Session G03: Energy (5:00pm - 5:45pm CST)

Broadening of reaction zones in turbulent lean methane-air premixed flames

Internal structure of extremely turbulent premixed flames stabilized on a large diameter Bunsen burner is investigated using simultaneous planar laser-induced fluorescence of formaldehyde molecule and hydroxyl radical as well as stereoscopic particle image velocimetry. Lean methane-air flames with the mean bulk flow velocities of 5 to 35 m/s are tested. Reynolds and Karlovitz numbers for the tested conditions change from 19 to 2729 and 0.3 to 76.0, respectively. Results show that the preheat and reaction zone thicknesses can increase to values that are about 6.2 and 3.9 times those of the laminar flame counterpart, respectively. While preheat zone broadening is expected from results in the literature, thickening of the reaction zone is not commonly observed, especially for a relatively large diameter Bunsen burner. Turbulent flow characteristics such as vorticity, swirling strength, and rotational kinetic energy of vortices are estimated to investigate the underlying reason for the broadening of preheat and reaction zones. The results suggest that penetration of energetic vortices into the preheat and reaction zones may be a potential reason for the broadening of these zones.   

Multidimensional direct numerical simulations of superknock in a thermally inhomogeneous DME/air mixture

Superknock propensity in a stoichiometric dimethyl-ether (DME)/air mixture with temperature inhomogeneities at conditions relevant to internal combustion engines is investigated using multidimensional direct numerical simulations. To examine the detonation development, simulations are performed by varying the temperature fluctuations, characteristic length scale and initial mean temperature, with initial temperatures lying in the low-, intermediate-, and high-temperature chemistry regimes. The volumetric fraction of the mixture regions that is prone to detonation development, $F_D$, which was proposed as a metric to predict the amplitude of knock intensity in previous studies by Luong et al. (Proc. Combust. Inst. 2020; Flow Turbul. Combust. 2020), is adopted. Regardless of the initial mean temperatures, $F_D$ serves as a reliable indicator of the subsequent detonation development that can capture the level of knock intensity. Partially, $F_D$ shows a good correlation with the heat release fraction of the mixture regions with pressure greater than the equilibrium pressure, $F_H$. The detonation regimes are also well captured by the predictive nondimensional numbers, $\varepsilon_p$ and $\xi_p$, that are extended from $\varepsilon$ and $\xi$ in the regime diagram of Bradley et al.   

Field characterization of the effect of wind veer on wind turbine power generation.

Wind direction variation with height (wind veer) plays an essential role in the inflow wind field for the current utility-scale wind turbines. We explore the interactions of wind veer and wind turbine blades and their impact on the turbine performance using a 5-year field dataset. Wind veer exhibits an appreciable diurnal variation that veering and backing winds tend to occur during nighttime and daytime, respectively. We further propose to divide the wind veer conditions into four scenarios based on their changes in turbine upper and lower rotors which have different influence on the lift and drag acting on different rotor sections: VV, (upper rotor: veering, lower rotor: veering), VB (upper rotor: veering, lower rotor: backing), BV (upper rotor: backing, lower rotor: veering), BB (upper rotor: backing, lower rotor: backing). Such division allows us to elucidate the impact of wind veer on turbine power generation. The clockwise-rotating turbines tend to yield substantial power losses in scenarios VV and VB and small power gains in scenarios BV and BB. The counterclockwise-rotating turbines follow exactly opposite trends as the clockwise turbine. The findings are generalizable to onshore and offshore wind sites with varying wind veer conditions for power evaluation.   

