In Phase I of the Offshore Code Comparison Collaboration 7 (OC7) project, we aim to address key questions related to the modeling of viscous hydrodynamic loads on floating platforms. This paper presents results from a collaborative effort within Phase I of the OC7 project. We investigate viscous forces acting on the rectangular members of semisubmersible platforms using high-fidelity computational fluid dynamics (CFD) simulations. The parametric CFD analyses include two-and-a-half-dimensional (2.5D) simulations of pontoon sections of two different semisubmersible platform designs (the VolturnUS-S and the INO WINDMOOR) and full three-dimensional (3D) structure simulations of the VolturnUS-S platform. We extract both the added mass and drag coefficients that are typically used in engineering-level hydrodynamic models from the 2.5D forced oscillation simulations with various Keulegan-Carpenter numbers and compare them to the coefficients from the 3D forced oscillation simulations. We observe consistent variation of the coefficients along the pontoon legs due to 3D platform and free-surface effects. With the breadth of the simulations and component-level detailed analyses, this work establishes a useful database and several empirical formulas for selecting hydrodynamic coefficients when modeling semisubmersible platforms in engineering tools.

Metered dose inhalers (MDIs) have been used for 60 years to treat a range of respiratory diseases including asthma. These portable, low-cost medical devices produce a respirable aerosol of micron-size droplets via flash-atomisation of a fluorocarbon propellant blended with drugs and other chemicals. MDIs produce complex multiphase flows in which turbulence, heat transfer and non-equilibrium flash evaporation play coequal roles in determining aerosol properties. Experimental studies of near-field spray structure are limited, as MDI sprays are not amenable to many conventional optical diagnostic techniques and the parameter space for the various propellants and nozzle geometries used in practice is large. This paper presents an alternative approach using multiphase large eddy simulation (LES). We employed a compressible Eulerian mixed-fluid approach using a custom solver written in OpenFOAM. Phase change was handled using a homogeneous relaxation model (HRM) previously developed for flash-evaporating fuel sprays (Neroorkar et al., 2012). Droplet formation, collisions and transport were modelled using a Sigma-Y specific surface area transport model (Rachakonda et al., 2018). We show that the HRM model can approximately reproduce experimental droplet size distributions obtained using laser diffraction and experimental line-of-sight integrated density distributions obtained using X-ray scattering. Results were obtained for three commonly used propellant-ethanol mixtures at a fixed ethanol mass fraction of 15%. Our model provides new insight into the features of the non-equilibrium flash-evaporation of the jet, how this process varies depending on propellant thermophysical properties, and what non-dimensional parameters are best suited to capturing these variations.

With only a few floating offshore wind turbine (FOWT) farms deployed anywhere in the world, FOWT technology is still in its infancy, building on a modicum of real-world experience to advance the nascent industry. To support further development, engineers rely heavily on modeling tools to accurately portray the behavior of these complex systems under realistic environmental conditions. This reliance creates a need for verification and validation of such tools to improve reliability of load and dynamic response prediction and analysis capabilities of FOWT systems. The Offshore Code Comparison Collaboration, Continued with Correlation and unCertainty (OC6) project was created under the framework of the International Energy Agency to address this need and considers a three-sided verification and validation between engineering level models, computational fluid dynamics (CFD), and experimental results. In this paper, a novel floating offshore wind platform, the Stiesdal TetraSpar, is simulated using CFD under the load conditions defined by Phase IV of the OC6 project. The comparison of these CFD results against the experimental results demonstrated the ability to predict the platform response to waves when imposing the measured wave signals as input. Although validation versus experiment was largely successful, the damping behavior was impacted by uncertainties likely originating from the mooring system and sensor umbilical cable. This extensive comparison effort with multiple CFD practitioners offers insight into best practices to achieve reliable results.

