates some major storm event contributed to significant losses and erosion of the test structure. It can be seen from the results that the real-time erosion pattern on the test site agrees reasonably with the surface shear patterns from the simulations. In the numerical simulation, commercial package ANSYS-FLUENT was used. A free-surface flow model is adopted with open channel wave boundary conditions. A grid-independence study was performed to determine to appropriate grid resolution to be used in the simulation. Parallel computing was conducted due to the expensive cost of this 3D simulation with relatively fine grid resolutions.

Topics: Waves , Levees , Shorelines
Commentary by Dr. Valentin Fuster
2016;():V01AT03A004. doi:10.1115/FEDSM2016-7596.

The paper presents a three-dimensional (3-D), time-dependent Euler-Lagrange multiphase approach for high-fidelity numerical simulation of strongly swirling, turbulent, heavy dust-laden flows within large-sized cyclone separators, as components of the state-of-art suspension preheaters (SPH) of cement kilns.

The case study evaluates the predictive performance of the coupled hybrid 3-D computational fluid dynamics–dense discrete phase model (CFD-DDPM) approach implemented into the commercial general purpose code ANSYS-Fluent R16.2, when applied to industrial cyclone collectors used to separate particles from gaseous streams. The gas (flue gases) flow is addressed numerically by using the traditional CFD methods to solve finite volume unsteady Reynolds-averaged Navier-Stokes (FV-URANS) equations. The multiphase turbulence is modeled by using an option of Reynolds stress model (RSM), namely dispersed turbulence model. The motion of the discrete (granular) phase is captured by DDPM methodology.

The twin cyclones of SPH top-most stage have been analyzed extensively both for the overall pressure drop and global collection efficiency, and for the very complex multiphase flow patterns established inside this equipment. The numerical simulation results have been verified and partially validated against an available set of typical industrial measurements collected during a heat and mass balance (H&MB) of the cement kiln.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A005. doi:10.1115/FEDSM2016-7609.

In this study, the complete dynamic performance of the high temperature and high pressure steam pressure relief valve (HTHP PRV) from pop up to reseating was simulated by CFD software which combined with moving mesh capabilities and multiple domains. An experimental setup was established for the testing of HTHP PRV in accordance with the standard of ASME PTC 25. The dynamic performance of HTHP PRV was recorded accurately. For the transient simulation of HTHP PRV, a domain with opening boundaries connected to the outlet of PRV was proposed to avoid the direct definition of the pressure at the PRV outlet and handle the critical flow. It also can describe the surrounding flow field and help us to understand the influence of the PRV discharge on the environment better. The simulation results were verified by experimental ones. The resultant force on the disk and the lift were monitored and analyzed. A detailed contour of the compressible steam flowing through the HTHP PRV was obtained, including small scale flow features in the back pressure chamber. The effect of the adjusting sleeve on the dynamic performance of HTHP PRV was also investigated in details. The blowdown increases linearly by 0.163% with the adjusting sleeve moves by each millimeter in the direction of departing from the disk. This study sheds a light of understanding of the dynamic characteristics of HTHP PRV.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A006. doi:10.1115/FEDSM2016-7614.

Supersonic steam ejectors are widely used in many industrial applications, for example for refrigeration and desalination. The experimental evaluation of the flow field inside the ejector is relatively difficult and costly due to the occurrence of shock after the velocity of the steam reaches over the sonic level in the ejector. In this paper, numerical simulations are conducted to investigate the detailed flow field inside a supersonic steam (water vapor being the working fluid) ejector. The commercial computational fluid dynamics (CFD) flow solver ANSYS-Fluent and the mesh generation software ANSYS-ICEM are used to predict the steam performance during the mixing inside the ejector by employing two turbulence models, the k-ω SST and the k-ε realizable models. The computed results are validated against the experimental data. The effects of operating conditions on the efficiency of the ejector such as the primary fluid pressure and condenser pressure are studied to obtain a better understanding of the mixing process and entrainment. Velocity contours, pressure plots and shock region analyses provide a good understanding for optimization of the ejector performance, in particular how to increase the entrainment ratio.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A007. doi:10.1115/FEDSM2016-7615.

Forward Osmosis (FO) driven asymmetric membrane filtration is a developing technology which shows promise for seawater desalination and wastewater treatment. Due to the fact that asymmetric membranes are widely used in conjunction with this technology, internal concentration polarization (ICP), a flow-entrainment effect occurring within such membranes, is a significant if not dominant source of overall osmotic pressure loss across the membrane. Accurate modeling of ICP effects is therefore very critical for accurate Computational Fluid Dynamic (CFD) modeling of asymmetric membranes. A related, dilutive effect known as external concentration polarization (ECP) also develops on both the rejection and draw sides of the membrane, further contributing to osmotic pressure loss. In order to increase the overall water flux, circular spacers can be implemented within the draw channel of FO cross-flow membrane exchange units to decrease the effects of ICP and draw ECP. The drawback of spacer inclusions is an increased pressure loss across the length of the feed channel. The system efficiency gained by the decrease in ECP must therefore be weighed against the energy cost of hydraulically making up lost channel pressure. To model the geometry of a FO cross-flow channel, the open source CFD package OpenFOAM is used. A compressible flow model with explicit boundary conditions is developed to simulate the flux transfer and ICP effects present within an asymmetric membrane when exposed to a NaCl solution. Results are validated by comparison with the numerical data generated by earlier models of asymmetric membranes implemented by other investigators using similar simulation conditions.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A008. doi:10.1115/FEDSM2016-7637.

The dynamics of the flow in a two-dimensional channel with an arrangement of periodic cavities in an oscillating lower bounding wall and a confining top wall is studied numerically in this paper. The open-source CFD code, OpenFOAM v.2.2.2, based on the Finite Volume method is used to solve the problem. The flow dynamics is studied with respect to variations in the location of the confining top wall (non-dimensionalized with the Stokes layer thickness z/δ), cavity-size-based Reynolds Number (Red) and the ratio of the Stokes Layer Thickness to the Cavity Size (δ/d). This is an extension of work done for a previous paper [20]. The evolution and transport of vortex structures in the vicinity of the are presented over the oscillation cycle time by means of instantaneous streamline plots. The effect of the location of the confining top wall on the flow structure is presented in some detail by comparison of the evolution of the streamlines to a case where the top wall is absent. The transfer of fluid mass in and out of the cavity is shown to be strongly dependent on the Red and the z/δ (for the same Red, smaller the z/δ, lower is the mass transfer efficiency).

Commentary by Dr. Valentin Fuster
2016;():V01AT03A009. doi:10.1115/FEDSM2016-7639.

A solar chimney is a natural ventilation technique that has a potential to save energy consumption as well as to maintain the air quality in the building. However, studies of buildings are often challenging due to their large sizes. The objective of the current study was to determine relationships between small- and full-scale solar chimney system models. In the current work, computational fluid dynamics (CFD) was utilized to model different building sizes with a solar chimney system, where the computational model was validated with the experimental study of Mathur et al. The window, which controls entrainment of ambient air, was also studied to determine the effects of window position. Correlations for average velocity ratio and non-dimensional temperature were consistent regardless of window position. Buckingham pi theorem was employed to further non-dimensionalize the important variables. Regression analysis was conducted to develop a mathematical model to predict a relationship among all of the variables, where the model agreed well with simulation results with an error of 2.33%. The study demonstrated that the flow and thermal conditions in larger buildings can be predicted from the small-scale model.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A010. doi:10.1115/FEDSM2016-7657.

Smoke sampling devices are used in several fields to study dynamics of smoke aerosols. An important criterion in designing smoke sampling devices is that flow paths leading to where the sample is characterized are constructed such that deposition of aerosol particles along the paths is minimized. Sampling devices often include a Venturi flow meter installed downstream of the smoke source, which may significantly alter the composition of the aerosol reaching the sample analyzer. The current work employs Computational Fluid Dynamics (CFD) to model particle deposition within the flow meter and to examine the effects of different design parameters. This study focuses on particles with sizes ranging from 0.01 to 100 microns, for which three main mechanisms for deposition can be identified: inertial impaction, gravitational sedimentation, and Brownian diffusion. It has been shown that inertial deposition is negligible for ultrafine particles (5–560 nm) and it becomes noticeable for particles in the micron size range. Also, deposition fractions increase with increasing particle sizes. Moreover, inertial particle deposition increases with increasing volume flow rates.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A011. doi:10.1115/FEDSM2016-7671.

A liquid mercury target is used at Oak Ridge National Laboratory’s (ORNL [1]) Spallation Neutron Source (SNS [2]) to generate neutrons. The mercury is flowing in a stainless steel containment vessel for neutron spallation, but also to cool the vessel itself. Computational Fluid Dynamics (CFD) simulations have been used to estimate the temperature and pressure fields needed for the thermal stress analysis. Because of the geometry complexity, the high turbulence number, and the computational time requirements, generating a quality mesh that can accurately capture the flow and heat transfer has always been a challenge. However, with today’s High Performance Computing (HPC) advances, larger and larger meshes can now be used and better accuracy can be achieved. In this study, two meshing methods were used for the SNS jet-flow target: automatic tetrahedral method (ANSYS meshing) and manual hexahedral meshing (ICEM-CFD). Both methods are compared in terms of quality, size, ease of generation, convergence, and user-friendliness. Both meshes were used with ANSYS-CFX to simulate the steady, Newtonian, single phase, isothermal, incompressible and turbulent flow in the target. The Shear Stress Transport (SST) k-ω model developed by Menter [3] was used for turbulence modeling. The accuracy of the CFD simulations are tested against experimental data presented in the current paper. An in-depth series of Particle Image Velocimetry (PIV) measurements performed on a “visual jet-flow target”, an acrylic replica target running with water, are presented in the paper. Since flow measurements in mercury are difficult, a water loop was built to investigate the flow in the target and a potential gas injection in the flow to mitigate the pressure wave [4]. A PIV system on a precise translation stage was setup on the water loop to perform detailed and accurate PIV measurements. Mean flow velocity fields were used to validate the CFD simulations. The paper concludes on the choice for mesh generation for future target analysis, and the path forward for CFD simulations for the future SNS targets.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A012. doi:10.1115/FEDSM2016-7675.

Cerebrospinal fluid (CSF) shunts are fully implantable medical devices that are used to treat patients suffering from conditions characterized by elevated intracranial pressure, such as hydrocephalus. One of the primary causes of CSF shunt failure is mechanical obstruction of the ventricular catheter, a component of the shunt system implanted directly into the brain’s ventricular system. This study aims to characterize the CSF flow through ventricular catheters via a 3-dimensional computational fluid dynamics (CFD) model. The fully-parametrized model has allowed for exploration of the catheter’s geometric design features, with the goal of reducing the incidence of catheter obstruction. As the first step towards this goal, a design optimization study was performed with the objective of achieving a uniform flow rate distribution among the catheter’s inlet holes. To perform this study, the CFD model was coupled with an optimization framework, and a large number of simulations were run on a high-performance computing system to determine the optimal design for target flow performance. This optimization study advances the field of CSF shunt design by providing systematically derived correlations between the catheter’s geometric parameters and CSF flow through the catheter’s inlet holes.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A013. doi:10.1115/FEDSM2016-7687.

Hole-pattern annular gas seals have been proven to be very effective in reducing leakage flow between high and low pressure sections in turbomachinery. This type of seal has two distinct flow regions: an annular jet-flow region between the rotor and stator, and cylindrical indentions in the stator that serve as cavities where flow recirculation occurs. As the working fluid enters the cavities and recirculates, its kinetic energy is reduced, resulting in a reduction of leakage flow rate through the seal. The geometry of the cylindrical cavities has a significant effect on the overall performance of a hole-pattern annular gas seal. Previous studies have been primarily focused on cylindrical cavities that are perpendicular to the axis of the seal and have indicated that the performance may be improved by varying the depths, spacing, and diameters of the cavities. However, to date the effects of elliptical cylinder cavities has yet to be investigated. In this study, the effects of elliptical shape hole pattern geometry on the leakage and dynamic response performance of an industry-relevant hole-pattern seal design are investigated using a combination of computational fluid dynamics (CFD), hybrid bulk flow/CFD analysis, and design of experiments techniques. A CFD model of the baseline hole-pattern seal was first developed and validated against experimental data. A design of experiments (DOE) study was then performed to investigate the effect that various elliptical shape cavities had on the leakage rate through the seal. CFD simulations were run for multiple geometry configurations of the cylindrical cavities to evaluate the seal performance at each of the design points. The design space was defined by varying the values of five geometrical characteristics: the major and minor radius of hole, the angle between the major axis and the axis of the seal, the spacing between holes along the seal axis, and hole spacing in the circumferential direction. Quadratic polynomial fitting was then used to analyze the sensitivity of different design variables with respect to the different outputs. This detailed analysis allowed for a greater understanding of the interaction effects from varying all of these design parameters together as opposed to studying them one variable at a time. Response maps generated from the calculated results demonstrate the effects of each design parameter on seal leakage as well as the relationships between the design parameters. The data from this analysis was also used to generate linear regression models that demonstrate how these parameters affect the leakage of the seal. The results of this study could aid in improving future designs of hole-pattern annular gas seals.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A014. doi:10.1115/FEDSM2016-7712.

