Session F13: Computational Fluid Dynamics: Applications (3:55pm - 4:40pm CST)

CFD verification and validation of added resistance in oblique waves

Today, almost all ship hulls are optimized to sail in calm water. In a realistic seagoing condition, the ship will be exposed to added resistance caused by the waves, which will influence the ships performance and fuel consumption. Physical towing tank testing of a broad spectrum of wavelengths, wave heights and headings is extremely expensive. CFD simulations offer the opportunity to include the performance in waves into the ship design phase. However, before trusting CFD results, it is very important to verify and validate the CFD model to ensure accuracy. The present study both verifies and validates the CFD model with towing tank results from the literature. The validated CFD model is used to study the added resistance in oblique waves with varying wavelength. Added resistance in head waves has been studies intensively, but only very few studies are conducted in oblique waves. The present study includes both a container ship and a tanker. The conducted verification and validation study shows that the discrepancy between the CFD results and towing tank results are within the towing tank uncertainty for most of the conditions.   

Direct numerical simulations of turbulent liquid jets

The breakup of an interface into a cascade of droplets and their subsequent coalescence is a generic problem of central importance to a large number of industrial settings such as mixing, separations, and combustion. Therefore, it is unsurprising that the breakup of liquid jets during injection (i.e. atomisation) has received great scientific interest, and this is the focus of this study. We use a hybrid interface-tracking/level-set method to treat the surface tension forces of the Navier-Stokes equations in a three-dimensional Cartesian domain. A turbulent water jet is gradually injected through a cylindrical nozzle into the computational domain, which is filled with a stagnant viscous oil. The atomisation induces the formation of outer and inner lobes which film thickness reduces overtime to generate holes. Those holes expand radially driven by surface tension to form ligaments, and subsequently droplets. The formation of the lobes can be explained though through its vortex-surface interaction. A full parametric study is presented, and the relevant mechanisms underlying the flow phenomena are elucidated.   

Benchmarking of CFD solvers for the simulation of two-phase jets

Multiphase flows have received significant interest due to their occurrence in a multitude of natural and industrial applications. In this study, we perform the benchmarking of solvers for the simulation of two-phase jets. The Navier-Stokes equation are solved under the assumption of the single-fluid formulation and therefore, the interfacial location must be also determined as part of the solution. We evaluate solvers based on both interface-capturing methods (such as Basilisk and OpenFOAM) and interface-tracking methods. We consider the scenario of a water drop formation from a cylindrical nozzle and the accuracy of the different solvers is compared with experimental results, in terms of jet-shape and drop-size. A thorough study in terms of the computational requirements (e.g., grid resolution) and the computational cost is also performed.   

Large eddy simulation of a cross-flow fan and its control

The flow characteristics of a cross-flow fan are investigated using large eddy simulation. The Reynolds number based on the blade chord length and the tip velocity at the outer radius of the fan is 5300. While the fan rotates, an eccentric vortex and a recirculation region are observed near the stabilizer and the rear guide, respectively. A through-flow region exists in between the eccentric vortex and recirculation region. To ana-lyze how the flow structures around the cross-flow fan affects the fan performance, the total pressure along the streamlines are examined. The flow separation and vortex shedding observed at the upstream region of the fan result in a considerable loss. To control this loss, a sinusoidal protrusion is applied to the leading edge of each blade. As a result, flow separation is reduced and the efficiency is increased.   

Using Machine-Learning to Speed-Up Optimisation in CFD: designing a micromixer.

In fluid dynamics applications, the use of Computational Fluid Dynamics (CFD) for design optimization is the recommended approach par excellence. Although Reynolds Averaged Navier Stokes simulations in CFD represent a lower cost option for prototyping, the number of computational simulations to achieve an optimal design may be time consuming. This process may speed-up by means of Machine Learning techniques. This presentation aims at showing an optimisation framework for the optimal design of a micromixer. This device consists of a rectangular pillar in a microchannel, which promotes vortex-shedding phenomenon to mix two fluids moving in parallel to each other. Since in the search of new configurations (longitudinal aspect ratio, blockage ratio and Reynolds number) an uncontrolled sampling may lead to undesired devices which do not lead to vortex-shedding, these are simulated and important computational resources may be wasted. By means of Machine-Learning classifiers, one can know prior simulation whether this new design is worth for simulation to achieve a final solution. This allows us to perform efficiently a multi-objective optimization of the micromixer device. This approach is extendable to other applications.   

A multi-fidelity simulation framework with uncertainty quantification for predicting natural ventilation in a slum house in Dhaka, Bangladesh

A preliminary study conducted in Dhaka, Bangladesh indicated that there might be an association between the presence of cross ventilation in a slum house and the incidence of pneumonia, which is the leading cause of death in children under five. The objective of this study is to establish a validated computational framework for accurately estimating household ventilation rate in terms of air change per hour (ACH) and to support further investigation of the association. To achieve this objective, we first perform high-fidelity large-eddy simulations (LES) that solve for three-dimensional flow and temperature fields in the home. The LES results are used to develop a relationship for ACH as a function of indoor-outdoor temperature difference and wind conditions. Then, the relationship is implemented into a computationally efficient low-fidelity model with uncertainty quantification (UQ) to predict the mean and 95{\%} confidence interval (CI) of both ACH and volume-averaged temperatures. The model predictions are validated against the field measurements, and the results are interpreted to improve our understanding of the ventilation mechanisms and identify robust ventilation strategies that will work under a variety of weather conditions.   