Following the operational success of the Hywind Scotland, Kincardine, WindFloat Atlantic, and Hywind Tampen floating wind farms, the floating offshore wind industry is expected to play a critical role in the global clean energy transition. However, there is still significant work needed in optimizing the design and implementation of floating offshore wind turbines (FOWTs) to justify the widespread adoption of this technology and ensure that it is commercially viable compared to other more-established renewable energy technologies. The present review explores the application of fully coupled computational fluid dynamics (CFD) modeling approaches for achieving the cost reductions and design confidence necessary for floating wind to fully establish itself as a reliable and practical renewable energy technology. In particular, using these models to better understand and predict the highly nonlinear and integrated environmental loading on FOWT systems and the resulting dynamic responses prior to full-scale implementation is of increased importance.

Abstract In this work we present a coupled computational fluid dynamics and fluid-structure interaction approach for simulating the dynamics of floating offshore wind turbines (FOWTs). The multiphase flow over the platform is solved using a cartesian cut-cell approach coupled with an environmental flow generation module which can model a variety of wind-wave conditions. This is in turn coupled with a 6-degree-of-freedom fluid-structure interaction model to simulate the platform motion, while the mooring lines are represented using a dynamic lumped element model. Finally, an actuator line model (ALM) is used to model the wind turbine rotor. We have applied this methodology to study both the aerodynamic loading on the rotor and the platform dynamics of a FOWT design consisting of a 3.6-MW horizontal-axis wind turbine attached to a Stiesdal TetraSpar platform in a variety of conditions such as free decay, regular and irregular waves. The FOWT geometry and cases presented here are based on those specified by Phase IV of the OC6 (Offshore Code Comparison Collaboration, Continued with Correlation and unCertainty) project under the International Energy Agency Wind Technology Collaboration Programme (IEA Wind) Task 30. We have selected a set of cases to study the response of the FOWT to various types of loading. The first set of tests are to determine the exact equilibrium location of the FOWT for numerical simulation. In these tests we will also investigate the effect of grid resolution on the results. Next, we simulate the free decay response of the FOWT in the surge, heave, and pitch degrees of freedom. After that we investigate the turbine thrust for both floating and fixed platform configurations. Finally, we present a set of cases that illustrate the platform response to both regular and irregular waves. The simulation results match very well with the experimental data for all the cases that we have studied. Thus, this approach can be seen as a viable method to carry out predictive simulations for studying FOWT platform motions, hydrodynamic loading, and aerodynamic performance.

