42nd Dayton-Cincinnati
Aerospace Sciences Symposium

List of Submitted Abstracts

* Note that appearance on this list does not guarantee that the abstract has been or will be accepted. All abstracts submitted prior to the deadline of 20 January, 2017 will be reviewed for suitability and technical content. Acceptance will be confirmed via email with the submitting author.

Acoustics

Abstract ID: 42DCASS-130

Flow Measurements from a Supersonic Rectangular Nozzle Exhausting Over a Flat Surface

Florian Baier (flobaier1@gmail.com)
University of Cincinnati
Pablo Mora
University of Cincinnati
Ephraim Gutmark
University of Cincinnati
Kailas Kailasanath
Naval Research Laboratory


Advances in jet technology have pushed towards faster aircraft, leading to more streamlined designs and configurations, pushing engines closer to the aircraft frame, even embedding them in an effort to reduce drag. This creates additional noise sources stemming from interactions between the jet flow and surfaces on the aircraft body, as well as interactions between the jet and the ground during takeoff and landing. This paper studies how the presence of a flat plate affects the flow structures in the supersonic jet from a rectangular converging-diverging nozzle, and how these impact the major or minor axis of a 2:1 aspect ratio (De=0.813in) nozzle. Comparisons are drawn between baseline cases without a plate to properly assess the impact of the plate at NPRs of 2.5, 3.0, 3.67, 4.0, 4.5, as well as temperature ratios from cold TR=1.1 up to heated TR=2.0 and TR=2.6. Streamwise particle image velocimetry (PIV) data was taken to extract average velocity data in the jet axis direction and turbulent kinetic energy (TKE) of the flow. Baseline results without a plate at an aspect ratio of 2:1 showed major trends in shock cell spacing, potential core length, and shear layer development. Increasing the pressure ratio at constant temperature is shown to increase the potential core length through a widening of the shock cell spacing, increasing the potential core length from ~6.1De to ~9.2De downstream of the nozzle exit from NPR=2.5 to NPR=4.5. Temperature is shown to shorten the length of the potential core, a 0.4De reduction in length for a TR increase from TR=1.1->2.6 at NPR=3.0. Flush mounting the flat surface is shown to result in a downstream shift of the maximum TKE from ~8De to ~11De downstream. In combination with previously published acoustic far field data these flow measurements will serve to gain a more in depth understanding of the noise components in embedded supersonic nozzles.

Abstract ID: 42DCASS-194

Aspects of Supersonic Jet Noise for Three-stream Engines

Christopher Ruscher (cjrusche@gmail.com)
Spectral Energies LLC
Sivaram Gogineni
Spectral Energies LLC
Alex Giese
Air Force Research Laboratory
Thomas Ferrill
Air Force Research Laboratory


Waiting for public release

Aircraft Design

Abstract ID: 42DCASS-040

The Design and Optimization of a Variable Camber Compliant Wing

Ryan Petrie (ryan.petrie@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-049

Design Optimization of a Heavy Lift VTOL UAS

Justin Ouwerkerk (ouwerkjn@mail.uc.edu)
University of Cincinnati
Kelly Cohen
University of Cincinnati
Shabban Abdallah
University of Cincinnati


The design and optimization of a Vertical take-off and Landing (VTOL), endurance, heavy lift drone is almost a complete misnomer. The majority of all heavy lift UAS systems are exactly the anti of most endurance designed UAS systems. For this project we will design a custom heavy lift UAS platform which will fly for over 30 minutes, and will prove to be robust enough to be reconfigurable for a variety of mission parameters. Using cutting edge manufacturing techniques, combined with various fluids and physics principle's, a unique, and highly configurable platform will be optimized. Of these principles propeller studies, namely the use of energized flow fields in fluid research, will be explored in an attempt to gain the maximum of efficiency from the designed propulsion platform. Cutting edge prototyping systems will be used to print lightweight, composite infused materials, and exotic metals parts which will allow for an overall frame package of just over a kilogram. A design optimization study will ensure the design is suitable, and as optimized as the rest of the platform, while utilizing FEA and CAD processes to ensure airworthiness. This design will prove suitable for various future applications, and modular enough to be easily reconfigured for any intended use.

Abstract ID: 42DCASS-096

Conceptual-Level Design Process for Hypersonic Systems

Jose Camberos (jose.camberos@us.af.mil)
Air Force Research Laboratory
Zachary White, Brendan Rooney, Dalton Baier
Air Force Research Laboratory


The Aerospace Systems Directorate, High-Speed Systems Division seeks to advance hypersonic technology development by establishing a conceptual design capability within the laboratory. A core team of researchers with unique sets of skills, talents, and tools can now postulate and design hypersonic air platforms for myriad military applications. It will also develop credible vision vehicles for specific functions such as an expendable weapon, ISR/Strike reusable platform, on-demand access to space, etc. The team will also conduct conceptual design of such vehicles including analysis of all major subsystems, with sufficient fidelity for quick initial sizing and basic functionality assessment. By applying modern Multidisciplinary Analysis & Design Optimization (MDAO) techniques, the team will advance the state of the art in hypersonic system design. Variable fidelity analyses of hypersonic systems will allow the team to assess design feasibility against specific Air Force requirements. This presentation will focus on the approach, applications, and initial results of this effort. Distribution A: Approved for public release: distribution unlimited. Case Number: 88ABW-2017-0239.

Abstract ID: 42DCASS-118

Distributed Lift: An Unconventional Approach

Muhammad Omar Memon (mmemon1@udayton.edu)
University of Dayton
Daniel Kowalski
University of Dayton
Aaron Altman
University of Dayton


Conventionally configured aircraft wings are difficult to manufacture, transport, and install on aircraft due to their sheer size. Additionally, at airport terminals, much of the space that a jet occupies is due to its large wingspan. Since aerodynamic lift is a strong function of the wing surface area, the size of the wing planform is determined based on the amount of lift desired. Theoretically (Reynolds scaling aside), the amount of lift produced by the mono-wing of an aircraft should remain the same if that mono-wing area was divided into many smaller wings placed along the fuselage. These mini-wings can be easily mass-produced, can have higher effective aspect ratio, can be made lighter, can improve maneuverability, and can improve damage tolerance. Preliminary investigation of the lift and drag forces on a fuselage with 200 rectangular wings each of aspect ratio 4 is conducted in this research. The wings are placed at a distance of two chord lengths apart along the fuselage to minimize the interference in the flow of the preceding wings. Lift and drag from the unconventional design obtained from the University of Dayton Low Speed Wind Tunnel will be compared to those of a traditional AR (Aspect Ratio) 4 rectangular mono-wing.

Abstract ID: 42DCASS-129

Trajectory Trades of Reusable Hypersonic Vehicle Design

Malia Stephens (malia.stephens.1@us.af.mil)
Air Force Research Laboratory


The US Air Force is aiming to develop a reusable hypersonic testbed to mature a wide range of technologies. In this presentation, the conceptual design for a hypersonic testbed and the results of a trajectory trade study of initial launch conditions will be presented. Trajectories of the hypersonic vehicle design were modeled using the One-Variable-At-A-Time (OVAT) approach. Variables traded include release altitudes, ignition time delays, thrust-to-weight factors, release speeds, and release flight path angles from an air-dropped condition. The trajectories were run with QuickShot, a trajectory tool developed by SpaceWorks Enterprises, Inc. and used to perform trajectory optimization and trade studies, expected casualty calculations, and survivability analyses. The trajectory data produced from the various initial conditions—including altitude, downrange, time, angle of attack, dynamic pressure, Mach number, and heat rate—was plotted and compared.