A commercial CFD code (ANSYS CFX, release 16.2) is used to predict the turbulent flow phenomena over a wavy wall. The present work will provide numerical simulations of flow in a channel with a wavy lower wall using a variety of turbulence models available in the CFD commercial code. Eddy viscosity models and Second Moment Closure models were used with wall function available. Those turbulence models had different predictions for the flow field, in which were evaluated: velocity profiles, pressure distribution, wall shear stress, recirculation region and turbulence quantities. A comparison between their predictions will be presented. The validation of results is performed by comparison to experimental data from previous studies and also LES simulations.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A015. doi:10.1115/FEDSM2016-7719.

In this study a numerical model of the insulin depot formation and absorption in the subcutaneous adipose tissue is developed using the commercial Computational Fluid Dynamics (CFD) software ANSYS Fluent. A better understanding of these mechanisms can be helpful in the development of new devices and cannula geometries as well as predicting the concentration of insulin in the blood. The injection method considered in this simulation is by the use of an insulin pump using a rapid acting U100 insulin analogue. The insulin is injected into the subcutaneous tissue in the abdominal region. The main composition of the subcutaneous tissue is blood vessels and adipose cells surrounded by interstitial fluid. The numerical simulation is conducted in a 2D-axisymmetric domain where the tissue is modeled as a fluid saturated porous media. Due to the presence of channel formation in lateral direction in the tissue, an anisotropic approach to define the permeability is studied having an impact on the viscous resistance to the flow. This configuration is resulting in a rather disk shaped depot following recent experimental findings. The depot formation is analyzed running Bolus injections ranging from 5–15 Units of insulin corresponding to 50–150μl. The depot formation model has been extended implementing the process of absorption of insulin from the depot to be able to run the simulation over longer timeframes where absorption starts playing an important role.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A016. doi:10.1115/FEDSM2016-7743.

Impinging jets are an attractive option for spot-cooling of high heat flux devices because of the stagnation region surrounding the point of impingement and the resulting high heat transfer. In small devices with a small jet (microjet) , however, the cooled region due to just a single jet is small. One way to potentially increase this area exposed to the impinging jet is by oscillating the heated surface. In the current paper, the flow structure and transport in a confined and submerged jet impingement arrangement impinging on a wall oscillating horizontally is numerically studied with respect to both parameters governing jet impingement :Jet Reynolds Number (from 40 to 200), distance from the jet inlet to the impinging wall (z/d ratios of 2 and 5) and an oscillation parameter (oscillatory peak Reynolds Numbers of 55 and 110). OpenFOAM v 2.2.2, an open-source CFD code based on the finite volume method is used to solve the problem. The Grid Convergence Index (GCI) is used to estimate discretization uncertainty and error bars on all of the parameters calculated. The flow structure in a confined submerged jet is made up of a double recirculation zone (mostly attributed to the confining top wall) the reattachment regions are associated with a secondary peak in the Nusselt Number. Heat transfer is not studied in this paper. The effect of the oscillating lower wall on the locations of the primary and the secondary recirculation zones are studied with respect to all the parameters mentioned above over a complete oscillation cycle. The local skin friction coefficients along different sections on the lower wall are computed along sections of the oscillating wall and compared to the case where there is no oscillation.The results are anticipated to have significant impact on the heat transfer enhancement possible in such an arrangement.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A017. doi:10.1115/FEDSM2016-7747.

Computational Fluid Dynamics (CFD) has become a widely used tool in industry as the cost for simulations is usually lower than the cost for multiple experiments. CFD is an effective tool for comparing design alternatives, investigating specific flow features and in some cases it may be the only feasible option for studying engineering flows. As a result, the demand for mechanical engineers with CFD skills keeps increasing. Nevertheless CFD is still not adequately presented in undergraduate engineering curricula, which can lead to expensive mistakes, if for example it is relied on without understanding its limitations. One excellent platform for CFD, which can be introduced to fluid mechanics classes, is the open-source environment OpenFOAM, which is widely used in both academia and industry. In addition to being open-source, OpenFOAM code can be viewed and modified by the user, and a wide range of modules for OpenFOAM are available with new modules being developed constantly. One major disadvantage, however, is that OpenFOAM has a rather steep learning curve and although there are many resources available online, it is difficult to find short introductory courses. A tutorial was developed to provide a brief introduction to OpenFOAM and allow the students to perform simple simulations. Upon completing the tutorial, the students can build their own simulations. The tutorial covers geometry, mesh, boundary and initial conditions, solvers, schemes, post processing, and some additional features, such as shell scripts and parallel processing. A large portion of the tutorial is devoted to the geometry and mesh generation as this is one of the more challenging aspects of OpenFOAM compared to conventional graphical user interface CFD packages. Nevertheless, the students are exposed to the importance of properly setting the other simulation parameters through simple examples — e.g., comparing 2D channel flow simulations using potential flow and using turbulence modeling. One crucial aspect of the tutorial is that students are encouraged to experiment with deliberate modifications of the simulations to experience and understand how some of them do not provide reasonable results. Although the tutorial is rather brief and does not cover the topics in much detail, it aims to familiarize students with the basics of OpenFOAM, so that they can better understand other relevant resources. The OpenFOAM tutorial offers an alternative introduction to CFD compared to commercial CFD packages, which may not be readily available. The tutorial has already been utilized for three consecutive years at the University of New Hampshire, mostly by undergraduate students who worked/are working on senior projects involving CFD. The feedback has been generally positive.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A018. doi:10.1115/FEDSM2016-7808.

Despite recent interest in complex fluid-structure interaction problems, the baseline fluid modeling capability for commercially available numerical methodologies used for multidisciplinary analysis is yet to be established. The current work is among those to first underline such a reference for coupled Lagrangian-Eulerian (CLE) and smooth particle hydrodynamics (SPH). These methodologies are quantitatively assessed using the classical 2-D lid-driven cavity and compared against an implicit Navier-Stokes solution in addition to other benchmarks from the literature. Qualitative comparison is made through the use of velocity magnitude contour plots with accompanying streamlines, whereas quantitative analysis is made using centerline velocity profiles for both U and V flows. Throughout the investigated Reynolds numbers (1000–20,000), SPH provides inaccurate results and is unable to represent vorticity in the cavity corners. Alternatively, CLE retains a high level of accuracy up to Re = 10,000, before deviating from published literature at Re = 20,000. In addition to being qualitatively similar, the centerline profiles consistently display ≤ 10% error when compared to the Navier-Stokes solutions. By establishing the limits of closed-system fluid modeling capability for SPH and CLE, this work can be extended to full fluid-scenarios.

Topics: Fluids , Modeling , Cavities
Commentary by Dr. Valentin Fuster
2016;():V01AT03A019. doi:10.1115/FEDSM2016-7857.

Flow-induced vibration (FIV) is a widespread problem in energy systems because they rely on fluid movement for energy conversion. Vibrating structures may be damaged as fatigue or wear occur. Given the importance of reliable components in the nuclear industry, flow-induced vibrations have long been a major concern in the safety and operation of nuclear reactors. In particular, nuclear fuel and steam generators have been known to suffer from flow-induced vibrations and related failures. Over the past five years, the Nuclear Energy Advanced Modeling and Simulation program has developed the integrated multiphysics code suite SHARP. The goal of developing such a tool is to perform multiphysics modeling of the components inside a reactor core, the full reactor core or portions of it, and be able to achieve that with various levels of fidelity. This flexibility allows users to select the appropriate level of fidelity for their computational resources and design constraints. In particular SHARP contains high-fidelity single-physics codes for structural mechanics and fluid mechanics calculations: the structural mechanics implicit code Diablo and the computational fluid dynamics spectral element code Nek5000. Both codes are state-of-the-art. highly scalable (up to millions of processors in the case of Nek5000) tools that have been extensively validated. These tools form a strong basis on which to build an FIV modeling capability.

This work discusses in detail the implementation of a fluid-structure interaction methodology in SHARP for simulating flow-induced viration based on the coupling between Diablo and Nek5000. Initial verification and validation efforts are also discussed, with a focus on standard benchmark cases: the flow past a cylinder, the Turek benchmark, and the flow in a Coriolis flow meter.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A020. doi:10.1115/FEDSM2016-7898.

Recently, the cooling system of hydraulic excavator is often designed using the thermal and fluid analysis to improve the cooling performance. The reliability of the analysis results is important, since it directly influences on the efficiency of development. In the present study, the uncertain parameters were estimated using the data assimilation method to increase the reliability of the thermal and fluid analysis in an engine room of a hydraulic excavator.

The ensemble Kalman filter (EnKF) was adapted as a data assimilation method, and the thermal and fluid analysis was conducted with the three-dimensional steady simulation based on the Reynolds-average Navier-Stokes equations. The estimated parameters were set to the total heat quantities released by heat exchangers and the flow rates of the coolants. The total heat quantity is a parameter used for the heat release calculation of a heat exchanger, and the flow rate of a coolant is specified at the inlet boundary. As measurement data, temperatures of coolants which were measured at the upstream and downstream of the heat exchangers were used. Initial parameters were generated by setting parameter values in a random manner.

The simulation using estimated parameters successfully predicted temperatures at the heat exchangers, where the maximum error was 3K. In addition, the reductions of the standard deviations of the uncertain parameters were confirmed. That means the reliability of the simulation was increased.

Commentary by Dr. Valentin Fuster
2016;():V01AT03A021. doi:10.1115/FEDSM2016-7934.

In this work, a sensitivity calculation approach called the adjoint method is explored to estimate and control the discretization error in computational fluid dynamic (CFD). This paper describes how the adjoint is applied to incompressible RANS simulations to approach the continuous solution which is unknown. While the adjoint method is already widespread in aerodynamics to optimize designs, here it is used as an error estimator. The error is thus calculated from the scalar product of sensitivities to an objective function selected by the user (efficiency, power, losses, etc.) with respect to residuals of governing equations. The key point is that adjoint method pinpoints the sensitive areas of the simulation. By adapting the mesh accordingly, it is possible to improve the numerical accuracy while keeping the mesh size manageable.

Topics: Simulation , Errors
Commentary by Dr. Valentin Fuster

7th Symposium on Bio-Inspired and Bio-Medical Fluid Mechanics

2016;():V01AT04A001. doi:10.1115/FEDSM2016-7507.

The top 200 meters of oceans abound in life forms since photosynthesis is possible in that layer. Competition and predator-prey (swordfish-flying fish, 102–104 to 1 mass ratio) interactions are intense here. Chased by predators, a flying fish (FF) — a pleuston — frantically escapes from the water and becomes airborne. Here we report the visual observations of oceanic surface and body distortions of FF to surmise the mechanisms of propulsion during taxiing and landing. FF leaps, not when it is chased, but when the additional energy required for further increase in speed underwater exceeds that required to leap.1 The higher metabolic cost of transport of regular flapping flight in air than in water is circumvented by gliding. We examine the BBCTV video2 by Richard Attenborough, the noted naturalist. An FF may camber its wings like parafoils and may also twist the outer half of the wings during taxiing and climbing. To produce thrust during taxiing, the FF sculls with the lower lobe of the tail fin to produce a reverse Karman vortex jet; there is rapid flicking of the lower lobe of the tail fin tangentially over the surface. The body acts as a chaotic damped and driven pendulum to produce the high-velocity wide flick. To damp after takeoff, it becomes a single asymmetric pendulum. Unpowered (foil) gliding follows. For descent, the wings are shaped, untwisted parafoils and, just prior to touchdown, travelling waves are superimposed, producing, in contrast to taxiing, an impressively smooth small-angle-of-attack tail touchdown on water without any nose-down. The spiked crowns of Richtmyer-Meshkov interface instability are visible on the ocean surface during leaping but not during landing. Trailing hydraulic jumps are observable during landing but not during leaping. The leap is a high-acceleration and Weber number dominated (inertia/capillary forces) phenomenon, but the landing involves little impact force and is dominated by Froude number forces (inertia/gravity forces). The evidence suggests that, prior to leaping and while still underwater, the FF reads the surface wind direction to align the flight path.

Commentary by Dr. Valentin Fuster
2016;():V01AT04A002. doi:10.1115/FEDSM2016-7586.