An experimental investigation to detect cavitation in the nozzles of working Diesel injectors was conducted. Cavitation behavior of Diesel injectors was characterized by a non-dimensional cavitation parameter and a coefficient of discharge. Transient behavior in Diesel injectors with different needle opening pressures and different numbersof nozzle holes was observed and measured. The behavior of sharp-edged single and multi-hole injector tips was found to be reasonably consistent with established characteristics of cavitating nozzles, as observed in steady-state experiments and as predicted by a one-dimensional model. The measurement of flow through a rounded multi-hole tip was consistent with the known behavior of non-cavitating nozzles. INTRODUCTION One important method of reducing emissions in Diesel engines is to improve fuel injector spray breakup, producing smaller and more disperse droplets. The flow inside the fuel injector nozzle is known to have a significant effect on the spray, but researchers have not discovered the exact nature of this effect [1]. Recent investigations have suggested that cavitation occurring within the fuel injector nozzle significantly affects spray breakup [2, 3]. However, much of what we know about cavitating nozzles has come from scaled-up models, with precisely determined geometry. Real fuel injector nozzles may have minute imperfections which can cause significant changes in the flow [4]. This investigation uses an experimental technique to indirectly detect the existence of cavitation in a variety of real injectors. The results of this technique should prove especially interesting in more complicated, multi-hole injectors. This experimental method will be applied to a single hole pump line injector and a multihole hydraulic electronic unit injector with sharp and rounded nozzle inlets. A One-Dimensional Model of Cavitating Nozzles Cavitation bubbles form because of the very low static pressure that occurs in high speed nozzle flow near a sharp inlet corner. This low static pressure is predicted by incompressible potential flow theory, which indicates that flow around a sharp corner, (e.g. a corner with a zero radius of curvature), will have infinite negative pressure. This physically impossible result is a direct consequence of the constant density restriction. In real injectors the fuel density decreases with decreasing pressure, most likely leading to a change in phase. The sharper the corner and the higher the velocity, the more likely cavitation is to occur. Sac Volume: Point 1 Contraction: Point c Cylinder: Point 2 Figure 1. Schematic of nozzle flow In the case of a sharp inlet, where the flow separates at the corner, the flow experiences a vena contracta. A diagram of the sharp entrance flow is shown in Fig. 1. Point 1 would be downstream of the injector needle and yet far enough upstream of the nozzle that the local velocity would be small, such as in the sac of the injector. Point c is downstream of the inlet, where the vena contracta effect is a maximum. In the case of a sufficiently rounded nozzle this point is nonexistent, in which case this analysis may not be useful. For convenience a ratio between the area at the contraction and the nominal nozzle area, known as the coefficient of contraction, is defined: Cc ≡ Ac A {1} Ac represents the effective flow area through the contraction and A represents the nominal nozzle area. The value of the contraction coefficient varies with the nozzle geometry and cavitation characteristics. For a very rounded entrance, the flow will not separate and the coefficient of contraction will be unity. For a short nozzle with a sharp entrance, conformal mapping by von Mises predicts a coefficient of contraction of 0.611 [5]. Experimental data seems to suggest that the coefficient of contraction is a constant with respect to Reynolds number, upstream pressure, and downstream pressure [6, 7]. Interestingly, the steady state coefficient of contraction for a sharp entrance seems to be around 0.61 for both cavitating and non-cavitating nozzles, as measured by Numachi [7]. When and how much the contraction area varies is very important and not well known. At increasingly high injection pressures the vapor region which bounds the contraction has been observed to elongate, apparently without further constricting the flow [8]. Another relevant integral property of the flow is the coefficient of discharge, Cd. The coefficient of discharge represents the efficiency of the nozzle between points 1 and 2 and thus is a measure of whatever losses occur in the nozzle The definition of the coefficient of discharge is: Cd ≡ ṁ A 2ρ P1 − P2 ( ) {2} In order to get closure for the one-dimensional model, an important assumption is made. We assume that the pressure at the point of contraction in a cavitating nozzle is equal to the vapor pressure, Pv. In this simplified view of the nozzle, all the losses are assumed to occur between c and 2. The mass flow rate behaves quite peculiarly under these assumptions. If the contraction pressure, Pc, is fixed at the vapor pressure, Pv, then the mass flow rate becomes independent of back pressure: ṁ = ACc 2ρ P1 − Pv ( ) {3} This peculiar behavior is similar to compressible choking, in that the mass flow rate depends only on the upstream pressure, but not on downstream pressure. This behavior was observed by Randall in cavitating venturi nozzles [9]. It is still unknown whether this flow is actually sonic because of the complexity of the twophase flow. We can combine the definition of Cd, continuity, and Bernoulli's equation to obtain the following expression for the coefficient of discharge of a cavitating nozzle: Cd = Cc P1 − Pv P1 − P2     1 2 {4} The pressure ratio in the right side of Equation 4 turns out to be a very useful cavitation parameter and is referred to as K in the remainder of this paper. K ≡ P1 − Pv P1 − P2 {5} Nurick plotted the coefficient of discharge versus the cavitation parameter, K on log-log axes in order to verify the square root dependence of Cd onK [6]. He observed that the data from the cavitating region lay on a straight line with a slope of one-half, where Cc is the value of the Y-intercept. At some point, the value of K is high enough that the nozzle no longer cavitates. The higher values of K occur when the difference between the upstream and downstream pressure is small. At high values of K the coefficient of discharge stays fairly constant or decreases with increasing K and thus falls to the right of the cavitating line. This variation occurs because the coefficient of discharge is no longer a function of K, but depends on the Reynolds Number instead. In order to further validate Nurick's findings, we have collected more data for sharp nozzles with an L/D ratio of about 4. The data come from real-scale experiments as well as scaled-up experiments and spans sixty years of research [8, 10, 11, 12, 13]. As shown in Fig. 2, this wide variety of sources tends to confirm Nurick's hypothesis. In an actual injector P1 represents the sac pressure. Unfortunately, it is extremely difficult to measure the sac pressure of a working injector. Instead, fuel injectors may be equipped with a pressure transducer just upstream of the needle. Because the exact sac pressure is not known, we must contend with the fact that our measured coefficient of discharge will actually include strong needle effects. During periods of low needle lift there will most likely be a large pressure loss across the needle [14]. For the beginning and ending portions of the injection, the needle will dominate the behavior of the coefficient of discharge. Due to needle effects, the measured value of K is only credible during large needle lift. For this reason, the bulk of the previous analysis should be applied to the portion of injection where the needle is nearly fully open. 0.5 0.6 0.7 0.8 0.9 1 1 2 3 Knox-Kelecy Hi royasu Reitz O h r n B e r g w e r k Gelalles T h e o r y C d