Influence of atmospheric boundary layer wind speed and direction shear on utility-scale yaw misaligned turbines

The intentional yaw misalignment of turbines in a wind farm to deflect energy deficit wake regions, or wake steering, has demonstrated potential as a wind farm control approach to increase collective power production. The potential for wake steering depends, in part, on the power reduction of yaw misaligned turbines. In the atmospheric boundary layer, the sheared wind speed and direction may change significantly over the rotor area, resulting in a relative inflow wind speed and angle of attack to the blade airfoil which depends on the radial and azimuthal positions. In order to predict the power production for an arbitrary yaw misaligned turbine based on the incident boundary layer velocity profiles, we develop a blade element model which accounts for wind speed and direction changes over the rotor area. A field experiment is performed using multiple utility-scale wind turbines to characterize the influence of yaw misalignment and the incident velocity profiles on the resulting turbine power production, and the model is validated using the experimental data. The power production of a wind turbine is asymmetric as a function of the direction of yaw misalignment and depends on a nonlinear interaction between the yaw, the incident wind conditions, and the turbine control system.   

A new analytical wavelet-Gaussian wake model

A new analytical wake model focusing on capturing near wake phenomena is proposed. A Ricker wavelet function summed with a normal Gaussian is used to capture flow acceleration and hub jet in the near wake and transition into a single Gaussian curve downstream. Simplifications are implemented to allow for inexpensive computational cost while still upholding the fidelity of the model. Parameters are considered to be functions of flow conditions and turbine properties and expressed analytically. Large eddy simulations and small scale wind tunnel experiments are used to validate the model.   

Dynamic effects of inertial particles on the wake recovery of a model wind turbine

Impacting particles such as rain, dust, and other debris can have devastating structural effects on wind turbines, but little is known about the interaction of such debris within turbine wakes. This study aims to characterize behavior of inertial particles within the turbulent wake of a wind turbine and relative effects on wake recovery. Here a model wind turbine is subjected to varied two-phase inflow conditions in a wind tunnel, with air as the carrier fluid ($Re_\lambda = 49$-$88$) and polydisperse water droplets (26 to 45 $\mu$m in diameter) at varied volume fractions ($\Phi_v = 0.24 \times 10^{-5}$ - $1.3 \times 10^{-5}$), comparing with sub-inertial particles [\textit{i.e.,} tracers] that follow the inflow streamlines. Phase doppler interferometry and particle image velocimetry were employed at multiple downstream locations, centered with respect to turbine hub height. Analysis considers energy and particle size distribution within the wake focusing on turbulence statistics and preferential concentration. Near wake statistics show similarities to those of turbines in single-phase flows, and show persisting velocity deficits at least as far as $9.5$ rotor diameters downstream.   

The Area Localized Coupled Model for Analytical Mean Flow Prediction in Arbitrary Wind Farm Geometries

We introduce the Area Localized Coupled (ALC) model, which extends earlier approaches that couple classical wake superposition and atmospheric boundary layer models to enable applicability to arbitrary wind-farm layouts. Coupling wake and top-down boundary layer models is particularly challenging since the latter requires averaging over planform areas associated with certain turbine-specific regions of the flow. The ALC model uses Voronoi tessellation with a cell around each turbine and a developing internal boundary layer description over Voronoi cells upstream of each turbine. Coupling is achieved by enforcing a minimum least-square-error in mean velocity in each cell between the top-down model and a given wake model (e.g. Jensen or Gaussian model). The ALC model, using a wake model with a top-hat to Gaussian profile, is applied to two wind farm geometries and compared with LES data for a circular wind farm generated at NREL using SOWFA, and a farm that has half of the turbines arranged in an array and the other half randomly distributed simulated using LESGO from JHU. The ALC model is shown to produce improved generated power predictions for both the farm and individual turbines over prevailing approaches for a range of wind inflow directions.   

Enhanced wind-farm performance using windbreaks

Many studies have considered wind farm performance optimization by manipulating the interaction between wind turbine wakes and the atmospheric boundary layer. Previous studies showed that windbreaks can increase the power production of a row of turbines, but found that the additional drag imposed by the windbreaks makes them ineffective when used in an infinite wind farm array. We use large eddy simulations of a wind farm with six rows to show that windbreaks can increase the power production of large wind farms. In a wind farm, the optimal windbreak configuration depends on a balance between this speed-up effect and the additional drag imposed by the windbreaks. Therefore, we find that the optimal windbreak height in a wind farm is significantly lower than the windbreak height used for stand-alone turbines. In large wind farms without windbreaks, the vertical kinetic energy flux that brings down high-velocity wind from above the wind farm to the hub-height plane plays a crucial role. However, surprisingly, we find that this effect is not important in wind farms with windbreaks. Instead, the turbines benefit from the favorable total pressure flux created by windbreaks of intermediate height.   