Abstract ID: 42DCASS-168

Turboelectric Distributed Propulsion System for a Next Generation Aircraft 

Hashim Abada (abada.2@wright.edu)
Wright State University
Rory Roberts
Wright State University
Mitch Wolff
Wright State University


 In next generation aircraft, advanced propulsion systems will be required which are significantly different than current aerospace systems. One concept is the replacement of the traditional turbofan engines with a series of embedded electrical fans. In addition to improved aircraft efficiency, this propulsion change will significantly reduce noise generation and providing the capability of short take-off and landing. The NASA proposed blended wing aircraft will have as many as 14 electric fans mounted on the upper aft surface of the aircraft wing. The primary function of these fans is to generate aircraft thrust with additional performance utilizing boundary layer ingestion to reduce fuel consumption and filling in the wake generated by the airframe with the thrust steam. The electrical power used to drive these electric fans will be generated by two wingtip turbo generators. A Simulink model of the electric fans will be presented based on propulsion fundamentals. The propulsion system will consist of a fan (low pressure compressor), power shaft, nozzle, and superconductive motor.  

Abstract ID: 42DCASS-171

Electrical Power-Generation for a NASA Next Generation Aircraft

Saif Alagele (al-agele.2@wright.edu)
Wright State University
Rory Roberts
Wright State University
Mitch Wolff
Wright State University


Next generation aircraft will need more an electrical power-generation. The power- generation used in this type of aircraft will be very high, for this reason a unique design is considered. The components of a typical design are very much like a turbojet engine, although their are differences between them. It has an inlet bell replaced with a diffuser in turbojet, a compressor, a combustor, a turbine, and an exhaust replaced with a nozzle. The turbine in this model has a high-pressure stage, low- pressure stage, and free stage. The HPT and LPT drive the compressor similar to a turbojet engine. The free stage in the turbine will be drive an electrical generator. It can be set up on the main shaft or an auxiliary shaft. This generator provides much of the needed aircraft power, with additional power provided thru Li- ion batteries. This model will be used as part of the power system for a next generation blended wing commercial aircraft with substantially greater efficiency proposed by NASA.

Abstract ID: 42DCASS-176

Aerodynamic Analysis of a Blended Wing Body Aircraft utilizing Electric Propulsion

Jay Abhilash Vora (vora.10@wright.edu)
Wright State University
Rory Roberts
Wright State University
Mitch Wolff
Wright State University


NASA has proposed a next generation commercial aircraft, which utilizes a Blended Wing Body (BWB) and Turboelectric Distributed Propulsion (TeDP) to improve performance of both fuel economy and noise. An array of electric fans are attached to the aft of the fuselage of the BWB. The fans have a continuous nacelle that ingests the blended wing boundary layer. CFD analysis of the effect of Boundary Layer Ingestion (BLI) and the aerodynamics of the BWB on the aircraft’s basic six -Degree of Freedom (6-DoF) aerodynamic model is required in the development of a simplified vehicle model which can used in an integrated power/thermal preliminary design analysis. The aerodynamic analysis will be accomplished using Cradle’s SC-Tetra CFD software for various flight parameters (Mach number, altitude, angle of attack, etc.). As an initial step towards this CFD analysis, the lift and drag on various NACA airfoils and 3D wings will be presented.

Abstract ID: 42DCASS-178

Modeling of a Fin-Plate Heat Exchanger for use in a NASA Next Generation Aircraft

Hayder Al-Sarraf (al-sarraf.2@wright.edu)
Wright State University
Rory Roberts
Wright State University
Mitch Wolff
Wright State University


NASA has been working on the design of a hybrid wing body aircraft that consumes less fuel and reduces pollution. They have proposed a next generation of aircraft called N3-X. One of the significant challenges of the N3-X aircraft is to manage the thermal loads on the board of the aircraft to reduce fuel burn. Therefore, the necessity an innovative heat exchanger is required. In this presentation, a fin–plate heat exchanger will be used to handle two types of fluids; kerosene and R-134a. Selection of the heat exchangers involves weight, efficiency, and cost. A Simulink based model will be presented. Eventually, the entropy destructed during heat exchange process will be calculated. In addition, fluids temperatures coming out of the heat exchanger will be computed.

Aviation History

Abstract ID: 42DCASS-192

The Proposed Southwest Ohio Regional Jetport, 1961-1962

Janet Bednarek (jbednarek1@udayton.edu)
University of Dayton


In the late 1950s and early 1960s the FAA pushed hard for the creation of new facilities – so-called jetports – to serve the new jet airliners entering service. The FAA envisioned these new facilities as serving metropolitan regions, rather than individual cities. The FAA waged its most fierce campaign in the Dallas-Fort Worth metropolitan area, but it also supported the creation of such facilities in many other parts of the country, including southwest Ohio. Reacting to the opening of Chicago’s new airport (O’Hare) in 1959, civic leaders including Dayton Mayor R. William Patterson called for the creation of an 11-county airport authority to build and manage a new jetport to serve the Dayton-Cincinnati metropolitan area in order to compete with Chicago. Though the state government in Ohio supported the effort, political fragmentation and resistance from both Cincinnati and Northern Kentucky (which housed the commercial airport serving the Cincinnati area) scuttled the effort.

Abstract ID: 42DCASS-198

McCook Field, at the Forefront of Twentieth Century Parachute Technology Development

Andrew Kididis (akididis@yahoo.com)
Air Force Life Cycle Management Center


A small group of engineers and technicians at McCook Field led the way in the development of the first practical in-flight aircraft escape capability through the development of the free-fall manual bailout parachute. Their work would set the standard for the bailout parachutes produced in the following decades that would save thousands of lives. They were innovators in all aspects from design and materials to testing and documentation. In addition to the work on the emergency personnel parachute, which was their original charter, they also explored other parachute applications such as aircraft recovery and aircraft landing deceleration. Their work influence a whole industry and it is fair to say that the technology and the personalities associated with the parachute work at McCook Field formed the foundation of the parachute industry in the United States for several decades to come.