Here, three-dimensional Reynolds-averaged Navier–Stokes (RANS) simulations are employed for in-nozzle flow modelling. The aim is to evaluate the capabilities and sensitivities of simulating in-nozzle flow phenomena by using homogeneous phase change models. Two commonly used homogeneous models are employed – (i) Homogeneous Equilibrium Model (HEM) and (ii) Homogeneous Relaxation Model (HRM) A simplified nozzle with reference experimental data is modelled in the developing, super cavitation, and hydraulic flip regimes using OpenFOAM. The influence of the inlet boundary condition and turbulence models are investigated. With HEM, three barotropic compressibility models are tested. The research seeks to understand the strengths and limitations of each approach by comparing different sub-models and their interactions with the flow fields and phase change phenomena. Ultimately, this study contributes to a better understanding of how different modeling choices can influence the accuracy and reliability of cavitation simulations using homogeneous phase change models. While the simulations demonstrated reliable predictions for low-order velocity statistics, substantial discrepancies were identified in the prediction of the cavitating region by both homogeneous models. Additionally, the relaxation model predicted only minor cavitation. In contrast, acceptable flow predictions were achieved using different barotropic compressibility models and inlet boundary condition types, as well as most RANS turbulence models, where the RNG k-ɛ model deviated more from the other turbulence models. Therefore, the added value of the present study is summarised as: (a) Four different cavitation regimes are simulated with HEM and HRM in a single study. (b) The more comprehensive HRM approach does not produce better velocity or cavitation predictions for a low-speed flow using the default relaxation time. (c) The choice of turbulence model can radically affect the cavitation prediction, in addition to velocity fields. (d) The HEM flow simulation has little sensitivity to the barotropic compressibility model.

The internal details of fuel injectors have a profound impact on the emissions from gasoline direct injection engines. However, the impact of injector design features is not currently understood, due to the difficulty in observing and modeling internal injector flows. Gasoline direct injection flows involve moving geometry, flash boiling, and high levels of turbulent two-phase mixing. In order to better simulate these injectors, five different modeling approaches have been employed to study the engine combustion network Spray G injector. These simulation results have been compared to experimental measurements obtained, among other techniques, with X-ray diagnostics, allowing the predictions to be evaluated and critiqued. The ability of the models to predict mass flow rate through the injector is confirmed, but other features of the predictions vary in their accuracy. The prediction of plume width and fuel mass distribution varies widely, with volume-of-fluid tending to overly concentrate the fuel. All the simulations, however, seem to struggle with predicting fuel dispersion and by inference, jet velocity. This shortcoming of the predictions suggests a need to improve Eulerian modeling of dense fuel jets.