Fluid dynamics of axisymmetric elongated bodies at high angle of attacks

In this study, flow over an axisymmetric body is investigated to determine the underlying physical phenomena responsible for the emergence of asymmetric flow structure at a high angle of incidence. Direct numerical simulations were conducted on curvilinear staggered grids with a sharp interface immersed boundary method. A pseudo-body-conformal grid is designed to provide higher resolution near the body. The cross-flow boundary layer and 3D vortical structures composed of two counter-rotating vortices are identified along the axisymmetric body like a cone and cone-cylinder configuration. The vortex strength and their interaction, instantaneous and average surface forces, shedding frequencies and angles were investigated in detail. It is discussed how the flow undergoes transition along the length of the body with the increase in the local Reynolds' number.?   

Assessment of the Coanda Effect in 2D and 3D CFD simulations of installed rectangular jets

The Coanda Effect describes the attachment of a jet flow to a nearby surface. In this paper, we show that a Computational Fluid Dynamics (CFD) solution of an installed rectangular jet introduces an anomalous Coanda Effect (i.e. bending of the jet towards the trailing edge of the flat plate positioned adjacent to the nozzle lower lip line) when the simulation is performed in $2$D. The latter is often used for fast estimation of a full $3$D RANS calculation (that can take up to 6 days for the residuals to converge to $1$x$10^{-4}$ on a grid of 7 million cells using 16 cores). Our results show that for a $3$D simulation, the jet flow passed smoothly over the surface without producing significant bending. To fix ideas and prove that a $2$D numerical simulation results in such anomalous deviation, we examine a rectangular nozzle of $8:1$ aspect ratio ($AR$) with a flat plate positioned parallel to the level curves of the jet at a transverse (stand-off) distance of $y_2/D_J = 1.9"$ and trailing edge length, $12"$. We compare a number of 2D simulations under various inflow conditions, turbulence models and flow solvers to show the degree of bending. In the talk, we discuss how the spanwise shear prevents the flow from bending in 3D simulations of the same configuration as the 2D cases.   

Rotational relaxation process for Argon-Nitrogen mixed gaseous thermal plasma using DSMC simulations.

In this study we investigate the rotational relaxation process for Argon-Nitrogen mixed gaseous thermal plasma with initial state composition 75 mol{\%} of Argon and 25 mole{\%} of Nitrogen, having two rotational degrees of freedom for Nitrogen molecules and with no internal degrees of freedom for Argon and electron using Direct Simulation Monte Carlo (DSMC) simulations. The Larsen-Borgnakke model is applied on a single molecular basis in which the relaxation collision number is approximated by the reciprocal of the fraction of inelastic collisions. The DSMC simulations are carried out for rotational relaxation collision number $Z_{r} =$\textit{ 7.5 }associated with \quad the Nitrogen molecule and $Z_{r} =$\textit{ 1 }for Argon and electron with viscosity temperature index $? =$\textit{ 0.75} (VHS model), $? =$\textit{ 1.0} (Maxwell model), and ? \quad $=$\textit{ 0.5} (HS model), having different collision rates. The DSMC simulations are compared with the theoretical predictions for translational and rotational temperatures, defined by $T_{tr\thinspace }= T_{eq} + (T_{tr,0\thinspace }-- T_{eq\thinspace })_{\thinspace }$\textit{exp(- }$\nu t/Z_{r}),$ and $T_{rot\thinspace }= T_{eq} -- (T_{eq}- T_{rot,0})_{\thinspace }$\textit{exp(- }$\nu t/Z_{r}) $respectively as well as for the molecular velocity distribution and rotational energy distribution, and found excellent agreement (error within 5{\%}), and the collision process do not lead to any distortion of the Maxwellian velocity distribution and the Boltzmann distribution for the energy. -/a   

Optimization of Extraoral Aerosol Suction Device using CFD

Dental professionals are exposed to contaminated aerosols and droplets produced during dental procedures. To prevent airborne disease transmission, extraoral suction is needed. Due to its bulky size, high cost, and loud noise, current Extraoral Suction Units are not widely adopted in dental offices. To fulfill this need, a smart extraoral suction cup is designed to attach to an HVE (high-volume evacuator) commonly present in dental offices. Computational fluid dynamics (CFD) simulation is performed using ANSYS to test the droplet and capture efficiency of twenty different designs with consideration of vacuum's pressure. To simulate aerosol, air and droplets were dispersed from a model mouth at various testing speeds under several factors, such as turbulence and dispersion angles. The optimal design had a simulated 84.15{\%} suction efficiency. To improve the design, a clear cover is added for increased droplet capture while maintaining vision for the dentist. CFD simulation was performed to optimize the device with clear cover designs. Appropriate distance between the device and a patient's mouth and device performance are validated. These findings result in a smart suction device with high capture efficiency for easy use with further improvement in the future.