CFD Applications

Abstract ID: 42DCASS-045

Optimization of a Scramjet Engine Using a Simplified CFD Model

Nate McGillivray (mcgillivray.4@wright.edu)
Wright State University
Rory Roberts
Wright State University
Mitch Wolff
Wright State University
Mark Hagenmaier
Air Force Research Laboratory
Dean Eklund
Air Force Research Laboratory


Various aspects of hypersonic vehicles are being rapidly explored for improved functionality. One of the main areas of consideration is the fueling of a Supersonic Combusting Ramjet (scramjet) engine. Using Computational Fluid Dynamics (CFD), computer simulations can be performed to analyze the flow physics of a given scramjet geometry. In addition, an optimization code, Dakota, is integrated with CFD to optimize a set of parameters for maximum thrust. Initially, the fuel injection and combustion is replaced with heat sources. Additionally, the 3D geometry is reduced to an axisymmetric 2D geometry. With this simplified model, the optimization and CFD algorithm is utilized to optimize the flow heat addition for maximum thrust. Different optimization methods have been explored to reduce computational times. A genetic algorithm was selected because of its robust abilities. Additionally, a sampling algorithm was selected because of its abilities to explore the whole design space and to enable sensitivity studies. Knowledge of the optimized heat distribution will assist in the optimization of fueling splits and injector locations for a more detailed combustion investigation in which similar optimization techniques can be applied.

Abstract ID: 42DCASS-087

Computational Prediction of HYMETS Arc-Jet Flow with KATS

Umran Duzel (umrandzl@gmail.com)
University of Kentucky
Olivia Schroeder
University of Kentucky
Alexandre Martin
University of Kentucky


Planetary entry is a challenging task as it involves extreme conditions that must be predicted. Investigation of hypersonic flow environments not only requires ground testing, but also needs models to better interpret the conditions. Therefore, high fidelity simulation tools are needed to predict the entry environment. The Kentucky Aerothermodynamic and Thermal-response Solver (KATS) is developed mainly to predict planetary entry conditions. In order to investigate and assess the arc-jet flows, KATS is used to compare experimental data obtained at the Hypersonic Material Environment Testing System (HYMETS) Arc-Jet facility. The ground testing case selected has a 6.5 MJ/kg enthalpy nozzle flow exiting into a vacuum test chamber at 2.28 mbar. The radial and axial velocity profiles which have been measured by using Planar Laser Induced Fluorescence (PLIF) are compared with KATS simulations.

Abstract ID: 42DCASS-102

Multiphase flow simulation using moment of fluid method

Yongsheng Lian (yongsheng.lian@louisville.edu)
University of Louisville


In this talk we will present the recent development of a new multiphase flow solver based on the moment of fluid method. The method is similar to the popular volume of fluid method but it considers the centroid information in the interface reconstruction. Numerical tests have shown that the new method can better capture the interface and also has advantages over the volume of fluid method when more than 2 materials are present. We will present recent simulation results in drop impact on hydrophobic and super-hydrophobic surfaces, drop impact on wet surfaces at both low and high speed, and shock/drop interactions.

Abstract ID: 42DCASS-120

One-Dimensional Model of Lithium-Ion Battery

Ashwin Borakhadikar (borakhadikar.2@wright.edu)
Wright State University
James Menart
Wright State University


As battery energy storage may be applied to a number of situations that differ in power and energy requirements, modeling of battery performance is required. This work involves the development of a one-dimensional computer model of Li-ion batteries which consists of 3 domains: the negative electrode, the separator, and the positive electrode. A finite volume technique is used to solve four governing, partial differential equations which simulate the dynamics of Li-ion batteries. These equations are conservation of charge in the solid electrodes, conservation of charge in the electrolyte, conservation of species in the solid particles, and conservation of species in the electrolyte. Another important equation included in this model is the Butler–Volmer kinetic equation. The solution of these equations is carried out with a MATLAB computer program. Initial results will be presented that show the effects of different charging and discharging conditions on the performance of Li-ion batteries.

Abstract ID: 42DCASS-143

1D-3D Multi-scale Computational Modeling for Blood Flow in Intracranial Aneurysms and Stenoses

Hongtao Yu (yu.41@wright.edu)
Wright State University
George P. Huang
Wright State University
Zifeng Yang
Wright State University
Bryan R. Ludwig
Miami Valley Hospital


It is believed that hemodynamic characteristics play significant roles in multiple intracranial vascular diseases. Recent computational fluid dynamics (CFD) studies based on patient-specific models frequently adopt a direct measurement of the flow and pressure waveforms from a patient or health person as input boundary conditions (BCs) for three-dimensional (3D) simulations. However, as an morphology develops, the feedback from this topological change to the alterations of BCs cannot be realized, and thus the impact on flow characteristics within the morphology is only computed locally. In the present work, a novel one-dimensional (1D) numerical model with the entire human cardiovascular network has been developed to compute the global hemodynamics. Meanwhile, a patient specific 3D morphology model is coupled with the 1D model. The proposed 1D-3D multi-scale modeling enables the investigation of the global hemodynamic changes corresponding to the local morphological effects. Validation studies were implemented by comparing the flow and pressure waveforms with experimental results, and complete 3D numerical modeling data reported in previous studies. After validation, the multi-scale modeling has been successfully applied to studies of a patient-specific aneurysm model and an internal carotid artery (ICA) stenosis model. It is demonstrated that the multi-scale modeling is able to accurately quantify the flow characteristics in the morphologies, and the changes in the flow shunting due to the local morphologies. Therefore, this multi-scale modeling is capable of providing more accurate BCs for studying intracranial vascular diseases such as aneurysm and stenosis.

Abstract ID: 42DCASS-152

Performance analysis of open and ducted wind turbines

Tariq Khamlaj (khamlajt1@udayton.edu)
University of Dayton
Markus Rumpfkeil
University of Dayton


In this talk, a computational approach, based on the solution of Reynolds-averaged-Navier–Stokes (RANS) equations, to describe the flow within and around a diffuser augmented wind turbine (DAWT) is reported. In order to reduce the computational cost, the turbine rotor is represented through a CFD-integrated blade element momentum model (Actuator disk model), employing realistic rotor data, chord length, twist angle as well as the airfoils lift and drag coefficients at different angle of attacks. Comparison with experiments carried out in similar conditions shows a good agreement suggesting that the adopted methodology is able to predict the performance accurately.

Abstract ID: 42DCASS-158

Modelling of Flame Acceleration due to Wall Friction in a Dusty-Gaseous Environment with Various Dust Distributions

Torli Bush (tabush@mix.wvu.edu)
West Virginia University
Sinan Demir
West Virginia University
Hayri Sezer
West Virginia University
V'yacheslav Akkerman
West Virginia University


While combustion of gaseous fuels and of dust is studied reasonably well, flame propagation in a combined dusty-gaseous environment, especially with a non-uniform dust distribution in the gas, still needs to be scrutinized both form fundamental viewpoint and practical perspectives. The latter includes, in particular, accidental methane-air explosions in the presence of coal dust, which frequently occur in coalmines and claims hundreds of lives every year. To reduce the risk of such disasters, a fundamental physical understanding of the dusty-gaseous combustion process is critically needed as an ultimate goal. This study is a step in this direction, which is being undertaken through the comprehensive computational simulations of the hydrodynamic and combustion equations by means of the fully-compressible Navier-Stokes solver adapted for parallel computations. The combustible coal dust particles are incorporated into the original CFD platform by means of the classical Seshadri formulation. Specifically, a real dusty-gaseous environment is replaced by an “effective” gaseous fluid with locally-modified, dust-induced flow and flame parameters. Flame propagation in pipes resembling the coalmine tunnels geometry is considered for various coal dust distributions: (a) homogenous, (b) linear, (c) cubic and (d) parabolic. As a result, the evolution of the flame shape and propagation velocity is tabulated as a function of various thermal-chemical flame parameters, and it is identified how non-uniform dust distributions promote and/or moderate flame acceleration caused by wall friction. Keywords: premixed flame acceleration; dusty-gaseous combustion; computational simulations.