High-fidelity computational fluid dynamics (CFD) simulations for design space explorations can be exceedingly expensive due to the cost associated with resolving the finer scales. This computational cost/accuracy trade-off is a major challenge for modern CFD simulations. In the present study, we propose a method that uses a trained machine learning model that has learned to predict the discretization error as a function of largescale flow features to inversely estimate the degree of lost information due to mesh coarsening. This information is then added back to the low-resolution solution during runtime, thereby enhancing the quality of the under-resolved coarse mesh simulation. The use of a coarser mesh produces a non-linear benefit in speed while the cost of inferring and correcting for the lost information has a linear cost. We demonstrate the numerical stability of a problem of engineering interest, a 3D turbulent channel flow. In addition to this demonstration, we further show the potential for speedup without sacrificing solution accuracy using this method, thereby making the cost/accuracy trade-off of CFD more favorable.

In this paper, dual-rotor counter-rotating (CR) configurations of vertical axis wind turbines (VAWTs) are briefly inspected and divided into three types. This investigation was focused on one of these types-the CR-VAWT with co-axial rotors, in which two equal rotors are placed on the same shaft, displaced from each other along it and rotated in opposite directions. For this CR-VAWT with three-blade H-Darrieus rotors, the properties of the design in terms of aerodynamics, mechanical transmission and electric generator, as well as control system, are analyzed. A new direct-driven dual-rotor permanent magnet synchronous generator was proposed, in which two built-in low-power PM electric machines have been added. They perform two functions-starting-up and overclocking of the rotors to the angular velocity at which the lifting force of the blades is generated, and stabilizing the CR-VAWT work as wind gusts act on the two rotors. Detailed in this paper is the evaluation of the aerodynamic performance of the CR-VAWT via 3D computational fluid dynamics simulations. The evaluation was conducted using the CONVERGE CFD software with the inclusion of the actuator line model for the rotor aerodynamics, which significantly reduces the computational effort. Obtained results show that both rotors, while they rotate in opposite directions, had a positive impact on each other. At the optimal distance between the rotors of 0.3 from a rotor height, the power coefficients of the upper and lower rotors in the CR-VAWT increased, respectively, by 5.5% and 13.3% simultaneously with some increase in their optimal tip-speed ratio compared to the single-rotor VAWT.

Fetal and neonatal alloimmune thrombocytopenia (FNAIT) can occur due to maternal IgG antibodies targeting platelet antigens, causing life-threatening bleeding in the neonate. However, the disease manifests itself in only a fraction of pregnancies, most commonly with anti-HPA-1a antibodies. We found that in particular, the core fucosylation in the IgG-Fc tail is highly variable in anti-HPA-1a IgG, which strongly influences the binding to leukocyte IgG-Fc receptors IIIa/b (FcγRIIIa/b). Currently, gold-standard IgG-glycoanalytics rely on complicated methods (e.g., mass spectrometry (MS)) that are not suited for diagnostic purposes. Our aim was to provide a simplified method to quantify the biological activity of IgG antibodies targeting cells. We developed a cellular surface plasmon resonance imaging (cSPRi) technique based on FcγRIII-binding to IgG-opsonized cells and compared the results with MS. The strength of platelet binding to FcγR was monitored under flow using both WT FcγRIIIa (sensitive to Fc glycosylation status) and mutant FcγRIIIa-N162A (insensitive to Fc glycosylation status). The quality of the anti-HPA-1a glycosylation was monitored as the ratio of binding signals from the WT versus FcγRIIIa-N162A, using glycoengineered recombinant anti-platelet HPA-1a as a standard. The method was validated with 143 plasma samples with anti-HPA-1a antibodies analyzed by MS with known clinical outcomes and tested for validation of the method. The ratio of patient signal from the WT versus FcγRIIIa-N162A correlated with the fucosylation of the HPA-1a antibodies measured by MS (r=-0.52). Significantly, FNAIT disease severity based on Buchanan bleeding score was similarly discriminated against by MS and cSPRi. In conclusion, the use of IgG receptors, in this case, FcγRIIIa, on SPR chips can yield quantitative and qualitative information on platelet-bound anti-HPA-1a antibodies. Using opsonized cells in this manner circumvents the need for purification of specific antibodies and laborious MS analysis to obtain qualitative antibody traits such as IgG fucosylation, for which no clinical test is currently available.