The wing motion of a flying insect such as a butterfly produces fluid force by manipulating the flow field around the wing. These forces enable a butterfly to rapidly accelerate, turn, and hover. A number of recent studies have examined the flow field around insect wings. In recent years, quantitative flow visualization techniques, such as PIV measurement, have been advanced rapidly, and the study of the flow field around insect wings using PIV has been actively performed. As a result, the two- and three-dimensional vortex structures and their dynamic behaviors have been investigated quantitatively. However, the dynamic behaviors of these vortex structures have not been related to the dynamic force characteristics. The purpose of the present study is to clarify the relationship between the dynamic lift generated by the flapping butterfly wing and the dynamic behavior of the vortex ring rolled up from the butterfly wing as well as to investigate the role of the vortex ring. We conducted a dynamic force measurement of a flapping Cynthia cardui using a six-axes sensor and three kinds of shafts: a straight shaft, a short L-shaped shaft, and a long L-shaped shaft. Moreover, a two-dimensional PIV measurement was conducted in the wake of the butterfly.

The butterfly is given not only negative dynamic lift but also the reactive force (positive lift) due to the jet flow induced by the vortex ring in the upward flapping. As a result, the butterfly produces dynamic lift in the downward flapping and produces not only negative dynamic lift but also dynamic lift in the upward flapping. Based on these results, it is considered that the vortex ring released into the wake contributes to the dynamic lift generated furing flight. That is, it was concluded that the vortex ring rolled up from the flapping wing has an important role in flight even after being released into the wake.

Topics: Vortices , Wings
Commentary by Dr. Valentin Fuster
2016;():V01AT04A003. doi:10.1115/FEDSM2016-7611.

Computational Fluid Dynamics (CFD) has become a routine tool in recent times for use in blood-contacting medical device design and analysis, such as prosthetic heart valves and ventricular assist devices (VADs). While CFD can aid in design by decreasing the need for expensive prototyping and iterative laboratory testing, standardizations are not currently available for CFD to be used for medical device safety analysis at the preclinical stage. To address this, the U.S. Food and Drug Administration (FDA)’s Center of Devices and Radiological Health (CDRH) has sponsored CFD “round-robins.” This paper focuses on Computational Round Robin #2 - Model Blood Pump. The exact geometries, flow conditions and fluid characteristics for the CFD analysis have been supplied to the participants. In the CFD analysis presented in this paper, a rotating fluid zone around the pump impeller was used to avoid the complexities of a dynamic mesh. The rotating fluid zone was modeled by including the centrifugal and Coriolis forces in the Navier-Stokes equations. The Shear Stress Transport (SST) k-ω turbulence model was used and the steady-state solutions for the desired flow conditions were calculated. Current experimental data is still being collected by FDA for the flow conditions given in the study. However, some of the pump operating characteristics are available from work of other investigators and are used to validate the CFD results.

Commentary by Dr. Valentin Fuster
2016;():V01AT04A004. doi:10.1115/FEDSM2016-7612.

Arteriovenous fistula (AVF) is one type of vascular access which is a surgically created vein used to remove and return blood during hemodialysis [1]. It is a long-term treatment for kidney failure. Although clinical treatment and technology have both achieved great improvements in recent years, the vascular access for hemodialysis still has significant early failure rates after the insertion of AVF in patients [2]. Studies have shown that stenosis in the vascular access circuit is the single major cause for access morbidity. Majority of efforts to understand the mechanisms of stenosis formation, and its prevention and management have largely focused on understanding and managing this complication based on the pathophysiology, tissue histology and molecular biology; however these efforts have not resulted in significant progress to date. We believe that the major impact in this area will come from continued and accurate understanding of the hemodynamics, and by development of techniques of intervention to modulate factors such as flow rates, pressures and compliance of the circuit.

The goal of this paper is to study anastomotic models of AV access using Computational Fluid Dynamics (CFD) and optimize them to minimize the wall shear stress (WSS). In order to achieve this goal, the commercial CFD software FLUENT [3] is employed in conjunction with a single objective genetic algorithm [4].

Computations for two types of AVF currently in use in clinical practice are performed. AVF with 25° angle/3–4mm diameter and 90° angle/3–5mm diameter are selected to conduct the optimization. A single-objective genetic algorithm is employed in the optimization process and a k-kl-ω turbulence model is employed in CFD simulations; this model can accurately compute transitional/turbulent flows. In order to optimize for the same flow conditions, a fixed boundary condition is used during the optimization process. Computations for 16 to 20 generations of the selected AVFs are obtained from the genetic algorithm solver. The maximum WSS in the two AVFs considered are 6997.8 and 7750 dynes/cm2; however, the maximum WSS in the shape-optimized AVFs are reduced to 3511.2 and 4293.9 dynes/cm2 respectively, which have decreased by 49.82% and 44.59% respectively. Thus, the probability of the formation of stenosis in AVFs and early failure rates of vascular access are reduced by using the optimized AVFs.

Commentary by Dr. Valentin Fuster
2016;():V01AT04A005. doi:10.1115/FEDSM2016-7682.

The purpose of this study was to investigate fin undulation as a form of locomotion. The analysis generated CFD simulations and models that identify characteristics that are known to indicate propulsive forces. A mechanical undulating fin was designed and built to experimentally validate these computational results. Comparing thrust data from the mechanical fin with the CFD results yielded qualitative agreement with various parameters including wave amplitude, wave speed, and wave number. Quantifying these characteristics are necessary towards understanding the mechanics of undulation and will aid in the design and control of underwater undulating robotics.

Topics: Robotics , Biomimetics
Commentary by Dr. Valentin Fuster
2016;():V01AT04A006. doi:10.1115/FEDSM2016-7699.

Three-dimensional numerical simulations are used to investigate the hydrodynamic performance and the wake patterns of a sunfish in steady swimming. Immersed boundary method for deformable attaching bodies (IBM-DAB) are used to handle complex moving boundaries of one solid body (fish body) attached with several membranes (fins). The effects of the vortices shed from both the dorsal and anal fins on the hydrodynamic performance of the caudal fin are analyzed by prescribing an undulatory swimming kinematics to a full body sunfish model. Results show that both the dorsal fin vortices and the anal fin vortices can increase the thrust and efficiency of the caudal fin comparing to caudal fin only case. This is because the dorsal/anal fin not only can feed vorticity into the caudal fin wake via vortex shedding, but also can modulate the flow in the downstream in a way of forming a jet with stronger backward component.

Commentary by Dr. Valentin Fuster
2016;():V01AT04A007. doi:10.1115/FEDSM2016-7803.

In this study, a shear-dependent continuum model for platelet activation, adhesion and aggregation is validated using computational fluid dynamics (CFD). To take the presence of red cells into account, a combination of excess-platelet boundary layer and enhanced mass diffusivity of platelets and large species is used to mimic this behavior. The model has been validated under three different shear conditions and two different heparin levels. Also three-dimensional simulations were carried out to evaluate the model’s prediction of thrombus growth rate for stenosed tubes under various flow conditions and stenosis degrees. For these cases, also the effect of change in platelet diffusivity has been investigated by using an empirical correlation for enhanced diffusivity of platelets. For all 3D simulations, results for thrombus growth rate as a function of local wall shear rate were compared to those of experiments and numerical studies in the literature and an acceptable agreement was achieved.

Topics: Thrombosis
Commentary by Dr. Valentin Fuster
2016;():V01AT04A008. doi:10.1115/FEDSM2016-7858.

The present work lays out an accurate, three-dimensional computational fluid dynamics (CFD) model of a human blood capillary. This model is composed of red blood cells and blood plasma inside a cylindrical section of a capillary. The plasma flow is resolved using an incompressible Navier-Stokes solver. At the level of capillaries, red blood cells must be individually handled to correctly resolve the hydrodynamics in the system. They cannot be lumped in with the plasma and considered as a non-Newtonian suspension because of the relative scale of the capillaries and the blood cells. Red blood cells act as highly deformable, fluid filled vesicles which readily deform from their typical biconcave shape when passing through narrow capillaries. In the present model, the deformed shape of red blood cells is predicted using a combination of analytical models and experimental data on cell deformation. The cell volume, cell surface area, and plasma layer thickness are found to be the key parameters which define red blood cell deformation in capillaries. The red blood cells are imposed in the flow using the immersed boundary method (IBM). To save computational resources while still yielding an accurate model, the deformed shape of each red blood cell is calculated once prior to running the simulation and then held constant throughout the run. In order to validate the model, two parameters — apparent relative viscosity and hematocrit ratio — were examined. The present model shows good comparison to experimental values for both these parameters.

Topics: Hydrodynamics
Commentary by Dr. Valentin Fuster
2016;():V01AT04A009. doi:10.1115/FEDSM2016-7891.

The aim of this work is to understand the physics underlying the mechanisms of two-dimensional aquatic pollen dispersal, known as hydrophily, that have evolved in several genera of aquatic plants, including Halodule, Halophila, Lepilaena, and Ruppia. We selected Ruppia maritima, which is native to salt and brackish waters circumglobally, for this study. We observed two mechanisms by which the pollen released from male inflorescences of Ruppia is adsorbed on a water surface: 1) inflorescences rise above the water surface and after they mature their pollen mass falls onto the surface as clumps and disperses as it comes in contact with the surface; 2) inflorescences remain below the surface and produce air bubbles which carry pollen mass to the surface where it disperses. In both cases dispersed pollen masses combined with others under the action of lateral capillary forces to form pollen rafts. The formation of porous pollen rafts increases the probability of pollination since the attractive capillary force on a pollen raft toward a stigma is much larger than on a single pollen grain. The presence of a trace amount of surfactant can disrupt the pollination process as the pollen is not captured or transported on the water surface.

Commentary by Dr. Valentin Fuster

86th Symposium on CFD Verification and Validation

2016;():V01AT06A001. doi:10.1115/FEDSM2016-7701.

Interface tracking simulation (ITS) is one of the promising approaches to describe heat transfer of boiling phenomena and their underlying mechanisms. Better understanding and modeling of this process will benefit various engineering systems. In modern nuclear reactors, study on nucleate boiling phenomena is very important for the prediction of the Critical Heat Flux (CHF) phenomena.

The presented research will implement and verify the capability of evaporation process modeling by the massively parallel research code, PHASTA. The comparison of the numerical results and the analytical results demonstrates that the overall behavior of the simulation compares well with the analytical solution. A second simulation of the single bubble growth with non-uniform temperature distribution demonstrates both condensation and evaporation modeling. In the third simulation flow boiling capabilities were preliminary tested with laminar flow demonstration case. These results will be applied to a larger scale, multi-bubble simulations and help modeling of nucleate boiling phenomena.

Commentary by Dr. Valentin Fuster
2016;():V01AT06A002. doi:10.1115/FEDSM2016-7718.

Highly concentrated slurries are found in many different industrial and environmental applications, such as hydro-transport systems of the oil sands industry, drilling and fracturing applications, and stirring vessels. When the volume fraction of particles is low, particles have little influence on the structure of the flow. However, even when average concentration is relatively low, there can still be some regions of high concentration. In highly concentrated flows, the effect of particles on the dynamics of the flow cannot be neglected. Under this condition, particle concentration can affect the turbulence intensity and erosion ratio.

Several experiments have been conducted to examine the effect of different parameters on erosion. Different models have been developed to predict erosion ratio in liquid and gas flows. However, previous studies mostly have examined dilute slurries and less attention has been paid to the effect of high concentration of particles on erosion. In this study, the erosion due to highly concentrated slurries is investigated using both experimental and numerical approaches. There are several parameters such as particle properties and shape, target material, fluid properties and dynamics of the flow that affect erosion ratio. In addition, higher fluid viscosity can significantly affect the flow dynamics and change the interaction behavior between fluid and solid particles.

Effects of particle size and velocity on erosion ratio are investigated for different sand concentrations. Experiments have been conducted for various concentrations, ranging from 1% to about 20% by mass and two different particle sizes, 75 μm and 300 μm. Erosion ratio was calculated based on two different approaches, mass loss and volume loss obtained from 3-D profilometry data. Scanning electron microscope (SEM) images were obtained for 1% and 15% concentration cases to examine the erosion on different parts of the specimens. In addition to the experimental work, a CFD model is setup to simulate the erosion results. The aim of this CFD simulation is to predict erosion rate of the specimen caused by submerged slurry jet flow by using Reynolds stress as turbulence model. The fluid flow solution is obtained using an Eulerian approach and a Lagrangian scheme is used to track the sand particles. In these models, the injected particles from the inlet impact the target wall in order to investigate the erosion.

Commentary by Dr. Valentin Fuster
2016;():V01AT06A003. doi:10.1115/FEDSM2016-7735.

Solid particle erosion has been recognized as a major concern in the oil and gas production industry. It has been observed that erosion can cause serious and costly damage to equipment and pipelines. Accordingly, different studies have been performed in order to investigate erosion caused by solid particles entrained in the flow. Both experimental and modeling approaches have been used in the past to analyze solid particle erosion under different conditions to be able to mitigate these problems.