Abstract ID: 42DCASS-163

Detached-Eddy Simulation of a Supersonic Reattaching Shear Layer

Tim Leger (leger.2@wright.edu)
Ohio Aerospace Institute
Nicholas Bisek
Air Force Research Laboratory
Jonathan Poggie
Purdue University


Simulations of a Mach 2.9 turbulent shear layer in a back-step/ramp configuration were performed using the delayed detached eddy simulation approach, and compared to a series of experiments carried out at Princeton University. Available experimental data include shear layer and reattaching boundary layer profiles, mean and fluctuating pressure, and mean skin friction. To reduce computational cost, only a third of the width of the configuration was simulated, and periodic spanwise boundary conditions were employed. A very careful study of spatial and temporal resolution was carried out. The computations predicted a greater flow turning angle at separation than measured experimentally. Accounting for the consequent shift in reattachment, the mean flow properties were predicted fairly accurately. The computations predicted a higher intensity of pressure fluctuation near reattachment than measured experimentally, but predicted the form of the pressure fluctuation spectrum fairly accurately down to about 20 Hz. The discrepancies between computation and experiment may be the result of three-dimensionality in the cavity flow and fluctuations in the incoming boundary layer that are not accounted for computationally. The successful prediction of the low-frequency pressure fluctuations may have applications in understanding the effect an engine exhaust jet reattaching onto an aircraft surface.

Abstract ID: 42DCASS-173

Evaluating the Influence of Vessel Wall Thickness and Blood Rheology on Hemodynamic and Mechanical Variables in Turner Syndrome

Dhananjay Radhakrishnan Subramaniam (subramdy@mail.uc.edu)
University of Cincinnati
Ephraim J. Gutmark
University of Cincinnati
Iris Gutmark-Little
Cincinnati Childrens Hospital


Aortic dissection in chromosomal abnormalities such as Turner syndrome (TS) is influenced by the dynamics of blood flow and biomechanics of the compliant aorta. The present study investigates hemodynamic and mechanical variables in a diseased aorta, using fully coupled flow-structure interaction (FSI) simulations. Anatomically accurate aorta geometry for a TS patient with moderate ascending aortic dilatation was reconstructed using magnetic resonance (MR) scans. A mean cardiac output of approximately 5 liters per minute and a pulse pressure corresponding to 55 mmHg was employed in this study. Blood was assumed to behave as a non-Newtonian fluid with a density of 1060 kg/m3. The shear-thinning behavior was modeled using a Carreau rheological model. The mechanical response of the vessel wall was approximated using a nearly incompressible, isotropic Arudda-Boyce material model. The shear modulus and limiting network stretch of the aorta wall were assumed to be 1 MPa and 1.01 respectively. An in-house mesh generation code was employed to model and discretize the local vessel wall thickness as a function of the corresponding lumen radius. Steady simulations of blood flow in a compliant aorta were performed using a k-ω SST turbulence model and a turbulence intensity of 2 percent. Forces arising from blood flow were passed to the vessel wall to compute displacements and the resulting displacements were relayed back to update the flow variables. The simulations were compared to the case involving a constant vessel wall thickness and Newtonian fluid assumption. Results for the flow model indicated that wall shear stress (WSS) distribution in the aorta was comparable for the Newtonian and non-Newtonian models (less than 5 percent difference). Mechanical stresses in the transverse and descending aorta were higher for the case involving variable wall thickness. These observations indicate that a Newtonian model could potentially predict hemodynamics in the TS aorta with reasonable accuracy, at least for the flow conditions tested. Results for the structural model indicate that the rupture potential of the TS aorta could be significantly influenced by variation in arterial wall thickness along the vessel length.

CFD Methods

Abstract ID: 42DCASS-076

Retrospective Cost Adaptive Reynolds-Averaged Navier-Stokes k-omega Models for Unsteady Turbulent Flow

Zhiyong Li (zli252@g.uky.edu)
University of Kentucky
Sean C.C. Bailey
University of Kentucky
Jesse B. Hoagg
University of Kentucky
Alexandre Martin
University of Kentucky


Turbulent flow arises in a vast array of aerospace technologies ranging from flow around launch vehicles to internal flow within propulsion and life-support systems. Thus, computational techniques for accurate turbulent-flow simulation can advance numerous aerospace technologies. We present a new data-driven adaptive computational model for simulating unsteady turbulent flow, where partial-but-incomplete measurement data is available. The model automatically adjusts the closure coefficients of the unsteady Reynolds-averaged Navier-Stokes (URANS) k-omega turbulence equations to improve agreement between the simulated flow and the measurements. A key enabling technology is retrospective cost adaptation (RCA), which was developed for real-time adaptive control technology but is used in this work for data-driven model adaptation. RCA has been successfully validated on numerous control applications that have significant transient behavior, which suggests that RCA is well suited for adaptation with unsteady flow. The new RCA-URANS k-omega model is verified on a steady test case (pipe flow) as well as 2 unsteady test cases: vortex shedding from a surface-mounted cube and flow around a square cross-section cylinder. The results demonstrate that the k-omega closure coefficients can be updated such that the periodicity in the simulated flow matches that of the measurement data. Thus, the RCA-URANS k-omega model effectively adapts to unsteady measurement data.

Combustion

Abstract ID: 42DCASS-002

Evaluating the Viability of Planar-Laser Induced Fluorescence to Determine the Constituents of Exhaust Plumes

Christine Robinson (christine.robinson.5@us.af.mil)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-071

Propagation Velocities for Neighboring Triple Flames

Stephen Grib (stephen.grib@uky.edu)
University of Kentucky
Michael W. Renfro
University of Kentucky


Triple flames propagate through mixtures faster than the laminar flame speed due to streamline divergence ahead of the flame base. When multiple triple flames are in close proximity, the bulk propagation velocity of the structure can be even faster due to additional streamline divergence. Practical turbulent flames in partially-premixed conditions encounter these situations, where multiple stoichiometric crossing are in close proximity, leading to multiple triple flames being formed. Individual triple flame propagation velocities has been characterized in previous studies, however similar characterization of neighboring triple flames velocities has not been conducted. The present work utilizes a laminar five slot burner, which allows both the concentration gradients and stoichiometric separation distance of two interacting triple flames to be varied. The bulk propagation velocity has been characterized based on distance between the flame base (or stoichiometric) locations and curvatures in order to better understand the conditions which lead to larger streamline divergence. Hydroxyl planar laser induced fluorescence and particle image velocimetry were used to obtain both flame curvature and velocity information.