IgG antibodies are important mediators of vaccine-induced immunity through complement- and Fc receptor-dependent effector functions. Both are influenced by the composition of the conserved N-linked glycan located in the IgG Fc domain. Here, we compared the anti-Spike (S) IgG1 Fc glycosylation profiles in response to mRNA, adenoviral, and protein-based COVID-19 vaccines by mass spectrometry (MS). All vaccines induced a transient increase of antigen-specific IgG1 Fc galactosylation and sialylation. An initial, transient increase of afucosylated IgG was induced by membrane-encoding S protein formulations. A fucose-sensitive ELISA for antigen-specific IgG (FEASI) exploiting FcγRIIIa affinity for afucosylated IgG was used as an orthogonal method to confirm the LC-MS-based afucosylation readout. Our data suggest that vaccine-induced anti-S IgG glycosylation is dynamic, and although variation is seen between different vaccine platforms and individuals, the evolution of glycosylation patterns display marked overlaps.

Background: The autoimmune bleeding disorder childhood immune thrombocytopenia (ITP) features antibody and T-cell immune responses against platelet self-antigens. This involves activation of B-cells, CD4 + T-cell help, and CD8 + T-cells. Novel prognostic biomarkers are needed to predict recovery and treatment responses. Compared to healthy children, changes in total lymphocytes, CD4 +, regulatory CD4 +, CD8 +T-cells, CD19 + B-cells, and NK cells were described in childhood ITP. Children with transient ITP were also found to have a fewer CD25 hi CD4 + T-cells than those who develop chronic ITP. In contrast, the platelet-stimulated peripheral blood mononuclear cells (PBMC) of children with chronic ITP produced more IL-2 than those of transient ITP. At present, none of these biomarkers have been validated for ITP prognosis, and consequently they are not in clinical use. This emphasizes the need for clinical-grade biomarkers to predict clinical responses in childhood ITP. Aims: To predict spontaneous recovery and IVIg response in newly diagnosed ITP by (1) validating previously suggested changes in immune cell frequencies, and (2) identifying novel immune cell subsets as predictors. Methods: Children with newly diagnosed ITP were randomized 1:1 to observation or IVIg treatment (TIKI trial). Recovery was determined by platelet counts according to IWG criteria 1, 4, 13, 26 and 52 weeks after the diagnosis. For validation, the CD4 +, regulatory CD4 + (CD25 +/CD127 lo), CD8 +, CD19 +, and NK cells were quantified at the time of diagnosis in a centralized laboratory by flow cytometry. For identification of novel predictors, CD4 + central (CD45RO + CD27 +) and effector memory (CD45RO + CD27 -) frequencies were determined. The primary clinical outcome was longitudinal recovery, either spontaneous or after IVIg (adjusted Cox-proportional hazard model). Secondary outcomes included the mean age-adjusted difference between transient/persistent and chronic ITP by multivariate regression, and comparison to age-appropriate healthy control data. Additionally, PBMC (N=6) were analyzed at diagnosis by single-cell RNA sequencing (scRNA-seq) combined with T- and B- cell receptor sequencing (scTCR-seq and scBCR-seq). Results: For validation of previously suggested predictors, the absolute CD3 +, CD4 +, CD8 +, CD19 +, and NK cell counts of newly diagnosed ITP patients were within the age-appropriate healthy reference range (N=158); they were not associated with recovery, and there were no age-adjusted differences between transient/persistent (N=139) and chronic ITP (N=19). The regulatory CD4 + T cell frequency was not reduced compared to age-appropriate reference data; not associated with recovery, and not different between transient/persistent and chronic ITP patients. For identification of novel predictors, we found that high CD4 +effector memory cells numbers were associated with a reduced recovery rate, with an adjusted hazard ratio for complete recovery over a one-year follow-up to be 0.55 (95% CI, 0.35 - 0.85; ≥ median; adjusted for age and treatment; N=150). Chronic ITP patients displayed a mean age-adjusted increase in effector memory CD4 +cells of 1.4 % (95% CI, 0.4 - 2.4; P= 0.005). The association with recovery was independent of a preceding infection, total leukocyte and lymphocyte counts, and the presence of anti-platelet IgG or IgM autoantibodies. ScRNA-seq analysis of 7965 PBMC also showed an effector memory CD4 + T-cell cluster that was expanded in chronic ITP, present in at a frequency of 8.1 ± 1.4 % (mean ± sem) in transient vs 14.8 ± 1.2 % in chronic ITP. This cluster also expressed high levels of fibronectin receptor integrin b1 ( ITGB1) interleukin 32 (IL32) interleukin 7 receptor (IL7R) andlymphotoxin beta (LTB). The effector memory phenotype of CD4 + ITGB1 + T-cells was confirmed by flow cytometry analysis in healthy individuals and ITP patients. Conclusions: Previously suggested changes in T-, B-, or NK cell frequencies could not be validated as predictors of spontaneous recovery or IVIg response. However, we identified the frequency of ITGB1-expressing effector memory CD4 + T cells as an independent predictor of spontaneous recovery and IVIg response. Thus, T cell phenotyping at the time of diagnosis may be suitable to determine prognosis and personalize treatment decisions in newly diagnosed childhood ITP.