Large-Eddy Simulations of a Cross-Flow Turbine with Intracycle Angular Velocity Control

Straight-bladed cross-flow turbines have certain advantages over axial-flow turbines such as no need for yaw control and ease of manufacturing and maintenance. Experiments and Reynolds-averaged Navier-Stokes (RANS) simulations have shown that optimized sinusoidal variation of angular velocity, instead of a constant angular velocity through the turbine rotation, increases the power conversion efficiency by up to 54\%. The relative flow velocity and effective angle of attack profile experienced by the blade are modified, hence also controlling the flow separation at the blade. High resolution visualization of flow structures is difficult in experiments, while RANS modeling has limitations in separating flows and at moderate Reynolds numbers. Hence large-eddy simulations (LES) are performed to investigate the effect of intracycle variation of angular velocity on the flow separation and vortex dynamics. The computed flow field is validated with particle image velocimetry (PIV) data from an experiment using a constant angular velocity turbine. The flow field for a sinusoidal angular velocity turbine is analyzed in terms of flow separation at the blade, the blade-vortex interactions, and their effect on turbine performance.   

Design and Optimization of an Internal Turbine to Interface with AeroMINE using the Proper Orthogonal Decomposition Method

The design of a low Reynolds number (order 100,000) turbine blade is presented that maximizes lift over drag using design of experiments and the Proper Orthogonal Decomposition (POD) method. This blade was created to interface to the novel wind energy harvester AeroMINE, which has no external moving parts. Power is extracted by a low-speed horizontal axis turbine mounted in the intake. Flow in the intake is driven by a pressure gradient caused by AeroMINE's external mirrored airfoil-pairs. As wind passes through these external foils, suction occurs on their low-pressure surfaces. This suction draws out air from air-jets located along the surfaces of the foils themselves and causes a pressure gradient within the intake. To begin the design process, an exploration space was set up using a D-Optimal design. Eight factors, which described the geometry of the airfoil, with three levels per factor were examined. Constraints on each factor were based on airfoil data found in the literature for low Reynolds number flows. The results of the design exploration provided an initial guess for the optimization method. The POD method was used to reduce the computational cost of the optimization procedure.   

Calibration Procedure for Gaussian-based Analytical Wake Model Using SCADA Data

Wind turbine wakes are responsible for the reduction of wind farm power generation. Many studies have been done using experimental and computational methods to model wind turbine wakes. However, the industry favors the analytical model for wind farm wake modeling because it can provide reasonably accurate results without the need for extensive simulation and experiment. In analytical wake modeling, using existing parameters from previous studies may cause inaccurate predictions due to differences of wind turbine specifications and specific wind farm variation. Calibrating the analytical model based on the specific wind farm setting can improve the prediction accuracy. We propose a procedure for wind farm wake modeling using a Gaussian-based analytical wake model and SCADA data. The wake growth rate varies across the wind farm based on the local streamwise turbulence intensity. A case study at a wind farm in Iowa will be presented. The wake model was calibrated by using the proposed procedure with turbine pairs selected from the wind farm. The results were compared with the industry standard wake model. This is the first time SCADA data was used to calibrate the Gaussian-based analytical wake model.   

Energy harvesting using horizontal axis wind turbines with hydrostatic transmission

The integration of a hydrostatic transmission provides flexibility and modularity in the turbine configuration, allowing installation of a pump in the turbine nacelle and a motor at ground level while having the possibility to decouple the input and output angular velocities of the power transmission system. Energy harvesting using wind turbines with hydrostatic transmission was studied experimentally using kW-level test units in the field. We tested and validated the viability of hydrostatic transmission capabilities and compared against laboratory experiments using an electric motor to drive the wind-turbine rotor. Power spectra as a function of the turbulent incoming flow reveal distinct modulation of the system. These modifications may reduce the turbine structure mass to 30{\%} and 5-15{\%} in the cost of energy. The measured power transmission efficiency was found to be in the range of 70-77{\%}.   