The goal of this paper is to use a Computational Fluid Dynamic (CFD) erosion model to predict erosion caused by particles flowing in 90 degree and long radius bends. The fluid flow model is coupled with a Lagrangian particle tracking approach. The CFD-based prediction procedure consists of three main steps: flow modeling, particle tracking and erosion calculation. The Reynolds Stress Model (RSM) is used as the turbulence model for all fluid flow simulations. Solid particles are injected from the inlet of the pipe and tracked throughout the bend. The effect of the number of particles released on the predicted maximum erosion magnitude has been investigated. In order to study the grid independency of the solution, erosion is predicted for 5 different grid spacings to accurately predict the flow and erosion rates. In order to assess the quality of the numerical predictions of the erosion rate, experimental data for single-phase (gas) flow with sand in a 3-inch pipe were used. The effects of particle size, fluid velocity, pipe diameter and radius as well as particle rebound model on erosion pattern and magnitude are also investigated. Comparison of these results with experimental erosion data demonstrates good agreement of the erosion trends.

It is found that the location of highest erosion for single-phase (gas) flow at low pressure containing sand is around 45° in the elbow. It has been also observed that the 300 μm particles cause approximately two times higher metal loss compared to the 150 μm particles. This higher erosion magnitude is not only caused by the increase in particle momentum but also by the significant increase in particle sharpness for the 300 μm sand. Moreover, simulation results indicate that the increase in gas superficial velocity leads to an increase in the erosion magnitude. According to the results, erosion ratios were reduced exponentially with the increase in pipe diameter at constant flow conditions and particle properties. Furthermore, two available rebound models in the literature were investigated, and simulations illustrate that both methods are in reasonable agreement with experimental data.

Topics: Erosion
Commentary by Dr. Valentin Fuster
2016;():V01AT06A004. doi:10.1115/FEDSM2016-7889.

In this work, fluid dynamics of a turbulent round impinging jet has been studied using Computational Fluid Dynamics (CFD) and the results have been compared with experimental data from the literature. The fluid was water with density of 1000 kg/m3 and the average velocity of the submerged jet was kept constant at 10.7 m/s while the liquid viscosity varied from 1 cP to 100 cP. Different turbulence models including k-ε, k-ω and Reynolds Stress Model (RSM) have been employed in ANSYS FLUENT and the predicted axial and radial velocity profiles at various distances from the wall are compared with LDV data. It was observed that at locations away from the target wall, predicted velocities are comparable to the measured velocities for all the viscosities. However, near the wall, the deviation between the CFD predictions and experimental measurements become noticeable. The performance of k-ω model and RSM are found to be better than the k-ε model especially for the highest viscous fluid, but no model was found to be superior for all conditions and at all locations.

Commentary by Dr. Valentin Fuster
2016;():V01AT06A005. doi:10.1115/FEDSM2016-7919.

Most methods presented in the literature for estimation of discretization errors focus primarily on steady flows. The transport of error in strongly transient flow has not been adequately addressed. Issues related to transient error calculations are discussed and some methods that are viable for such applications are proposed. Examples are presented on simple flows such as transient Burgers equation followed by applications to more complex flows, e.g. two-phase gas-solid flow relevant fluidized beds. It is demonstrated that error estimation can be made with reasonable accuracy using a combination of various methods.

Commentary by Dr. Valentin Fuster

Symposium on Development and Applications of Immersed Boundary Methods

2016;():V01AT07A001. doi:10.1115/FEDSM2016-7882.

This paper studies energy harvesting of a two-dimensional foil in the wake downstream of a cylinder. The airfoil is passively mobile in the transverse direction. An immersed-boundary method with a fluid-structure interaction model is validated and employed to carry out the numerical simulation. For improving the numerical stability, a modified low-storage 3rd-order Runge-Kutta scheme is implemented for time integration. The performance of this temporal scheme on reducing spurious pressure oscillations of the immersed-boundary method is demonstrated. The simulation shows a foil emerged in a vortical wake achieves better energy harvesting performance than that in a uniform flow. The types of dynamic response for the energy harvester are identified and the properties of vortical wakes are found to be of pivotal importance in obtaining the desired periodic response of the foil.

Commentary by Dr. Valentin Fuster

11th Symposium on DNS, LES and Hybrid RANS/LES Methods

2016;():V01AT08A001. doi:10.1115/FEDSM2016-7501.

This work is devoted to the analysis of the interaction of two coaxial jet with swirling flow using Large Eddy Simulation methodology to reproduce the case of Roback and Johnson. The eight - flat - vanes in the annular nozzle are the precursor of the swirling annular jet with high swirl number.

An implicit LES model was used, this model uses mesh size as filter width so no sub-grid model is required. It is the numerical error who plays the role of dissipative part of the stress tensor in the sub-mesh. Mesh prerequisites are Δy+ = 1 and uniform hexahedral mesh. The resolution scheme used is a Total Variation Diminishing (TVD) with coefficients looking for good precision. PISO is the pressure-velocity coupling used. Besides, multigrid resolution improves the performance towards the full convergence.

Influence of the swirl number on the flow pattern is analyzed. Also the impact of conical diffuser on the mixing is presented.

Topics: Jets , Swirling flow
Commentary by Dr. Valentin Fuster
2016;():V01AT08A002. doi:10.1115/FEDSM2016-7585.

The paper presents a result of the direct numerical simulation with the lattice Boltzmann method which was conducted for quantitative prediction of turbulent broadband noise. For better prediction of broadband noise with high frequency, which is generally generated in high Reynolds number flows, not only high grid resolution is required for a flow simulation to capture very small eddies of the sound source inside the turbulent boundary layer, but also the computation of acoustic field is often needed. In such case, the direct simulation of flow field and acoustic field is straightforward and effective.

In this study, the direct simulation with the lattice Boltzmann method was conducted for a flow around the NACA0012 airfoil with the Reynolds number of two hundred thousand. In order to efficiently simulate this high Reynolds number flow with the LBM, the multi-scale approach was introduced in conjunction with the Building-cube method, while keeping the advantage of the LBM with the Cartesian mesh. At the condition with angle-of-attack of 9 degrees, a laminar separation bubble arises on the suction surface near the leading-edge and the suction boundary layer downstream of it is turbulent due to the separated-flow transition. As a result, turbulent broadband noise is generated from the boundary layer over the airfoil with the separated-flow transition. In the paper, as for prediction of such broadband noise, the computed frequency spectrum of far-field sound is validated to agree with the experimental result. In addition, through the detailed analyses of turbulent properties of the turbulent boundary layer on the suction surface, the validity of the present direct numerical simulation is demonstrated.

Commentary by Dr. Valentin Fuster
2016;():V01AT08A003. doi:10.1115/FEDSM2016-7698.

The log-layer mismatch arises when a Reynolds-averaged Navier-Stokes (RANS) model is blended with a large-eddy simulation (LES) model in a hybrid fashion. Numerous researchers have tackled this problem by simulating a turbulent channel flow. We show that the log-layer mismatch in hybrid RANS-LES can be reduced substantially by splitting the mean pressure gradient term in the wall-normal direction in a manner that keeps the mass flow rate constant. Additionally, an analysis of the wall-normal variation of the friction velocity shows a constant value is recovered in the resolved LES region different than the value at the wall. Second-order turbulence statistics agree very well with direct numerical simulation (DNS) benchmarks when scaled with the friction velocity extracted from the resolved LES region. In light of our findings, we suggest that the current convention to drive a turbulent periodic channel flow with a uniform mean pressure gradient be revisited in testing eddy-viscosity-based hybrid RANS-LES models as it appears to be the culprit behind the log-layer mismatch.

Commentary by Dr. Valentin Fuster
2016;():V01AT08A004. doi:10.1115/FEDSM2016-7714.

Savonius-style wind turbines are a class of vertical axis wind turbine usually used for off-grid applications. It appears to be promising for energy conversion because of its better self-starting capability and flexible design promises. The blades are characterized by relatively large surface, which are thin circular shape to produce large drag, which is used for power generation. Typically, the suction side of the advancing blade is submitted to strong adverse pressure gradient, causing a well known vortex shedding process, which is responsible for the wake flow. This topic has been the subject of many researches in the past decades, as it obviously depends on tip speed ratio (TSR) and directly influences the turbine efficiency. The flow on the pressure side of the blade is generally considered as fully attached and is characterized by high pressure, low velocity level that produces most of the drag used in the energy conversion. However, because of the gap between the two blades, the flow is accelerated on the pressure side of the returning blade and a thicker boundary layer develops at this side. Because of the concave curvature of the blade and the small scale of the turbine, centrifugal instabilities may occurs, depending on the flow regime that can cause natural transition on the blade. Moreover, these vortices induce different mechanisms of ejections and sweeps, causing thereby strong transverse variations of the drag coefficient, which results in the formation of hot spots near solid walls. This can leads to a rapid degradation of mechanical structures and materials fatigue. In this paper, Direct Numerical Simulations (DNS) are carried out in order to capture the flow instabilities and transition to turbulence occurring on the pressure of a conventional design Savonius wind turbine blade. Simulations are conducted with the open source code Nek5000, solving the incompressible Navier-Stokes equations with a high order, spectral element method. Because of the relatively high Reynolds numbers considered (Reξ = 90,000), the computational domain of the Savonius blade is reduced to the pressure side, whereas no turbine rotation is considered, which avoid the large scale vortex shedding that occurs on the suction side. The results suggest that Gortler vortices can occurs and cause the flow to transit to turbulence, which modify the pressure distribution and the drag force significantly.

Commentary by Dr. Valentin Fuster
2016;():V01AT08A005. doi:10.1115/FEDSM2016-7761.

Particle-resolved direct numerical simulations are performed using fictitious domain approach [1] to investigate the effect of an oscillatory flow field over a rough wall made up of a regular hexagonal pack of fixed spherical particles, in a setup similar to the experimental configuration of [2]. Turbulent flows at Reynolds numbers, Reδ = 200 and 400 (based on the Stokes-layer thickness δ) are studied. The unsteady nature of hydrodynamic forces on particles and their cross-correlations with measurable flow variables are investigated. Temporal correlations showed drag and lift to be positively correlated with a phase difference, which is approximately equal to the Taylor micro-scale related to drag/lift correlations. Spatio-temporal correlations between the flow field and particle-related quantities showed that the lift force is well correlated with the streamwise velocity fluctuations up to distances of the same order as the particle diameter, beyond which the cross correlation decays considerably. On the other hand, the pressure fluctuations are correlated and anti-correlated with the lift force in the front and aft regions of the particle, respectively, as a result of wake effects. Further statistical analyses showed that the near-bed velocity and pressure fluctuations fit poorly with Gaussian distributions. Instead, a fourth order Gram-Charlier distribution model is proposed that may have consequences on the Gaussian descriptions of sediment pick-up functions typically used in quantification of turbulent transport of sediment particles.

Commentary by Dr. Valentin Fuster
2016;():V01AT08A006. doi:10.1115/FEDSM2016-7795.

Temporally developing DNS for channel flow at Reτ = 180 and 590 are performed to understand the turbulence generation mechanism for wall bound flow transition. Simulations for Reτ = 180 were performed for initial turbulence intensities TI = 0.1%, 1%, 2.5% and 5% and for Reτ = 590 with TI = 1% and 5%. The results in the fully developed turbulent region were compared with Moser et al. (1999) DNS data to validate the predictions. The near-wall vortical structures, mean and turbulent stresses and energy spectra, and turbulent kinetic energy (TKE) and stress budget in the pre-transition, transition and turbulent regions are analyzed to understand the turbulence onset, growth and decay mechanism. Finally, molecular diffusion and pressure strain time scales in the pre-transition to turbulent regions are analyzed to evaluate the turbulence onset criteria.

Commentary by Dr. Valentin Fuster

28th Symposium on Fluid Machinery

2016;():V01AT09A001. doi:10.1115/FEDSM2016-7518.

In the previous study (AJK2015-09034 A meridian profile obtained by no restriction in optimum process), an optimum meridian profile of impeller and guidevane by no restriction was obtained. In case of no restriction, all the design parameters and specifications are variable optimum parameters. As the result, the combination of the best design parameters and specifications were selected. In optimum process, blade number, outlet impeller mid span diameter, rotational speed of impeller and specific speed were also variable optimum parameters. As the variable design parameters need to change gradually, the blade number considering as solidity is used as real number in optimum process. There are two kinds of object functions in previous case study. One object function is composed of efficiency and initiated suction specific speed. The ideal goal of efficiency is 100%. The goal of initiated suction specific speed is 1000. The best specific speed of best efficiency was 680 and 852. The other object function is composed of only efficiency. The best specific speed of best efficiency was 520 and 539. The unit of specific speed is composed of m, min−1 and m3/min. In this case study, the initial conditions of design parameters for all design specifications are born from the best one combination of optimum design parameters obtained by no restriction in the previous case study. In this method, the optimum profile for various specifications are obtained by the optimum profile of the best specific speed in condition of changing the specific speed little by little. In previous study, the suitable initial value of all design parameters for each design specifications was not able to obtain. In this case study, the two optimum meridian profiles of low and high specific speed (200 and 3000) was born from one same meridian profile obtained by no restriction in previous optimum process, that is, it become clear that all design parameters calculated by no restriction produces initial value of all design parameters for all specifications in case study of the two kind of specific speed 200 and 3000.