One control strategy to increase power production in wind farms is angling wind turbine rotors, in order to steer wakes away from downwind turbines. Although rotor yaw is the most common approach to wake steering, tilting the rotor vertically to steer the wake downward can also increase total farm power. In this study, large eddy simulations of a 15 MW turbine are performed for rotor tilt angles of 0 degrees, 15 degrees, and 30 degrees with below-rated turbulent inflow. Wake characteristics are analyzed, including using circulation to quantify the curled wake's counter-rotating vortex pair and quantifying wake shapes by fitting Legendre polynomials to wake edge polar coordinates. Tilting the rotor causes downward wake steering, wider and vertically compressed wake cross-sections, and stronger counter-rotating vortices. Although the wake velocity deficit recovers similarly for tilted and non-tilted wakes, the power available to a downwind rotor recovers faster because the tilted wake is steered away from the downwind rotor area and is replaced by high-speed air from above. This also causes higher effective wind shear across the downwind rotor. Additional simulations double the gap between the ground surface and the rotor bottom, which affects the wake geometry as well as the downwind power recovery and wind shear. (c) 2022 Elsevier Ltd. All rights reserved.

Past Lagrangian/Eulerian modeling has served as a poor match for the mixing limited physics present in many sprays. Though these Lagrangian/Eulerian methods are popular for their low cost, they are ill-suited for the physics of the dense spray core and suffer from limited predictive power. A new spray model, based on mixing limited physics, has been constructed and implemented in a multi-dimensional CFD code. The spray model assumes local thermal and inertial equilibrium, with air entrainment being limited by the conical nature of the spray. The model experiences full two-way coupling of mass, momentum, species, and energy. An advantage of this approach is the use of relatively few modeling constants. The model is validated with three different sprays representing a range of conditions in diesel and gasoline engines.