Effect of Aspect Ratio on Cross-Flow Turbine Performance

Cross-flow turbines convert kinetic power in wind or water currents to mechanical power. Unlike axial-flow turbines, the influence of geometric parameters on turbine performance is not well-understood, in part because there are neither generalized analytical formulations nor inexpensive, accurate numerical models that describe their fluid dynamics. Here, we experimentally investigate the effect of aspect ratio – the ratio of the blade span to rotor diameter – on the performance of a straight-bladed cross-flow turbine in a water channel. To isolate the effect of aspect ratio, all other non-dimensional parameters are held constant, including the relative confinement, Froude number, and Reynolds number. The efficiency is found to be invariant for the range of aspect ratios tested (0.95 – 1.63), which we ascribe to minimal blade-support interactions characteristic of the particular turbine investigated. Finally, a subset of experiments is repeated without controlling for the Froude number and the efficiency is found to increase, a consequence of Froude number variation that could mistakenly be ascribed to aspect ratio. This highlights the importance of rigorous experimental design when exploring the effect of geometric parameters on cross-flow turbine performance.   

A low-order wake interaction model for ocean current turbine arrays operating in turbulent flow

Ocean Current Turbines (OCTs), which function similarly to wind and tidal turbines, represent a promising technology for harnessing the energy from oceanic currents. The power extracted by the turbines can be significantly affected by the turbulence intensity in the upstream flow. For turbines distributed in an arrayed configuration, the highly turbulent wakes behind each upstream turbine must also be considered. In this presentation, we describe a low-order analytical wake interaction model capable of estimating the total array power of OCTs operating in any stacked configuration, and embedded in a flow with Turbulence Intensity (TI). The model incorporates both near and far-wake effects associated with each turbine, and has been validated using high-resolution Large Eddy Simulations (LES) performed using the STAR-CCM software. The simulations were driven by realistic ocean turbulence conditions derived from Gulf Stream current measurements. The analytic model was validated over a wide range of turbulence intensities and OCT array configurations, and can also be applied to evaluate the performance of wind turbine installations.   

Geometric Optimization of Self-Reacting Point Absorber for Specific Sea-Sites

Self-reacting point absorbers (SRPA) are devices that are designed to capture power from the ocean. The wave climate is site-specific. As a result, the power capturing performance of the SRPA is improved by optimizing the device geometry for selected sea sites. The SRPA consists of two concentric bodies, torus (shallower body at the water surface) and float (deeper body acts as reference), heaving at different frequencies. The hydrodynamic coefficients of both bodies are calculated using boundary element solver (NEMOH); the time-domain modeling of multi-body dynamics of SRPA is modeled using open source code WEC-Sim. We will discuss results from our Genetic Algorithm optimization study where the float and the torus are optimized independently. The optimized geometry of the SRPA increases the power capture performance by 50{\%} for both selected sea-sites.   

Predicting Energy Harvesting Efficiency in Two Tandem Oscillating Foils

Oscillating foils in synchronized pitch/heave motions can be used to harvest energy from moving water and offer an alternative to rotary turbines. By understanding the wake structure and its correlation with the pitch/heave kinematics, one can develop predictive models for how foils can coordinate in array configurations. In order to establish a relationship between foil kinematics and wake characteristics, a wide range of kinematics is explored in a 2-foil configuration with inter-foil spacing from 4-9 chord lengths separation. With an in-depth wake analysis, the trailing foil efficiency is normalized by the mean convective velocity and the turbulent kinetic energy in the wake. This normalization accounts for the mean flow in addition to the energy transported by the coherent leading edge vortices (LEVs) shed from the leading foil. Using the mean wake velocity, a model is developed to predict the trailing foil's efficiency demonstrating four different regimes. These regimes are distinguished by the kinematics of the leading foil, which dictates the LEV strength and the trailing foil efficiency profile.   