Abstract ID: 42DCASS-070

NOx Formation in Light-Hydrocarbon, Premixed Flames

Rob Hughes (rthughes94@gmail.com)
University of Kentucky
Jose Grana-Otero
University of Kentucky


This study explores the formation of NOx pollutants in light-hydrocarbon, premixed flames. The goal is to uncover the main production/consumption paths for these hazardous species. To do this, steady planar flames with various equivalency ratios were calculated numerically. It is shown that since these species participate in very small concentrations, the steady state approximation is accurate in most of the flame structure, except perhaps in the preheat region. This is a considerable simplification that permits to obtain, explicitly, algebraic relationships among the concentrations of NOx species and of the temperature. In addition, it turns out that only a very reduced number of reactions, especially close to the lean flammability limit, are enough to accurately represent the exact distributions. These results yield simplified flow maps revealing clearly and quantitatively the main reaction paths. Future work includes exploiting these results in order to reduce NOx emissions.

Abstract ID: 42DCASS-108

Effect of Rayleigh-Taylor Instability on Turbulent Flame Speed in a Backward-facing Step Curved Duct

Brandon Long (Brandon.long.14.ctr@us.af.mil)
University of Dayton Research Institute
Alejandro Briones
University of Dayton Research Institute
Scott Stouffer
University of Dayton Research Institute
Brent Rankin
Air Force Research Laboratory


Rayleigh-Taylor instabilities induced by centrifugal forces can be used as a mechanism to enhance turbulent fluctuation and flame wrinkling, which in turn enhances turbulent flame speeds. Faster fuel consumption rates translate to shorter residence times and more compact combustors. A numerical investigation of propane-air flowing through a U-shaped centrifuge duct is conducted using a customized commercial code. This research is motivated by the desire to reduce the size of modern combustors for improving engine thrust-to-weight ratios. Efficient compact combustion systems are also needed to enable reheat cycles which involve inter-turbine burners positioned between the high- and low-pressure turbines. A backward facing step on the outer wall is used to stabilize the flame. The U-duct is a geometry that has been used previously to study the effects of buoyancy-like forces on reacting flows. This curved channel with a backward stabilizing step induces a large centrifugal acceleration on the flow which can be varied by changing geometry or flow conditions. The centrifugal force acts from the high-density reactants towards the low-density products creating a Rayleigh-Taylor instability. The instability causes significant turbulent mixing of the reactants and products. Transient, three-dimensional simulations are performed using Large Eddy Simulation (LES) with the sub-grid kinetic energy transport closure model. The chemical kinetics are modeled using two-step global chemistry. The inlet flow velocity is varied while the equivalence ratio, inlet temperature, and exit pressure are maintained constant. The flame front will be tracked using progress variable gradient. Here the turbulent flame speed and centrifugal force will be computed in order to determine the local and transient effect of Rayleigh-Taylor instability on turbulent flame speeds. The turbulent flame speeds will be locally computed as the ratio of the fuel consumption rate and the product of the unburned density and the flame sheet normal gradient magnitude. In addition, the global fuel consumption rate will be computed as well as the net centrifugal force to also quantify the net effect of the latter on the former. The results are expected to indicate that as the inlet flow velocity is increased, Rayleigh-Taylor instability increases with increasing centrifugal acceleration, which, in turn, will promote faster turbulent flame speeds and fuel consumption rates.

Abstract ID: 42DCASS-109

Mechanism of Valved Pulsejet Operation

Justas Jodele (jodelejb@mail.uc.edu)
University of Cincinnati
Vijay Anand
University of Cincinnati
Ethan Knight
University of Cincinnati
Ephraim Gutmark
University of Cincinnati


An experimental analysis of valved pulsejets and the concomitant results is to be discussed. By altering the combustor length, the tail pipe length and by adding a flare at the aft-end, twelve different pulsejet configurations are tested. An axially-distributed array of piezoelectric pressure sensors and ionization probes reveal the fluid and combustion dynamics inside these devices. Evidence is attained to support the claim that a valved pulsejet behaves like a Helmholtz resonator. Each cycle of a pulsejet is composed of a compression wave, an expansion wave, and temporally and spatially restrained combustion events. Altering the geometry also induces an amplitude modulated low frequency instability inside the pulsejet that is characterized by sinusoidally-varying peak cycle pressures. The operating frequency, peak pressures and combustion activity of the pulsejets are characterized to reveal interesting features of valved pulsejet operations that have been hitherto unknown. A mechanism of valved pulsejet operation is proposed.

Abstract ID: 42DCASS-110

Rotating detonations through ethylene-air mixtures in a hollow combustor

Vijay Anand (ganeshvn@mail.uc.edu)
University of Cincinnati
Andrew St. George, Ephraim Gutmark
University of Cincinnati


A hollow rotating detonation combustor operating with ethylene-air mixtures is to be discussed. Three air flow rates and diverse equivalence ratios are tested. Rotating detonation waves are produced in the hollow combustor at certain conditions of air flow rate and equivalence ratio. Transverse high frequency combustion instabilities are also produced depending on the flow rate. At higher flow rates, the rotating detonation waves exhibit highly unstable propagations that are characteristic of stuttering and galloping detonations observed in decoupled planar detonations. The results of this study marks the first time in known literature that rotating detonation waves are produced through a non-premixed ethylene-air mixture.

Abstract ID: 42DCASS-116

Power Loss Pathways and Energy Balance of Two Small Four-Stroke Internal Combustion Engines

Jason Blantin (jason.blantin@afit.edu)
Air Force Institute of Technology
Marc D. Polanka
Air Force Institute of Technology
Paul J. Litke
Air Force Institute of Technology
Jacob A. Baranski
Innovative Scientific Solutions Inc.


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-117

Experimental Characterization of Heat Transfer Coefficients in a Rotating Detonation Engine

Samuel Meyer (samuel.meyer.2@us.af.mil)
Air Force Institute of Technology
Marc D. Polanka
Air Force Institute of Technology
Frederick R. Schauer and Richard J. Anthony
Air Force Research Laboratory


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-140

Effect of Gas Compression on Propane-Air Flames in Industrial Passages

Sinan Demir (sidemir@mix.wvu.edu)
West Virginia University
V’yacheslav Akkerman
West Virginia University


Accidental gas explosions in various industrial and laboratory applications oftentimes result in fatal disasters. Among various factors driving these explosions such as turbulence, acoustics, wall friction and obstacles, combustion instabilities and specific (finger) flame shape play a significant role and constitute the focus of the present work. Specifically, an analytical predictive scenario of premixed propane-air flame propagation in a pipe-like passage is developed and the characteristic stages of the flame evolution are identified. For simplicity, the two-dimensional (2D) planar and cylindrical-axisymmetric configurations are considered. Starting with an incompressible formulation, we subsequently extend the analysis to account for gas compressibility, because the latter inevitably becomes substantial as soon as the flame spreading velocity starts approaching the sound threshold. In particular, the effects of equivalence ratio on the premixed flame evolution are systematically investigated. It is shown that gas compressibility moderates flame acceleration, and its impact depends on the type of the fuel, its thermal-chemical parameters as well as geometry of the problem. While the effect of compression is relatively minor for the fuel-lean and fuel-rich propane-air flames (say, a 5-25% reduction in the flame velocity is observed), thereby justifying the incompressible formulation in that case; it appears significant for near-stoichiometric propane-air combustion (yielding an up to 70% reduction in the flame velocity), and therefore, this effect should be incorporated into a rigorous formulation. In addition, moderation of flame acceleration due to compressibility is shown to be higher in the cylindrical geometry as compared to 2D planar one. In summary, the incompressible and compressible models are compared, with the detailed parametric study performed for various lean and rich propane-air mixtures. Keywords: propane-air flame; expanding or finger flame; combustion instability; flame acceleration industrial pipe-like passage.