If the ideal goal value of efficiency is 100% and the goal value of the initiated suction specific speed is 1000, the initiated suction specific speed looks like as restriction. If the ideal goal value of the initiated suction specific speed is very large value and the efficiency is not 100% but small value, the efficiency looks like as restriction. The object function contains the efficiency and suction specific speed. The coupling constant combined efficiency and suction specific speed is important constant for the object function. As the result of this case study, one of a meridian profile obtained by no restriction in optimum process was able to become the same initial condition of the optimum method for all specifications.

Topics: Suction , Impellers , Design , Blades
Commentary by Dr. Valentin Fuster
2016;():V01AT09A002. doi:10.1115/FEDSM2016-7575.

Micro hydro turbines with capacity ranging from 5–100 kW can provide a source of renewable energy which has high potential to grow due to the sheer number of available sites with a relatively small head difference and flow rate. Such systems would complement existing conventional large-scale hydro installations.

The paper presents a study of a 30 cm diameter Kaplan hydroturbine geometry with performance optimization through Computational Fluid Dynamics (CFD) method based on Large Eddy Simulation (LES) of transient turbulence modeling. The work under this study include: selection of horizontal or vertical setup, tail water head test, draft tube length test, draft tube angle test, intake tube geometry test and blade geometry combination test. All of these geometric parameters will be tested and compared based on power output, mass flow rate and efficiency. The optimized geometry was selected for system manufacturing and experimenting by Cadens LLC in a project site in Sullivan, WI.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A003. doi:10.1115/FEDSM2016-7587.

Back blades are usually assembled on the outside surface of impeller back shroud as a sealing device in centrifugal slurry pumps. The presence of solid particles in slurry leads to an obvious problem about the abrasion of the flow components of pump. Especially, the life of sealing devices, like the back blades, the oil seals and the shaft sleeve, is only a quarter or less of other components. Hence, an important engineering significance lies in the research on the abrasion of back blades.

In this paper, a single-stage horizontal type centrifugal pump was chosen as the main study model. The 3D model of the entire flow field was meshed by hexahedral structured grids. Based on the Particle model, which is an Eulerian multiphase method, the internal two-phase flow in the centrifugal slurry pump was simulated by using ANSYS CFX software. Six optimized design cases with the variation of back blades were analyzed to study the influence of vane profile and blade number of back blades on the abrasion characteristic and sealing performance. The main conclusions obtained are as follows: the volume fraction of solid phase achieved by simulation is in good agreement with the test results; the effect of vane profile on the flow of particles in the passages and the pressure on the seal is small; the usage of less back blade number will lower the flow constraint of blades on the particles and increase the area of axial vortex in each single passage, which means that the impact velocity of particles towards the pressure side grows and the pressure on the seal increases significantly. Based on the simulation mentioned above, two better cases were selected and manufactured for trial. Then, a wear test rig was set up to study the wear pattern of impeller during the operation of pump. Through the comparison of these two impellers after the wear test, it is found that: the back blades with the back forward shape can effectively reduce the abrasion of back blades at the pressure sides near the trailing edge; thickening the trailing edge of back blades to increase the life of back blades is feasible in practical application. The optimization design of back blades was preliminarily achieved which could provide some reference for the optimization design of back blades in centrifugal slurry pumps.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A004. doi:10.1115/FEDSM2016-7588.

Flow of steam, different from other gas flows, involves droplet generation in flow expansion process. This phase transition affects not only the flow fields, but also machine performance including efficiency. In addition, it is totally harmful for machine structures as blades and casing. Therefore, prevention or preparation of droplet generation in steam flows is dreadfully important in stable machine operation.

Nowadays, Computational Fluid Dynamics (CFD) is widely used in machine design and optimization process. Thus, simulation with CFD should consider this droplet generation phenomena to predict internal flows precisely. Many studies that analyze steam condensing flow in nozzles, cascades and steam turbines were carried out. Though, the flows of wet-steam which include non-equilibrium phase-transition phenomena are still difficult to predict, especially in the 3D rotating cases as steam turbines. Therefore, more studies are required to get comparable results with experiment.

In this study, non-equilibrium wet-steam model was implemented on T-Flow to simulate realistic non-equilibrium steam condensing flow. In the cases of White cascade, characteristics of wet-steam flow were studied and pressure distributions were compared with experimental results for model validation. To use implemented wet-steam model for calculating flows in rotation, especially in steam turbines, a study of steam condensing flow in single stage steam turbine was conducted. Interaction between the stator and rotor using frozen rotor or mixing plane method in steady calculations were compared in order to find the effects of used interface on flow fields and steam condensation.

As a result, condensing flows were predicted well even in the rotating cases by using non-equilibrium wet-steam model. The wet-steam parameters (nucleation, droplet size, wetness) are differed throughout the spans due to 3D effects and influenced by selection of interface as expected. In addition, droplet generation enhances entropy rise throughout the domain. The case using mixing plane seems to be overestimate the size of high wetness zone and it is recommended to use frozen rotor in multi-phase calculations. However, to apply this model in general cases, comparison with experimental data from real steam turbines should be conducted in further studies.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A005. doi:10.1115/FEDSM2016-7621.

The axial-flow pump system research and technological innovation in China are introduced. The research trend on hydraulic performance of the axial flow pump system are discussed. The applications of axial-flow pump hydraulic model and recent progress are presented. Through inductive characteristics of axial flow pump system, the classification of the system is put forward according to the installation position of the motor namely the shaft extension type axial-flow pump system and the tubular axial-flow pump system. The innovation applying of the different types of axial-flow pump systems in a number of pumping station projects are introduced respectively. From the perspective of technological development and application prospects of axial-flow pump system are analyzed. The limitations of the traditional pump selection method are discussed, and the applicable rationality of a new method of pump selection for the axial-flow pump system, is introduced. The variable angle adjustment formula and applicability based on the test data of axial-flow pump system are introduced. The harmfulness to the pump units from the intake vortex and the safety measures are analyzed. The research results of the vortices in the pump sump and the measures of vortices prevention and elimination are described. As regarding the real and potential problems in research development on axial-flow pump system, the suggestions for further deepen researches are presented.

Topics: Pumps , Axial flow , China
Commentary by Dr. Valentin Fuster
2016;():V01AT09A006. doi:10.1115/FEDSM2016-7622.

A reduction of fan power is required to develop high efficiency air conditioner. Fan-power reduction is achieved by reducing of the pressure loss of flow channels and/or improving the efficiency of a fan. An indoor unit of the air conditioner in the present study is installed on a ceiling in a room. The indoor unit consists of a centrifugal fan, a heat exchanger, an air inlet and 4 air-outlet nozzles. One of the areas of the highest pressure-loss in the indoor unit is around the air-outlet nozzles since the cross-section of the flow channel in the air-outlet nozzles is smaller than those in other areas. In addition, the path of air flow after passing through the heat exchanger is sharply turned downward by a cabinet wall of the indoor unit. The air flow separates in the air-outlet nozzles when the air flow gets over a drain pan which receives water condensed on the surface of the heat exchanger. As a result, the effective cross-section of the air-outlet nozzles is further reduced due to the flow separation. This is main cause of the pressure-loss in the air-outlet nozzles.

The optimum nozzle shape to suppress flow separation in air-outlet nozzles of the indoor unit of an air conditioner was determined. An edge, shaped on the wall of the drain pan, minimized the flow separation by corresponding to the location between the edge and the attachment point of the flow separation. The location of the edge is defined by two parameters, and the influence of the parameters on reduction of fan power was determined by using Computational Fluid Dynamics (CFD). A CFD model of a whole indoor unit (including fan, heat exchanger, air-outlet nozzles) was applied to accurately predict fan power for different locations of the edge. Furthermore, the flow separation in the air-outlet nozzle was visualized on the basis of the CFD results. To obtain the appropriate combination of parameters to suppress the flow separation, a response surface based on the CFD results and approximate values given by a Kriging-based method, was used. The Kriging-based model is one of the response-surface methods and is characterized by approximating a nonlinear function based on Bayesian probabilistic estimation. The response surface provided the area containing the appropriate parameters for reducing fan power (“parameter area”, hereafter). The parameter area of fan-power reduction on the response surface was found. The CFD results confirm that the flow separation corresponds to the edge location given by this parameter area and that the edge minimizes the flow separation. To experimentally verify the effect of the edge on fan-power reduction, four points in the parameter area for fan-power reduction were selected, and four nozzle shapes with these parameters were prototyped. It was found that fan power was reduced (at most) by 9.9% by the optimized nozzles shapes in comparison with the current shape of air-outlet nozzles.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A007. doi:10.1115/FEDSM2016-7627.

A bionic blade with convex domes was applied in a double suction centrifugal pump to improve erosion resistance of the blade surfaces in this study. The hydraulic performance of the pump was simulated and the numerical results were in good agreement with the experiment data. The erosion rates of the smooth blade and bionic blades with convex domes at different heights (1.0 mm, 1.5 mm, 2.0 mm ) were numerically predicted. The results showed that the pump with bionic blades had a higher head and a lower efficiency than those of the pump with smooth blades. A comparison of the erosion rates indicated that the bionic blades exhibited much better erosion resistance than the smooth surface ones. The high erosion-rate area was reduced remarkably and the erosion region became more dispersed on the whole bionic blade surface. The pressure side of the blade with 2.0 mm-height convex domes showed better anti-erosion performance than those with other two heights, while the suction side with 1.0 mm-height domes showed better anti-erosion performance.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A008. doi:10.1115/FEDSM2016-7643.

In order to meet the requirement of coal mine flooding emergency rescue, a high power, high head and small volume high-speed wet submersible pump is designed. The high speed rescue pump applies the wet motor and pump integrated structure to achieve the best effect . When high speed rescue pump works, the temperature rise of the motor is high, which may cause the damage of the whole unit if the heat which produced by motor can not be taken away fully. The design of the cooling circuit is critical for the performance of the high speed rescue pump. This paper gives two design methods of the cooling circuit of high speed rescue pump. The design performance parameters: Capacity Q = 200m3/h, Head H = 50m, Rotate speed n = 6000r/min, Power P = 600kW. Two cooling circuits contains the normal and reverse one, which are based on theoretical deduction, numerical simulation and experimental verification. First and foremost, two theoretical models of cooling circuit are established by the theory of convective heat transfer .The heat balance and distribution are calculated by theoretical derivation. Then, both three-dimensional models of the circuit are built by CREO and simulated by ANSYS. The method of flow-heat coupling is used to simulate the whole inner flow field of the high speed rescue pump at different running conditions by considering the transformation of thermal performance parameter of cryogenic fluid caused by temperature change. In the simulation ,the information , such as temperature , flow field, pressure distribution of the whole cooling circuit together with temperature and velocity in the gas gap where temperature changed greatly, the convective heat transfer between fluid and motor ,and the flow rate of the cooling fluid are also gained. The analysis results show that: from the comparison of the pressure distribution of the two cooling forms, under the same inlet and outlet liquid condition ,the minimum and maximum pressure value of the reverse circuit are much higher than the corresponding value ,which means the reverse cooling method is better than the normal method as the aspect of cavitation performance. The temperature rise of reverse cooling circuit with the value 1.5K is smaller than the value of the normal cooling circuit. As the key part of the cooling circuit, the motor gas gap has a significant influence on the performance of the circuit. The velocity and temperature distribution is given to study the law of the flow and thermal field in the gap which can supply an intuitive understanding of the key part. At last, an experiment of a model pump is carried out on the test table validated the reliability of the reverse cooling circuit. It can be also concluded that the cooling circuit can satisfy with mode demand of the working condition of the high speed rescue pump.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A009. doi:10.1115/FEDSM2016-7644.

Due to the advantages of high head and no leakage, multistage canned motor pump is widely used in oil industry, chemical industry, national defense and atomic energy. In order to meet the needs of the market, the multistage canned motor pump is designed. This paper introduced the hydraulic design and structural design. In order to optimizing the performance of the pump, this paper designed and used multistage canned motor pump DBP15–50×8 as the research object. Three-dimensional model of the main flow passage components is built and the mesh is generated respectively by using Pro/E and ICEM software, and we calculated the whole internal flow field of the pump that was selected by using ANSYS CFX14.0 software, achieving the pressure and velocity distribution in the pump and the internal details of flow in impeller and other main flow components. The post-processing showed the fluid in sliding bearing section rotates around the shaft, so the local flow is disorder. The comparison of the performance prediction and the experiment shows that the error is low. The cavitating turbulent flow in the flow field was numerically simulated by using the cavitation model. The cavitation phenomena didn’t occur in the experiments. The condition meets the result of numerical simulation.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A010. doi:10.1115/FEDSM2016-7734.