Starting with two well-tested, one-dimensional models of non-evaporating, mixing-limited sprays, governing equations for liquid mass and two-phase momentum for each model can be manipulated to reveal the formal similarity between momentum and liquid volume fraction. The consequence of this similarity is that momentum, when properly non-dimensionalized, is equal to the liquid volume fraction at any time and at any axial location within a non-evaporating, mixing-limited spray with a constant rate of injection. An alternative, the more well-known similarity between mass fraction and velocity, is also mathematically evident. We compare predictions of this mathematical analysis to high-fidelity, first-principles simulation results of a non-evaporating spray to assess the validity of the theoretical similarity. The analysis of the simulation not only confirms the mathematical derivations but also points to subtlety in the definition of the spray velocity. In particular, the density-weighted velocity is required to observe similarity. The requirement of density-weighted velocity means that similarity tests require knowledge of both phase velocities. The agreement also works to confirm that the first-principles simulations are indeed mixing-limited, despite the finite nature of domain size and resolution.

The early and late portions of transient fuel injection have proven to be a rich area of research, especially since the end of injection can cause a disproportionate amount of emissions in direct injection internal combustion engines. The presented work simulates a gasoline direct injector operating under cavitating conditions by employing a more gradual and easily implemented model of closure that avoids spurious water-hammer effects. The results show cavitation at low valve lift for both flash-boiling and nonflash-boiling conditions. Further, this study reveals post-closure dynamics that result in dribble, which is expected to contribute to unburnt hydrocarbon emissions. Flashing versus nonflashing conditions are shown to cause different sac and nozzle behaviors after needle closure. In particular, a slowly boiling sac is observed for the flash-boiling condition which causes spurious postinjection behaviors. However, post-needle closure, traces of ambient gas and liquid fuel, which are the main source of dribble, are observed for the nonflashing condition. The current work indicates more tip wetting for the flash-boiling condition compared to the nonflashing condition during both pre-and post-needle closure phases.

The acceleration of microparticles to supersonic velocities is required for microscopic ballistic testing, a method for understanding material characteristics under extreme dynamic conditions, and for projectile gene and drug delivery, a needle-free administration technique. However, precise aerodynamic effects upon supersonic microsphere motion at sub-300 Reynolds numbers have not been quantified. We derive drag coefficients for microspheres traveling in air at subsonic, transonic, and supersonic velocities from the measured trajectories of microspheres launched by laser-induced projectile acceleration. Moreover, the observed drag effects on microspheres in atmospheric (760 Torr) and reduced pressure (76 Torr) are compared with existing empirical data and drag coefficient models. We find that the existing models adequately predict the drag coefficient for subsonic microspheres, while rarefaction effects cause a discrepancy between the model and empirical data in the supersonic regime. These results will improve microsphere flight modeling for high-precision microscopic ballistic testing and projectile gene and drug delivery.

In fluid physics, data-driven models to enhance or accelerate solution methods are becoming increasingly popular for many application domains, such as alternatives to turbulence closures, system surrogates, or for new physics discovery. In the context of reduced order models of high-dimensional time-dependent fluid systems, machine learning methods grant the benefit of automated learning from data, but the burden of a model lies on its reduced-order representation of both the fluid state and physical dynamics. In this work, we build a physics-constrained, data-driven reduced order model for the Navier-Stokes equations to approximate spatio-temporal turbulent fluid dynamics. The model design choices mimic numerical and physical constraints by, for example, implicitly enforcing the incompressibility constraint and utilizing continuous Neural Ordinary Differential Equations for tracking the evolution of the differential equation. We demonstrate this technique on three-dimensional, moderate Reynolds number turbulent fluid flow. In assessing the statistical quality and characteristics of the machine-learned model through rigorous diagnostic tests, we find that our model is capable of reconstructing the dynamics of the flow over large integral timescales, favoring accuracy at the larger length scales. More significantly, comprehensive diagnostics suggest that physically-interpretable model parameters, corresponding to the representations of the fluid state and dynamics, have attributable and quantifiable impact on the quality of the model predictions and computational complexity.