Abstract ID: 42DCASS-156

Temperature Distribution of the Exit Plane from an Ultra Compact Combustor

Edwin Hornedo (edwin.hornedorodriguez@afit.edu)
Air Force Institute of Technology
Marc D. Polanka and Brian T. Bohan
Air Force Institute of Technology
Larry P. Goss
Innovative Scientific Solutions Inc.


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-153

Preliminary Experimental Characterization of Centerbodiless RDE Emissions

William Stoddard (stoddawa@mail.uc.edu)
University of Cincinnati
Vijay Anand, Andrew St. George, Robert Driscoll , Brian Dolan, Rodrigo Villalva , Justas Jodele , Jarred Wilhite, Ephraim Gutmark
University of Cincinnati


Rotating Detonation Engines (RDE’s) or Rotating Detonation Combustors (RDC's) are a detonation based combustion method that shows potential capability of high efficiency, compact geometry, and continuous operation. In order to incorporate an RDC into a larger engine, in is important to check practical issues on a system level. One such issue is the level of pollution generated by the new combustion method. To that end, experiments are being conducted to test the level of emissions generated by an RDE. Of particular interest are nitrogen oxides (NOx). Recent tests of a new centerbodiless RDE configuration, where the inner wall has been replaced with a core of flowing air have found stable operating conditions with an H2-O2 enriched air mixture. A modified version of the RDE run as a C-RDE proof of concept with a water jacket has been created and run for a preliminary test, run on Hydrogen and Oxygen enriched air at detonating conditions. A high speed PCB pressure probe is used to confirm wave speed. A water cooled gas probe measures emissions. Other configurations like it will be built and compared for emissions at different mixtures.

Abstract ID: 42DCASS-155

Developing a Calculator for Generating Surrogate Jet Fuels with Target Physical and Chemical Properties

David Bell (belld5@udayton.edu)
University of Dayton
Joshua S Heyne
University of Dayton


In the interest of improving national security and decreasing transportation’s carbon footprint, fuels derived from non-conventional, and in some cases renewable, feedstocks are of escalating interest. These non-conventional, or alternative, fuels have encountered difficulties penetrating the market due to the extensive certification process. A significant reason for this extensive certification process is the uncertainty of novel fuels and their properties on the combustion behavior of combustors. Here we aim to model certain chemical and physical properties, via surrogates, to ideally map these properties to combustor Figure of Merit behavior. Previously, only chemical properties had been considered to mimic the prevaporized combustion behavior of conventional and alternative jet fuels. Additionally, the National Jet Fuels Combustion Program (NJFCP) has focused their efforts towards developing a method for rapid characterization of surrogate jet fuel’s chemical and physical properties to assist in finding these relationships. Here we report on developing surrogate fuels to match target physical and chemical properties to assist in developing these relationships. In this research, a calculator is created to generate surrogate fuels with desired physical and chemical properties using mixing rules and an optimization method for determining the ideal mixture.

Abstract ID: 42DCASS-161

Predicting Combustion Phenomena Based on Random Forest Modelling

Jeremy Carson (carsonj2@udayton.edu)
University of Dayton
Joshua Heyne
University of Dayton


The National Jet Fuels Combustion Program (NJFCP) seeks to streamline the process for certifying alternative jet fuels for use in aircraft engines, a process which is currently plagued with extraneous investments of capital, namely excessive costs and extensive testing with little guarantee of success. This current process creates a disconnect between refineries that can produce new fuels and the aviation industry that would utilize those fuels in their engines. To minimize this disconnect, the NJFCP is looking to map alternative jet fuels’ unique chemical and physical properties to three key Figures of Merit (FOM) that were agreed upon by the program’s Original Equipment Manufacturers (OEM): Cold Start, Lean Blowoff (LBO), and High Altitude Ignition. By correlating the fuels’ properties to the FOMs, the program seeks to shift the fuel certification process from an experimental approach to a predictive modelling approach. To that end, a statistical analysis of experimental data taken across numerous rigs at various institutions sing a random forest model has been proposed. The current results from the current analyses of LBO and preliminary ignition tests show the Derived Cetane Number impacting LBO performance and the Distillation Curve properties having significance to ignition.

Abstract ID: 42DCASS-177

Computational Study of Non-equidiffusive Flame Propagation in Obstructed Channels

Gbolahan Idowu (gidowu@mix.wvu.edu)
West Virginia University
Abdulafeez Adebiyi
West Virginia University
Damir Valiev
Umea University
V’yacheslav Akkerman
West Virginia University


Bychkov et al. [Phys. Rev. E (2008) 164501] have revealed a mechanism of ultra-fast premixed flame acceleration in channels equipped with a comb-shaped array of tightly-packed obstacles. Specifically, the analytical theory of the mechanism has been derived and substantiated by the computational simulations. However, similar to the multitude of theoretical studies, the Bychkov formulation employed the approach of equidiffusive combustion such that the Lewis number characterizing the thermal-to-mass diffusivity ratio was assumed to be unity, Le = 1. At the same time, the hydrocarbon and hydrogen flames are usually non-equidiffusive in the practical reality. To reduce such a gap between the theoretical models and the experimental observations in the present work, we investigate the dynamics and morphology of non-equidiffusive premixed flames in obstructed pipes, by means of the computational simulations of reacting flow equations with fully-compressible hydrodynamics and Arrhenius chemical kinetics. The channels of blockage ratios 0.33~0.67, with the scaled spacing between the obstacles being dz/R = 0.25, and having widths of 48 ≤ D/Lf ≤ 96, where Lf is the flame thickness, are considered. The Lewis numbers are taken in the range 0.2 ≤ Le ≤ 2.0. It is demonstrated that Le plays a key role for combustion in obstructed conduits. Specifically, due to the flame front thickening, a flame accelerates slower for Le > 1 as compared to the Le = 1 flames. In contrast, Le < 1 flames acquire stronger distortion of the front shape and thereby accelerate much faster than the equidiffusive ones. The later effect can be related to the onset of the diffusional-thermal combustion instability.

Abstract ID: 42DCASS-180

Characterization of a Methane-Air Microflame Burner with TaN Seed Particles through Emission Spectroscopy

Zhaojin Diao (zhaojin.diao@uky.edu)
University of Kentucky
Michael Winter
University of Kentucky


Three microflame burner designs developed for flame synthesis, consisting of six fuel nozzles in a concentric configuration surrounding a seventh nozzle feeding an air/particle stream into the complex flame structure were characterized with optical methods. Three different pitch sizes were investigated while operated with methane and air at atmospheric pressure. Temperatures of tantalum nitride (TaN) seed particles were determined from thermal emission of under different fuel-air flow rates conditions. The analysis is currently being extended to the emission of C2 and H2O emission. Vertical profiles of particle temperatures were measured by comparing the spectral shape of the continuum radiation between 600 nm and 1,000 nm with Planck radiation. Pre- and post-test analyses of the seed particles were conducted with Scanning Electron Microscopy (SEM), Energy dispersive X-ray (EDX) and X-ray photoelectron spectroscopy (XPS) to monitor changes of the particle morphology and surface condition. It was found that the particle surface undergoes oxidation to Ta2O5 while passing through the flame. From a comparison of the spectral shape of the measured data with theoretical Planck emission it was concluded that this layer is thinner than 10 nm since no influence of the characteristic spectral signatures of Ta2O5 transmission were seen in the continuum spectra.