Power density of a super-critical carbon dioxide cycle is very high due to its fluid-like density. For this reason, generally size of turbines are very compact compared to that of the air Brayton cycle. However, such an advantage sometimes becomes a challenge for aerodynamic design, because low volume flow rate of the turbine requires design point at a very low specific speed. One of the solution for the challenge is to design a turbine stage as a partial admission stage in which flow enters the turbine nozzle over only a portion of its annulus. Then it secures a sufficient turbine inlet area, even though performance degradation should be taken in to account.

In this study, aerodynamic design of an axial turbine has been carried out and its performance has been assessed with numerical simulations. One of design requirements for the axial turbine was to minimize rotor inlet and outlet pressure difference to avoid potential axial thrust. In spite of a small amount of expansion ratio in the turbine stage, the absolute pressure difference could cause severe damage to rotor dynamic system and require complicated bearing system. For this reason, in this study, the turbine was designed as impulse type axial turbine with partial admission.

Required rotating speed and resultant low volume flow rate restricted mean diameter and blade height at the stage inlet. The final design has a very low aspect ratio, less than unity. The number of nozzle and rotor are 12 and 34, respectively. The rotating speed of the rotor is 45,000 rpm. The ratio of nozzle arc to blade pitch is approximately 3, which determines efficiency deterioration due to the partial admission.

During the numerical simulations, to implement real gas property, Redlich-Kwong-Aungier cubic equation was used. As the turbine operating point is far from its critical point, the Redlich-Kwong-Aungier cubic equation showed a good agreement with real supercritical gas property.

To assess full and partial admission turbine performance, steady state numerical simulations have been performed. The full annulus CFD domain was constructed for the partial admission stage. At the design condition, there was 15% isentropic efficiency drop in case of the partial admission stage relative to the full admission stage. Also similar amount of power output penalty was investigated from the partial admission case. As the nozzle was choked at the design condition, the mass flow rate was conserved regardless of the admission type. Then in the flowing region, design velocity triangle in front of the rotor well established, while additional loss was generated along the circumferential direction over non flowing region.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A011. doi:10.1115/FEDSM2016-7739.

The unit of a Pumped Storage Power Station experienced abnormal noise and vibration in the guide vanes at the slight opening when the pump turbine was in the process of startup in the pumping mode. Based on this phenomena, the three dimensional model of the pump turbine was established, RNG k-epsilon two equations turbulence model was selected for the flow numerical simulation in the pump turbine because this model can simulate both the flow separation and vortex dynamics, and it is more accurate in the near wall areas. The governing equations were discretized with the finite volume method. The computation was carried out with three steps, 1.steady calculation, 2.unsteady calculation with constant guide vane opening, 3.unsteady calculations with the increase of the opening of guide vanes, by using the results of the last step as the initial condition. According to the three dimensional simulation results, the main flow between the guide vanes was deflected from attaching to the one vane to the other vane with the opening of the guide vanes. The calculation of complete 3D flow indicated that the deflections of the flows between the different adjacent guide vanes were basically the same, however, the deflections starting times had a few differences. The variation of the torque on the guide vane was also investigated, and the results shown the abrupt changes occurred during the deflection process of the main flow. When the torque produced by the servomotor cannot adapt quickly enough to the abrupt changes, the vibration and loud scrape noise might occur.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A012. doi:10.1115/FEDSM2016-7740.

To investigate the influence of relative positions between a radial diffuser and an annular volute on the unsteady pressure at the centrifugal pump outlet, experiment tests were carried out with five positions between the diffuser and volute in an open test rig. Statistical and frequency spectrum analyses were carried out to obtain the pressure fluctuation amplitude range and the frequency domain respectively. The results showed that the relative position has greater influence on the pressure at large flow rate than at part load condition. The dominant frequency and the Power Spectrum Density (PSD) values are affected by diffuser azimuthal position and the harmonic frequencies are determined by number of blades and vanes. The investigation can give a reference to optimize the relative angle between diffuser and volute to reduce the pressure fluctuations.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A013. doi:10.1115/FEDSM2016-7780.

New equations for hydraulic efficiency conversion from a model to a prototype centrifugal pump including mixed and axial flow types have been developed and proposed in this paper. In order to establish a set of conversion equations applicable for all type numbers of pumps, the following factors related to the conversion equations were examined.

1) The ratio of scalable loss to total hydraulic loss was examined by using CFD (Computational Fluid Dynamics) analysis. The ratio is related with the effect of actual complex velocity distributions in the flow passages in the impeller and diffuser/volute casing.

2) The conversion equation was constructed by two terms dealing with flow passages in two major hydrodynamic components, impeller and diffuse/volute separately, where contributions of each component to the efficiency step-up was expressed explicitly.

3) The friction coefficient ratio between a model and prototype pump was expressed in a simplified equation, applicable for both hydraulically-smooth and transitional surfaces. This expression was found to be useful to determine the relationship between equivalent machined surface roughness and uniform sand roughness, as the friction coefficient diagram expressed for uniform sand roughness is used for conversion of hydraulic losses between the model and prototype pumps.

4) The difference between the friction coefficients for decelerating flow and for a flat plate with uniform flow was examined by CFD analysis, and it was found that the friction coefficient for a flat plate can be used for the conversion without causing any substantial deviation.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A014. doi:10.1115/FEDSM2016-7816.

When a turbocharged engine which was designed to work at sea level works at high altitude area, its power will decrease because of the decrease of the air density. And the expansion ratio of the turbine will increase a lot because of the decrease of the out pressure, leading to a bad efficiency of the turbine. In order to recover the power of the turbocharged engine when it worked at the high altitude area, the efficiency of the turbine should be increased when it worked at high expansion ratio condition.

This paper shows a numerical investigation of the flow fields in a radial inflow turbine at design and off design condition. The comparison of the flow characteristics between the design and off design conditions is researched. When the turbine work at a high expansion ratio, there will occur a supersonic zone at the leading edge on the suction side and a shock wave will occur at the trailing edge, causing a flow separation and making a lot of losses, leading the decrease of the turbine efficiency.

Topics: Design , Turbines
Commentary by Dr. Valentin Fuster
2016;():V01AT09A015. doi:10.1115/FEDSM2016-7939.

An axial-flow pump has a relatively high discharge flow rate and specific speed at a relatively low head and it consists of an inlet guide vane, impeller, and outlet guide vane. The interaction of the flow through the inlet guide vane, impeller, and outlet guide vane of the axial-flow pump has a significant effect on its performance. Of those components, the guide vanes especially can improve the head and efficiency of the pump by transforming the kinetic energy of the rotating flow, which has a tangential velocity component, into pressure energy. Accordingly, the geometric configurations of the guide vanes such as blade thickness and angle are crucial design factors for determining the performance of the axial-flow pump. As the reliability of Computational Fluid Dynamics (CFD) has been elevated together with the advance in computer technology, numerical analysis using CFD has recently become an alternative to empirical experiment due to its high reliability to measure the flow field. Thus, in this study, 1,200mm axial-flow pump having an inlet guide vane and impeller with 4 blades and an outlet guide vane with 6 blades was numerically investigated. Numerical study was conducted using the commercial CFD code, ANSYS CFX ver. 16.1, in order to elucidate the effect of the thickness and angle of the guide vanes on the performance of 1,200mm axial-flow pump. The stage condition, which averages the fluxes between interfaces and is accordingly appropriate for the evaluation of pump performance, was adopted as the interface condition between the guide vanes and the impeller. The rotational periodicity condition was used in order to enable a simplified geometry to be used since the guide vanes feature multiple identical regions. The shear stress transport (SST) k-ω model, predicting the turbulence within the flow in good agreement, was also employed in the CFD calculation. With regard to the numerical simulation results, the characteristics of the pressure distribution were discussed in detail. The pump performance, which will determine how well an axial-flow pump will work in terms of its efficiency and head, was also discussed in detail, leading to the conclusion on the optimal blade thickness and angle for the improvement of the performance. In addition, the total pressure loss coefficient was considered in order to investigate the loss within the flow paths depending on the thickness and angle variations. The results presented in this study may give guidelines to the numerical analysis of the axial-flow pump and the investigation of the performance for further optimal design of the axial-flow pump.

Commentary by Dr. Valentin Fuster
2016;():V01AT09A016. doi:10.1115/FEDSM2016-7949.

Leading edge protuberance modifications on airfoils or wings have attracted extensive attentions as a new passive technique for separation control. In this paper, the hydrodynamic performance of a NACA 634-021 foil and a modified foil with leading-edge protuberances were numerically investigated using Spalart-Allmaras turbulence model. Compared to the sharp decline of baseline lift coefficient, the stall angle of the modified foils was advanced and the decline of lift coefficient became mild, and the post-stall performance was improved. A special bi-periodic flow pattern may occur and stay extremely steady at a wide range of attack angles, accompanied with a relatively steady lift. The transformation from single-periodicity to bi-periodicity occurred within a small range of range of attack angle. A couple of counter-rotating streamwise vortex was formed on the shoulder of each protuberance, altering the vorticity line to share a similar shape as the leading-edge profile. At larger angles of attack, the development of streamwise vortex would be accompanied with transformation to lateral vortex, where strong interaction may happen and give rise to the occurrence of bi-periodic condition. The formation mechanism and control method of the special phenomenon should be investigated more deeply in the future.

Topics: Airfoils
Commentary by Dr. Valentin Fuster

17th International Symposium on Fluid-Structure Interaction and Flow-Induced Noise in Industrial Applications

2016;():V01AT11A001. doi:10.1115/FEDSM2016-7506.

This contribution is the second part of three papers on Adaptive Multigrid Methods for the eXtended Fluid-Structure Interaction (eXFSI) Problem, where we introduce a monolithic variational formulation and solution techniques. To the best of our knowledge, such a model is new in the literature. This model is used to design an on-line structural health monitoring (SHM) system in order to determine the coupled acoustic and elastic wave propagation in moving domains and optimum locations for SHM sensors. In a monolithic nonlinear fluid-structure interaction (FSI), the fluid and structure models are formulated in different coordinate systems. This makes the FSI setup of a common variational description difficult and challenging. This article presents the state-of-the-art in the finite element approximation of FSI problem based on monolithic variational formulation in the well-established arbitrary Lagrangian Eulerian (ALE) framework. This research focuses on the newly developed mathematical model of a new FSI problem, which is referred to as extended Fluid-Structure Interaction (eXFSI) problem in the ALE framework. The eXFSI is a strongly coupled problem of typical FSI with a coupled wave propagation problem on the fluid-solid interface (WpFSI). The WpFSI is a strongly coupled problem of acoustic and elastic wave equations, where wave propagation problems automatically adopts the boundary conditions from the FSI problem at each time step. The ALE approach provides a simple but powerful procedure to couple solid deformations with fluid flows by a monolithic solution algorithm. In such a setting, the fluid problems are transformed to a fixed reference configuration by the ALE mapping. The goal of this work is the development of concepts for the efficient numerical solution of eXFSI problem, the analysis of various fluid-solid mesh motion techniques and comparison of different second-order time-stepping schemes. This work consists of the investigation of different time stepping scheme formulations for a nonlinear FSI problem coupling the acoustic/elastic wave propagation on the fluid-structure interface. Temporal discretization is based on finite differences and is formulated as a one step-θ scheme, from which we can consider the following particular cases: the implicit Euler, Crank-Nicolson, shifted Crank-Nicolson and the Fractional-Step-θ schemes. The nonlinear problem is solved with a Newton-like method where the discretization is done with a Galerkin finite element scheme. The implementation is accomplished via the software library package DOpElib based on the deal.II finite element library for the computation of different eXFSI configurations.

Commentary by Dr. Valentin Fuster
2016;():V01AT11A002. doi:10.1115/FEDSM2016-7668.

Efficiency of different types of immersed boundary methods in the fluid structure interaction (FSI) analysis is studied for different cases. Two different formulations of smoothed profile method (SPM) [1, 2] as diffuse interface approaches are compared with the ghost fluid method (GFM) [3, 4] as sharp interface method (SIM) [5]. First, the original SPM which has two pressure Poisson equations (SPM2P) is modified to a novel formulation for SPM with only one pressure Poisson equation (SPM1P) and then validated. The efficiency study is performed for SPM1P, SPM2P and SIM. The results show that when the solid object is fixed, the explicit solution of SIM is faster than the two SPMs. However, when the solid is moving and strongly coupled formulations is used, SPM1P will be the fastest method. It is shown that the efficiency of the strongly coupled formulations depends on the number of subiterations required in each time step to reach the converged implicit solution. SPM1P and SPM2P need less number of subiterations in comparison with SIM and they are faster. When the added mass effect is high, the efficiency of SPM becomes more noticeable as the required number of subiterations is significantly less in SPM. Finally, SPM1P is faster than SPM2P in all cases however, the accuracy of SPM2P in predicting the flow pattern is better than SPM1P.