Fluids for flow batteries: the case of MEEPT

We report viscous flow properties of a redox-active organic molecule, MEEPT, a candidate for non-aqueous redox flow batteries, and its radical cations. A microfluidic viscometer enabled the use of small sample volumes in determining viscosity as a function of shear rate and concentration in acetonitrile, both with and without supporting salts. From concentration-dependent viscosity measurements, molecular information, such as intrinsic viscosity, hydrodynamic diameter, and the Huggins coefficient were inferred. Model fit credibility was assessed using the Bayesian Information Criterion. We found that the MEEPT and its charged cation are ``flowable'' and do not flocculate at concentrations up to 0.5 M. MEEPT has a hydrodynamic diameter of $\sim$ 8.5 {\AA}, which is comparable to molecular dimensions of single molecules obtained from density function theory calculations. The results suggest that MEEPT is a promising candidate for redox flow batteries in terms of its viscous flow properties. Reference: Wang, Y., et al., Phys. Fluids, in press, http://doi.org/10.1063/5.0010168   

Fire Extinguisher Innovation Based In Nature

Many of today's greatest innovations are based upon principals and mechanisms found in nature. One mechanism which shows great potential for use in contemporary technology is the defense mechanism of the Bombardier Beetle, which uses exothermic reactions to heat and pressurize a chemical compound before releasing it towards its attacker. A fire extinguisher using this same principal has the potential to be lighter as well as have a greater range and effectiveness, than traditional fire extinguishers. We endeavor to design and build a mock-up of such a fire extinguisher. We first focus on the pressurized heating chamber where cartridge heaters, driven by batteries, initiate the phase change of the water. We use ANSYS simulations to test different chamber geometries and heater configurations to optimize the balance between heating requirements and weight limitations.   

Supersonic Mixing and Combustion in a Cavity Flameholder

Efficient mixing and flame stabilization remain central challenges to supersonic combustion due to their relatively long time scales with respect to residence time. Cavity flameholders can aid such efforts, but can also introduce losses, and the entrainment and flameholding mechanisms remain incompletely described. We study these in detail by numerical simulations of the Illinois Arc-heated Combustion Tunnel (ACT) II facility, where a round underexpanded ethylene jet issues horizontally into a cavity under a $M = 3$ oxidizer freestream. During sustained combustion, a flame stabilizes around the jet and acts as a heat source on the freestream. Results are compared against wall-pressure measurements and shock-angle imaging from corresponding experiments. We focus on the relative effects that mean versus turbulent and inert versus burning flow have on mixing and entrainment and, in the sustained-flame case, the relative effects of flow versus chemical reactions via local Damköhler-number statistics. Particularly, we assess how recirculation and turbulence affect jet-flame structure and stabilization.   

System for the Generation, Compression, and Storage of Hydrogen Gas from Renewable Sources.

The purpose of this research was to implement a control system to support the autonomous generation, compression and storage of hydrogen gas from renewable sources. The system was designed using commercial off the shelf components and was required to be mobile for operating in remote locations. $\backslash $A photovoltaic (PV) array was utilized to power the system. Ambient moisture in the air was collected via dehumidifiers and processed through an electrolyzer to produce hydrogen gas. The hydrogen gas produced from the electrolyzer was required to be dry, so a vacuum dehumidifier with a semi-permeable membrane was utilized to remove residue moisture from the hydrogen gas. An electrochemical hydrogen (ECH) compressor was used to compress the hydrogen gas for storage. The system performance was constrained by the compressor flow rate, and output from other components had to be subordinate to this. $\backslash $Hydrogen produced from this system can be stored for an indefinite about of time in order to provide steady stable power regardless of fluctuations in demand. The hydrogen can then be used in fuel cells to provide electricity or to power long endurance drones from remote locations.$\backslash $pard-/abstract-$\backslash $\tex