Abstract ID: 42DCASS-181

Modelling of Flame Propagation in Channels with Nonslip Walls: The Impacts of the Lewis Number and the Thermal Expansion Coefficient

Swathi Reddy Shetty (swreddyshetty@mix.wvu.edu)
West Virginia University
Sinan Demir
West Virginia University
Damir Valiev
Umea University
V’yacheslav Akkerman
West Virginia University


Equidiffusive combustion is a conventional approach, which is widely employed in the analytical and computational studies. However, the real laboratory and industrial premixtures are usually fuel-rich or fuel-lean, being thereby strongly non-equidiffusive. The standard dimensionless measure of non-equidiffusivity is the Lewis number (Le) defined as the thermal-to-mass diffusivity ratio. To bridge such a gap between the theoretical and experimental studies, in the present work we analyze the effect of Le on flame propagation in pipes by means of numerical simulations with fully-compressible hydrodynamic and Arrhenius chemical kinetics. The roles of the thermal expansion coefficient (the density drop at the front),a, and the Reynolds number associated with flame propagation, Re = R/PrLf (where Lf is the thermal flame thickness and R the channel radius) are also investigated. Specifically, we employ the quantities a, Le and Re in the ranges 5 ≤ a ≤ 8, 0.2 ≤ Le ≤ 2.0, and 5 ≤ Re ≤ 30, respectively, and quantify various flame propagation regimes in the a-Le-Re space. The thermal boundary conditions include both adiabatic and isothermal walls. Mechanistically, the walls are nonslip. It is shown that flames generally tend to accelerate exponentially in channels with adiabatic, nonslip walls, although acceleration moderates to a linear trend for small a. In isothermal channels, in contrast, flames spread steady or near-steady or may even extinguish. A non-unity Lewis number influences the flame dynamics notably in both adiabatic and isothermal channels. Specifically, Le < 1 flames get extra strong corrugation due to the diffusional-thermal instability and thereby spread much faster than Le = 1 ones. In contrast, Le > 1 flames are thickened and propagate slower.

Abstract ID: 42DCASS-182

Different Regimes of Premixed Flame Propagation in Open Obstructed Pipes

Abdulafeez Adebiyi (abadebiyi@mix.wvu.edu)
West Virginia University
Damir Valiev
Umea University
V’yacheslav Akkerman
West Virginia University


It is known that premixed flames accelerate extremely fast and may even trigger detonation when propagating in semi-open obstructed pipes (one end of a pipe is closed; the flame is ignited at the closed end and propagates towards the open one). However, industrial and laboratory conduits oftentimes have both ends open, or vented, with a flame ignited at one of these ends. The latter constitute the focus of the present work. Specifically, premixed flame propagation through a tooth-brush-like array of obstacles, in-built in a pipe with both ends open, is studied by solving fully-compressible hydrodynamic and combustion equations with Arrhenius chemical kinetics. The computational approach employs a cell-centered, finite-volume numerical scheme, which is of 2nd-order accuracy in time, 4th-order in space for convective terms, and of 2nd-order in space for diffusive terms. For simplicity, the pipes are imitated by the two-dimensional (2D) planar Cartesian channels of half-width R. The pipes of three different R (R/Lf =12, 24 and 36, where Lf is the thermal flame thickness) are considered, with three various blockage ratios, BR = 1/3, 1/2 and 2/3, for each R. In relatively narrow pipes, R/Lf =12 and 24, the oscillations of the burning rate are observed for all BR which conceptually differs from fast flame acceleration in semi-open pipes. Such a difference is devoted to the fact that while the entire new gas volume, generated in the pockets between the obstacles, is pushed towards a single exit in a semi-open pipe, in an open-open one, it is distributed between the upstream and downstream flows, thereby moderating flame propagation. The flame oscillations are non-linear in all cases, and the non-linearity is stronger for larger BR. The oscillation period increases with BR, the oscillation amplitude is almost BR-independent, and the average burning rate reduces with BR. The oscillations are steady for BR = 2/3 and 1/2, while they slightly damp for BR = 1/3. This oscillating regime resembles the flame pulsations observed previously in open-open unobstructed pipes. Moreover, this result also supports the recent experiments and modelling of flames in open-open obstructed pipes, as well as our latest viscous theoretical formulation on flame dynamics in open-open obstructed pipes, which all show steady or quasi-steady flame propagation prior to the onset of flame acceleration. Indeed, the oscillations can be treated as the fluctuations around a quasi-steady solution. It is also noted that in a wider pipe, R/Lf =36, the oscillations are terminated soon, being subsequently replaced by flame acceleration. The latter regime resembles that in semi-open pipes, although acceleration in open-open pipes is much weaker. The last result also supports our recent inviscid theoretical formulation on flame acceleration in open-open obstructed pipes.

Experimental Applications

Abstract ID: 42DCASS-059

Investigation of Dynamic Store Separation out of a Cavity Utilizing a Low Speed Wind Tunnel

Drew Bower (andrew.bower@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-085

Process of Providing a Small-Payload Return from International Space Station

James Sparks (devin.sparks@uky.edu)
University of Kentucky
Evan C. Whitmer, Gabriel I. Myers, Courtney C. Montague, Suzanne W. Smith, and Alexandre Martin
University of Kentucky


The design of an efficient Thermal Protection System (TPS) remains one of the most challenging tasks of planetary exploration missions. Because of the harshness of re-entry environments, no ground tests can replicate these conditions, and engineers must rely on numerical models, which often lack validation data. In order to provide a path toward inexpensive validation, the KRUPS spacecraft is being designed at the University of Kentucky. The current project intends to release a KRUPS spacecraft from two separate rockets into microgravity atmosphere, and obtain data for the TPS. These launches consist of an important step toward the next phase, releasing the spacecraft from the International Space Station (ISS). These launches will raise the Technical Readiness Level (TRL) to a TRL 6 by demonstrating data acquisition, communication and TPS designs, which is required to release from the ISS.