Commentary by Dr. Valentin Fuster
2016;():V01AT11A003. doi:10.1115/FEDSM2016-7861.

In this work, we present an approach for solving fluid structure interaction problems by combining a non-linear structure solver with an incompressible fluid solver using immersed boundary method. The implementation of the sharp-interface immersed boundary method with the fluid solver is described. A structure solver with the ability to handle geometric nonlinearly is developed and tested with benchmark cases. The partitioned fluid-structure coupling algorithm with the strategy of enforcing boundary conditions at the fluid/structure interaction is given in detail. The fully coupled FSI approach is tested with the Turek and Hron fluid-structure interaction benchmark case. Both strong coupling and weak coupling algorithms are examined. Predictions from the current approach show good agreement with the results reported by other researchers.

Commentary by Dr. Valentin Fuster
2016;():V01AT11A004. doi:10.1115/FEDSM2016-7868.

The drill bit blaster (DBB) studied in this paper aims to maximize the drilling rate of penetration (ROP) by using a flow interrupting mechanism to create drilling fluid pulsation. The fluctuating fluid pressure gradient generated during operation of the DBB could lead to more efficient bit cutting efficiency due to substrate depressurization and increased cutting removal efficiency and the vibrations created could reduce the drill string friction allowing a greater weight on bit (WOB) to be achieved. In order to maximize these mechanisms the effect of several different DBB design changes and operating conditions was studied in above ground testing. An analytical model was created to predict the influence of various aspects of the drill bit blaster design, operating conditions and fluid properties on the bit pressure characteristics and compared against experimental results. The results indicate that internal tool design has a significant effect on the pulsation frequency and amplitude, which can be accurately modeled as a function of flowrate and internal geometry. Using this model an optimization study was conducted to determine the sensitivity of the fluid pulsation power on various design and operating conditions. Application of this technology in future designs could allow the bit pressure oscillation frequency and amplitude to be optimized with regard to the lithology of the formations being drilled which could lead to faster, more efficient drilling potentially cutting drilling costs and leading to a larger number of oil and natural gas plays being profitable.

Topics: Bits (Tools)
Commentary by Dr. Valentin Fuster

10th International Symposium on Flow Applications in Aerospace

2016;():V01AT12A001. doi:10.1115/FEDSM2016-7573.

An overview of twenty years of adjoint-based aerodynamic design research at NASA Langley Research Center is presented. Adjoint-based algorithms provide a powerful tool for efficient sensitivity analysis of complex large-scale computational fluid dynamics (CFD) simulations. Unlike alternative approaches for which computational expense generally scales with the number of design parameters, adjoint techniques yield sensitivity derivatives of a simulation output with respect to all input parameters at the cost of a single additional simulation. With modern large-scale CFD applications often requiring millions of compute hours for a single analysis, the efficiency afforded by adjoint methods is critical in realizing a computationally tractable design optimization capability for such applications.

Commentary by Dr. Valentin Fuster
2016;():V01AT12A002. doi:10.1115/FEDSM2016-7603.

Droplet impingement is the basic module in both ice accretion and anti-icing numerical calculation. A three dimensional finite volume approach with the capacity of modeling the in-flight droplet impingement on a wide range of subsonic regime is therefore established in this research, using OpenFOAM®. The Eulerian model is applied to estimate the droplet flow field with the same computational grid sets as those of the air flow calculation. The roughness effect caused by ice accretion is considered in the wall function modeling. Thus, the collection efficiency could be investigated for further icing numerical simulations. This approach is validated on both cylinder and sphere benchmark cases. The results are compared with the corresponding experimental and LEWICE (LEWis ICE accretion program) simulation data.

Topics: Drops , Ice , Flight
Commentary by Dr. Valentin Fuster
2016;():V01AT12A003. doi:10.1115/FEDSM2016-7685.

The flowfield and pressure distribution of a rotor blade in reverse flow is studied using stereoscopic particle image velocimetry. The 2-bladed teetering rotor with rigid NACA0013 untwisted untapered blades, with manually set collective and cyclic pitch, is operated at advance ratios from 0.7 to 1.0. Results are presented from azimuths 240 and 270 degrees, where the velocity field in chordwise sectional planes at two radial stations are analyzed, at two advance ratios. The paper is focused on two aspects. First is calculation of the total circulation around each blade section, and around the strong sharp-edge vortex seen below the blade in these sections. The second is the surface and flowfield pressures derived from the 3-component velocity field obtained from closely-spaced planes, after interpolation to satisfy the mass continuity equation. The pressure extraction technique is being developed using a yawed-cylinder test case, and shows good success in satisfying continuity between data planes.

Commentary by Dr. Valentin Fuster
2016;():V01AT12A004. doi:10.1115/FEDSM2016-7688.

Narrowband excitation of fin buffeting is known to exist on several modern aircraft configurations at high angles of attack. For a fixed angle of attack and model geometry, narrowband peak frequency is a linear function of freestream speed. Under these conditions, counter-rotating vortex pairs conforming to the Goertler vortex mechanism are known to develop and amplify in the flowfield over the wings. This phenomenon is explored for relevance to reverse flow over rotor blades at high speeds as well. A 42-degree delta wing with rounded leading edges is used in a low-speed wind tunnel to confirm the phenomenon. The presence of a non-zero yaw angle can increase the strength of the Goertler vortices and also change the location of maximum intensity. Small fences on the surface have been shown to eliminate these narrow-band fluctuations. Dielectric Barrier Discharge plasma actuators offer a possible means to eliminate the narrowband excitation without obtrusive surface fences. An array of such actuators is used to generate counter-rotating vortices. Incense smoke entrained into the flow is illuminated with a laser sheet from a laser pointer. Video images are used to capture velocity in the potential flow region around the vortices. The induced velocity is used to calculate vortex strength. Scaling laws are used to estimate the frequency of the actuators, as well as the magnitude of the velocity. The scaling estimation shows that a plasma actuator is viable for model-scale configurations. Continuing experiments for the final paper plan to apply the actuator under delta wing high angle of attack operation.

Topics: Actuators , Vortices , Wings
Commentary by Dr. Valentin Fuster
2016;():V01AT12A005. doi:10.1115/FEDSM2016-7741.

The vast majority of bird scale ornithopters still utilize single active degree of freedom wings in which the flapping motion is actuated at the root of the wing. Yet, as we look to nature, we see that birds utilize more than one active degree of freedom. The purpose of this study is to determine the effect of dynamic wing twist and wing folding on lift and thrust produced by a flapping wing as well as their effects on power consumption. The method of analysis this study utilizes is a version of MST, a Modified Strip Theory, in order to model the aerodynamics of the wing. Both non-folding and folding wing scenarios are considered where the parameters varied include dynamic wing twist amplitude, time averaged wing twist, and dynamic wing twist and flapping phase offset. Furthermore, unlike many other theoretical studies, when examining power consumption both the aerodynamic force as well as inertial effects are considered as inertial effects can be of the same order as aerodynamic force. Moreover, the negative power occurring on the upstroke cannot be always considered to lead to energy transfer back into the system as many studies assume. Thus, this study discusse the impact of negative power and its implications on ornithopter design.

Commentary by Dr. Valentin Fuster
2016;():V01AT12A006. doi:10.1115/FEDSM2016-7742.

Computational fluid dynamics (CFD) was used to investigate the fluid mechanics for undulatory stingray locomotion. This method of undulatory propulsion can be utilized to generate non-turbulent thrust with minimal disturbance to the immediate environment, ideal for exploratory vehicles for underwater environments. Undulatory locomotion was modeled as a two-dimensional fin in free flow with a deforming non-slip boundary to represent a propagating sinusoidal wave with a linearly increasing amplitude, constant frequency, wavelength and flow velocity. In the presented computational study, we varied the amplitude, wavelength, frequency, and flow velocity parametrically and examined the effect on thrust, lift, and pitching moment. Average net thrust was found to increase with wavelength and frequency, whereas for this two-dimensional case amplitude showed negligible effects. For the parametric cases, a theoretical efficiency for forward propulsion was then calculated for a continuous fin. The amplitude was found to increase the input power required for actuation, but decreased output power for forward thrust. Variation of the other parameters showed that the output power depends nearly linearly on the input power, regardless of the particular kinematics or swimming speed.

Commentary by Dr. Valentin Fuster
2016;():V01AT12A007. doi:10.1115/FEDSM2016-7782.

Insects, sustaining flight at low Reynolds numbers (500<Re<10,000), fly utilizing mechanically simple kinematics (3 degrees of freedom) at an extremely high flap frequency (150–200 Hz), resulting in a complicated vortical fluid field. These flight characteristics result in some of the most agile and maneuverable flight capabilities in the animal kingdom and are considered to be far superior to fixed wing flight, such as aircraft. Bees are of particular interest because of the utilization of humuli to attach their front and hind wings together during flight. A Cartesian-based adaptive meshing implementation of the Lattice-Boltzmann Method is utilized to resolve the complex flow field generated during insect flight and is verified against experimental and computational results present in the literature in two dimensions. The Lattice-Boltzmann Method was found to agree well in both qualitative and quantitative comparisons with both two-dimensional computational and three-dimensional experimental results.

Commentary by Dr. Valentin Fuster

11th Symposium on Active Fluid Dynamics and Flow Control — Theory, Experiments and Implementation

2016;():V01AT13A001. doi:10.1115/FEDSM2016-7508.

In this study, we have examined slot location and velocity ratio of a tangential co-flow jet in dynamic stall motion of an airfoil at Reynolds number 1×106, for active flow control (AFC) purposes. The airfoil is the symmetrical NACA 0012 with a pitching motion between AOAs 5 deg. and 25 deg. about its quarter-chord with a sinusoidal motion. We have utilized Computational Fluid Dynamics (CFD) tool to numerically investigate the impact of jet location and jet velocity ratio on the aerodynamic coefficients. We have placed the jet location upstream the counter clock-wise (CCW) vortex which is formed during the upstroke motion near the leading-edge. We have also selected several other locations nearby to perform sensitivity analysis. Our results showed that placing the jet slot within a very small range upstream the CCW vortex has tremendous effects on both lift and drag, such that maximum drag which occurs at maximum incidences reduced by 80%. There was another unique observation: putting jet at separation point leads to an inverse behavior of drag hysteresis curve in upstroke and downstroke motions. Drag in downstroke motion is significantly lower than upstroke motion, whereas in uncontrolled case the converse is true. In addition, by implementing jet flow lift is significantly enhanced during both upstroke and downstroke motions. Finally, it should be indicated that this study provides initial steps in investigations of applying synthetic jet actuator (SJA) on a pitching airfoil at high Reynolds number 106 with effects of changing momentum ratio and SJA frequency, which will be presented in the near future.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A002. doi:10.1115/FEDSM2016-7520.

Numerical simulation experiments on vortex shedding and corresponding drag coefficients from a two-dimensional bluff body are performed over a range of Reynolds numbers from one to four million. Active control is implemented on the body via velocity boundary conditions in the form of blowing and suction jets. These controls range in velocity from half to double the free-stream inlet velocity. An overall drag coefficient reduction in excess of 75% is observed for maximum power input to the actuators. In addition, a trend of increasing Strouhal number for each successive increase in actuator power (and corresponding reduction in drag) is noted. Important physical mechanisms involving near-body wake flow are analyzed to determine optimal wake flow pattern and corresponding control schemes.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A003. doi:10.1115/FEDSM2016-7526.

Cyanobacteria are photosynthetic micro-organisms colonizing all aquatic and terrestrial environments. The motility of such living micro-organisms should make their diffusion distinct from typical Brownian motion. This diffusion can be investigated in terms of global behavior (Fickian or not) and in terms of displacement probabilities, which provide more detail about the motility process. Using cyanobacterium Synechocystis sp. PCC 6803 as the model micro-organism, we carry out time-lapse video microscopy to track and analyze the bacteria’s trajectories, from which we compute the mean-squared displacement (MSD) and the distribution function of displacement probabilities.

We find that the motility of Synechocystis sp. PCC 6803 is intermittent: high-motility “run” phases are separated by low-motility “tumble” phases corresponding to trapped states. However, this intermittent motility leads to a Fickian diffusive behavior, as shown by the evolution of the MSD with time.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A004. doi:10.1115/FEDSM2016-7532.

The interest in the active flow control based on dielectric barrier discharge (DBD) plasma actuators has increased rapidly in the past decade. Because of its features such as light weight, low power consumption, fast response and flexibility, the DBD plasma actuator is a promising technology in advancing the aerodynamic performance and maneuvering of unmanned aerial vehicles. In this study, DBD plasma actuators are employed on a full span delta wing with a 75 degree swept angle to control the leading edge vortices (LEV), which generate the vortex lift on the delta wing. The experiment is conducted in a low speed closed-loop wind tunnel and the Reynolds number based on the delta wing chord is 50,000. To fix the stagnation points, both leading edges are beveled on the windward sides at an angle of 35 degrees and actuators are insulated at the leading edges. These actuators are driven independently at a frequency of 20 kHz and a voltage of 12 kV in both continuous mode and periodic mode. The DBD actuators are calibrated using a pitot tube. Smoke flow visualization result indicates that the breakdown points of leading edge vortices can be significantly affected by DBD plasma actuators at the leading edge. In the asymmetric control case (only an actuator on one side is powered), the breakdown point of the LEV on the controlled side is greatly advanced while the one on the uncontrolled side is delayed; in the symmetric control case (actuators on both sides are powered), the control shifted the breakdown points of both LEVs further downstream. Particle image velocimetry (PIV) demonstrates clearly that the control caused by DBD actuators at the leading edge can influence the separation at the leading edges and also the shear layer vortices, which form the substructures around the primary vortices. As a result, breakdown points of LEVs are affected. Interestingly observed, the control leads to a contrary flow phenomenon: in the asymmetric case, the breakdown point of the LEV on the controlled side is advanced while, in the symmetric control case, the breakdown points of the LEV on both sides are delayed. The effects of reduced frequency and duty cycle on the control authority are also investigated experimentally. Control efficiencies of both continuous mode and periodic mode are discussed.

Topics: Actuators , Vortices , Wings
Commentary by Dr. Valentin Fuster
2016;():V01AT13A005. doi:10.1115/FEDSM2016-7583.

Active flow separation control using dielectric barrier discharge (DBD) plasma actuators oriented in the spanwise direction has been successfully investigated by the authors using an integrated numerical simulation tool that couples the unsteady Reynolds averaged Navier-Stokes (URANS) or large eddy simulation (LES) solver for incompressible flows with the DBD electro-hydrodynamic (EHD) body force model. Although many experimental and numerical investigations have indicated that the spanwise-oriented DBD plasma actuator is an effective flow control method, the application is difficult to extend from model-scale to full-scale problems, partly due to the required high amplitude and high bandwidth excitation. Also, the flow control mechanism associated with a spanwise-oriented DBD actuator is mainly direct momentum injection, therefore, the effectiveness of actuation is sensitive to the location of the DBD actuator relative to the location of flow separation. On the other hand, a few experimental studies have shown promising results using the DBD Vortex Generator (DBD-VG) consisting of multiple plasma DBD actuators oriented in the streamwise direction. By generating streamwise vortices extending a long distance downstream, the DBD-VGs enhance the mixing of the inner and outer layers of turbulent boundary layer flows. As a result, the boundary layer can better withstand an adverse pressure gradient. When applied to flow separation control, the effectiveness of the DBD-VGs should be less sensitive to location of flow separation.

The present work extends the capability of the integrated numerical simulation tool from a single spanwise-oriented DBD plasma actuator to multiple DBD plasma actuators oriented in any direction, including the streamwise direction. As a demonstration of the new capability in the DBD-URANS coupled solver, numerical simulations of flow induced by a DBD-VG actuator with an array of exposed electrodes in a quiescent environment, as well as in a turbulent boundary layer over a flat plate, are carried out. The numerical simulation successfully reproduced the longitudinal vortices embedded in the boundary layer.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A006. doi:10.1115/FEDSM2016-7590.

The effects of hydrodynamic shear stress on the growth rate of cyanobacteria Synechocystis sp. and Chlamydomonas reinhardtii microalgae cells were studied in agitated photobioreactors, since they have different motility rates and sizes. An experimental setup was designed and constructed to monitor the growth rate of the micro-organisms versus the shear rate; experiments were carried out in a well controlled environment, under constant atmospheric pressure and 20 °C temperature. Digitally controlled magnetic agitator-photobioreactors were placed inside a closed chamber with air flow for 4 weeks, under a uniform full-time light intensity provided by two 6-watt white fluorescent light sources.

To study the effects of shear stress produced by mechanical agitation on the growth rate of a micro-organism, different agitation frequencies were tested. All reactors were filled with 150 ml of culture medium and micro-organism suspension, with initial dilution factors (mlsuspenion/mltotal volume) of 1/30 and 1/300 for Synechocystis and C. reinhardtii respectively. The vessels were placed on different agitating systems at the desired agitator rotation speed, and were sealed with a cotton membrane from the top in order to permit air exchange with the external environment. The micro-organisms’ growth was monitored daily by measuring the optical density of the suspensions using a spectrophotometer and was then correlated with the cellular concentration, which was measured in turn using a microscopic cell counter. Throughout the experiments pH levels and temperature were measured regularly and adjusted to 7 and 20 °C respectively in order to maintain the photosynthetic activity of the species.

In addition, to measure the shear stress inside the agitated reactors, a mathematical model was derived to determine the global shear stress magnitude. To determine the local shear stress distribution, the velocity field in the reactor was measured for different agitation frequencies using PIV. Different zones of high and low shear stress were identified.

The results showed that the growth rate is independent of the shear stress magnitude for Synechocystis; Synechocystis showed strong resistance, unlike C. reinhardtii, which showed linear dependence of growth rate and shear stress.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A007. doi:10.1115/FEDSM2016-7628.

This work describes the use of a synthetic jet (SJ) array for mild control of flow separation over a straight wing model. Experiments were performed in a subsonic wind tunnel to show improvement of the wing aerodynamic performance. A tomographic particle image velocimetry system was used to measure and analyze the three-dimensional flow-field with and without the SJ actuation. It was observed that, although the SJ array is relatively weak, it can still made impacts on the separated flow. After the SJ actuation, the large-scale vortex structures in the shear layer were broken into small discrete structures and the near-wall flow was substantially improved. Subsequently, Proper Orthogonal Decomposition (POD) analysis was also conducted and the effectiveness of the present SJ array was further discussed.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A008. doi:10.1115/FEDSM2016-7647.

Several recent studies have examined the fundamental behavior of synthetic jets as substitutes for continuous jets [1–4]. In addition, attempts to control fluid machines with synthetic jets have also begun [5, 6]. However, little attention has been given to the effects of an asymmetric flow field on the behavior of synthetic jets [7, 8]. There have been few reports on the influence of a large-scale asymmetric boundary on the motion of synthetic jets [7]. In this study, an attempt was made to describe the flow around a rectangular cylinder with an asymmetric slot for synthetic jets. The main results are as follows: (1) the continuous jets proceed toward the nearest rigid wall by the Coanda effect independently of the slot geometry (with/without beak), (2) when the nondimensional stroke of the synthetic jets is large, the flow patterns are similar to those of continuous jets, (3) when the nondimensional stroke is small and the rectangular cylinder is placed near the wall opposite the side of the beak, the synthetic jets turn to the upstream direction under the present range of conditions, and (4) the flow rate in the duct depends on the non-dimensional stroke length when the rectangular cylinder is placed near the wall opposite the side of the beak.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A009. doi:10.1115/FEDSM2016-7655.

A numerical study of separation control has been made to investigate aerodynamic characteristics of a NACA0012 airfoil with a tangential synthetic jet. Simulations are carried out at the chord Reynolds number of Re=1,000,000. The present approach relies on solving the Unsteady Reynolds-Averaged Navier–Stokes (URANS) equations. The turbulence model used in the present computation is the K-ω SST equations. All computations are performed with a finite volume based code. We have varied the synthetic jet position on the suction side of the airfoil at various locations from 4% of the chord all the way up to 60% of the airfoil chord. The jet oscillating frequency of fj = 15 Hz, (which corresponds to the non-dimensional oscillating frequency of Display FormulaFjet+ = 1 when the jet is placed at the 12% chord location), and the blowing ratio of Vj/U = 2 are used during the control cycle. All the cases considered here are for the airfoil at the constant angle of attack of α = 19°, where the airfoil stalls in the uncontrolled base flow. We found that stall characteristics are significantly improved by controlling the formation of separation vortices in the flow. The airfoil lift is more than doubled by placing the tangential synthetic jet anywhere between 20% chord to 50% chord location. This corresponds to a 25% improvement over the best cases reported by Chapin and Benard (2015) for a cross flow synthetic jet.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A010. doi:10.1115/FEDSM2016-7817.

Large-eddy simulation and linear stability analysis were performed on a NACA 0025 airfoil at a chord Reynolds number of 105 and four angles of attack. The computations showed that the initial vortex roll-up quickly breaks down to three-dimensional turbulence. Flow separation was observed at all angles, whereas only the lowest angle of attack formed a laminar separation bubble due to flow transition occuring close to the airfoil surface. A Chebyshev collocation method was employed to solve the viscous and inviscid stability equations. Linear stability analysis demonstrated that high-frequency disturbances occur in the laminar separation bubble case, whereas lower frequencies are present for the fully separated angles of attack. The maximum disturbance growth rates were dampened with the addition of viscosity but negligible change in peak frequency was noted.

Topics: Stability , Airfoils
Commentary by Dr. Valentin Fuster
2016;():V01AT13A011. doi:10.1115/FEDSM2016-7847.

The ability of active flow control through synthetic jet actuator to reattach the stalled flow on a NACA 0025 at low Reynolds number was investigated experimentally. Wind tunnel tests were performed where boundary layer velocity profiles were measured by hot-wire anemometry. The effects of excitation frequency and momentum input, quantified by the blowing ratio were the focus of the measurements. Using both low- and high-frequency actuation, the boundary layer was reattached in a time-averaged sense, however the mechanism of control was not the same. Low-frequency excitation lead to oscillations in the boundary layer at the control frequency due to the passage of a vortex transporting high-momentum to the inner boundary layer. Conversely, high-frequency control caused a steady reattachment of the boundary layer.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A012. doi:10.1115/FEDSM2016-7915.

The Lattice Boltzmann method is actually considered as one of the simplest approach. The flow and heat transfer distribution in a duct containing a backward and an obstacle are studied for different geometric and physical parameters.

The LBM is applied to solve the backward-facing step flow problems for an expansion ratio H/h = 2 in rectangular duct and to determine the effect of the obstacle on flow and heat transfer distribution. The obstacle is situated in the bottom wall of the duct. The effect of various obstacle lengths (0.5<h/S<1.5) will be also considered. All this results were observed for a Prandtl number of 0.71 and different range of Reynolds number (1–200).

This study showed the instabilities of velocity, recirculation length, temperature and Nusselt number with the obstacle downstream the step. These results were compared to with published experimental.

Commentary by Dr. Valentin Fuster
2016;():V01AT13A013. doi:10.1115/FEDSM2016-7918.

Couette-Taylor-Poiseuille flow CTPF consists on the superposition of Couette-Taylor flow to an axial flow. The CTPF flow hydrodynamics studies remain rather qualitative or numerical or are restricted to relatively low Taylor and/or axial Reynolds numbers. For more comprehensive and control of CTPF, especially for relatively high Taylor numbers and high axial Reynolds numbers, we investigated experimentally CTF with and without an axial flow, using the electro-diffusion ED method. This technique requires the use of Electro-Diffusion ED probe which allows the determination of the local mass transfer rate from the Limiting Diffusion current measurement delivered by the ED probe while it is polarized by a polarization voltage. From the local mass transfer (the Sherwood number), we determined the wall shear rate using different approaches. The results illustrate that low axial flow can generate a stabilizing effect on the CT flow. The time-evolutions of the local mass transfer and the wall shear rate are periodic. These evolutions characterize the waviness or the stretching of the vortices. However, Taylor Wavy Vortex Flow TWVF is destabilized under the effect of relatively important axial flow. The time-evolutions of wall shear rate are no longer periodic. Indeed, Taylor vortices are overlapped or completely destructed.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2016;():V01AT13A014. doi:10.1115/FEDSM2016-7927.

The effect of plasma flow control on reducing aerodynamic drag for ground vehicles is investigated. The experiments were carried out for a simplified ground vehicle using single dielectric barrier discharge (SDBD) plasma actuators. The plasma actuators were designed to alter the flow structure in the wake region behind the vehicle. The Ahmed body was modified to allow eight different vehicle geometries (with backlight or slant angles of 0° and 35°). Each of these were further modified by rounding the edges with different radii. Flow visualizations such as particle streams and surface oil were used to quantify features of the local flow field. The drag on the models was measured using a force balance as well as by integrating the mean velocity profiles in the model wakes. The results indicated that flow modifications needed to be applied symmetrically (upper to lower and/or side to side). This was demonstrated with the 0° backlight angle (square-back) that had all four side-corners rounded. Plasma actuators were applied to all four of the rounded edges to enhance the ability to direct the flow into the wake. Wake measurements showed that steady actuation at a fixed actuator voltage reduced the drag by an average of 20% at the lower velocities (below 15 m/s) and by 3% at the highest velocity tested (20 m/s). Model constraints prevented increasing the plasma actuator voltage that was needed to maintain the higher drag reduction observed at the lower speeds.

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