Abstract ID: 42DCASS-090

Comparisons of Measured and Modeled Aero-thermal Distributions for Complex Hypersonic Configurations

Denton Sagerman (sagermand1@udayton.edu)
University of Dayton
Markus P. Rumpfkeil
University of Dayton
Nastasija Dasque
Air Force Research Laboratory
Barry Hellman
Air Force Research Laboratory


The ability to quickly and accurately predict the thermal signature of a complex geometry is important in the early design stages for any aircraft. Due to the lack of hypersonic facilities with this capability, a recent effort has been made to quantify the ability of the Mach 6 tunnel at Wright-Patterson Air Force Base (WPAFB) for this task. The Mach 6 High Reynolds Number Facility at WPAFB in Dayton, Ohio, has been non-operational for the past twenty years, but a recent resurgence in the need for hypersonic test facilities has led to the reactivation of the tunnel. With its restoration, the facility is to include new capabilities to assess hypersonic aero-thermodynamic effects on bodies in Mach 6 flow. Using temperature sensitive paint (TSP) and three complex geometries commonly used in the hypersonic community, experimental tests were conducted inside the Mach 6 tunnel to capture the temperature contours and some pressure data for these geometries at various angles of attack. These results were then compared to numerical analyses conducted using the panel code CBAero, the Euler code Cart3D, the coupled Euler/Boundary layer solver UNLATCH, and Navier-Stokes solutions from FUN3D. Due to the experiments in the tunnel never reaching steady state since paint adherence was affected after about 10 seconds in the high-speed flow, the comparison to steady numerical analysis proved difficult. As a result, the capabilities of the Mach 6 tunnel, in terms of having a quantifiable measure between the experimental and numerical temperature distributions, could not be assessed and instead general qualitative comparisons were made.

Experimental Methods

Abstract ID: 42DCASS-057

Performing Frame Transformations to Correctly Stream Position Data on Multiple Hosts

Tom Franco (francotg@mail.uc.edu)
University of Cincinnati
Hans Guentert
University of Cincinnati
George T. Black
University of Cincinnati


Unmanned Aerial Vehicles (UAV) are starting to become a more common occurrence in today’s society. What started off as highly classified military weapons with little known information, have become part of everyday life for the common individual. That being said, UAV’s still carry a great deal of importance for paving the way for unmanned flight. UAV’s hold major potential for many amazing technological advances within the near future. Drones have become such a common backyard toy for individuals all over the world as well as the way of the future. Major corporations, such as Amazon, are starting to test drones for delivering small packages. Uber has stated that they want to get to the point where cars will be self-driving. It’s crazy to think that if an order from amazon is processed, it could arrive at the desired destination the same day within minutes of being processed. To get to that point, there is a lot to consider. First, and most importantly, the drone has to be self-aware with no sort of human control. The drone also has to be able to communicate and relay its position to not only some form of GPS, but also to a company tracker and the customer as well. How would it go about this? What sort of factors make this possible fantasy of the future a real conceptual idea? The drone has to communicate with numerous devices, properly orientate itself and data being streamed has to be associated with the proper direction in order for the pattern to be flown correctly. Since there are a variety of potential directions for the drone to move, odds are there will be some sort of data conversion involved. This likeliness comes from the fact that the drone is centered around some local coordinate frame and the environment containing the drone will have some other world coordinate frame associated with it. The purpose of this paper is to determine the appropriate transformations required to have the data from multiple sources match up correctly. Once this is achieved, testing can be conducted to collect data and once the data is analyzed, it should be clear to determine if the data conversion has been properly integrated.

Abstract ID: 42DCASS-091

Calibration methodology for a dynamic two strut store balance

Ryan Schmit (ryan.schmit@us.af.mil)
Air Force Research Laboratory
James Grove
Air Force Research Laboratory
Rudy Johnson
Air Force Research Laboratory


Stores in weapons bay cavities may experience significant shaking prior to and through separation to clear the aircraft. The unsteady shear layer over the weapons bay and the self-amplifying aerodynamic feedback mechanism causes the store vibrations. Measuring the store loads and moments inside the weapons bay is difficult due to the induced mechanical vibrations imparted by unsteady aerodynamics forces on the store/sting/balance. This work is attempting to assess the feasibility to remove the structural vibration from the acquired force balance data to eventually improve store trajectory separation prediction.

Abstract ID: 42DCASS-150

A Cleanly Seeded PIV Analysis of Two Geometries in Supersonic Flows

Christopher Hoskins (christopher.hoskins@afit.edu)
Air Force Institute of Technology
Mark Reeder
Air Force Institute of Technology


Awaiting Public Release Clearance.

Facilities

Abstract ID: 42DCASS-111

Design and Achievements of the Modular Polysonic Research Facility at Parks College

Sally Warning (sally.warning@gmail.com)
St. Louis University
Miranda Pizzella
St. Louis University
Mary Jennerjohn Christianer
St. Louis University
Mark McQuilling
St. Louis University


A new wind tunnel has been designed and constructed at Parks College of Saint Louis University. It shares the pressurized air used for the neighboring blowdown supersonic wind tunnel and exhausts to atmosphere. The polysonic facility is modular in its design, allowing for a number of test section sizes and cross section shapes to complete various test objectives at high subsonic and low supersonic Mach numbers including straight and curved flow paths. The first test section employed for testing investigates the interaction between a normal shock wave and boundary layer in a large aspect ratio rectangular cross section. This presentation will cover the design rationale of the wind tunnel components and initial testing achievements.

Abstract ID: 42DCASS-131

Overview of the Plasma Research Facilities at the University of Kentucky

Helmut Koch (helmut.koch@uky.edu)
University of Kentucky
Michael W. Winter
University of Kentucky


The utilization of ground test facilities is crucial in the development process of reliable thermal protection systems for atmospheric entries. However, even high power facilities are not capable of reproducing real-flight conditions, and therefore, an extrapolation of the ground test data to the actual flight situation is necessary, which is achieved through aerothermochemistry simulation codes. In order to generate validation data for these codes, smaller, more flexible, and less expensive facilities are much more suitable than large high power facilities, which are usually very expensive in operation and follow a tight testing schedule. Furthermore, basic research projects, parametric investigations of individual processes, as well as the development of diagnostic methods can be achieved much easier in smaller facilities. For this purpose, three low power plasma sources have been built up in the Department of Mechanical Engineering at the University of Kentucky. The generation of high enthalpy plasma flows is achieved with a 10.5 kW radio frequency (RF) and a 3 kW microwave plasma generator (MWG). While the RF generator is well suited to achieve high temperature, low pressure plasmas which are likely in non-equilibrium, the MWG can work up to atmospheric pressure and typically produces plasmas close to equilibrium. Both systems work electrode-less and are therefore capable of being operated even with highly reactive gases, e.g. pure oxygen or CO2. The plasmas are not intended to reproduce re-entry conditions but to provide a highly controlled environment which can be rebuilt by simulation codes. Targeted near-future applications comprise the investigation of gas-surface interactions with material samples and the characterization of remote recession measurement techniques. Recently, a third plasma generator - an inertial electrostatic plasma generator - has been successfully brought into operation. It is currently used for student education purposes, but as a near-future goal the utilization as a low thrust electric propulsion system for high precision applications will be investigated. The presentation gives an overview on the recent innovations, and first characterization data will be presented to provide a preliminary overview of the operational envelope of the facilities.

Abstract ID: 42DCASS-197

Transient Startup Simulations for a Large Mach 6 Quiet Ludwieg Tube

Joseph Jewell (jjewell@gmail.com)
National Research Council
Thomas J. Juliano
University of Notre Dame


A viscous, two-dimensional axisymmetric time-accurate code is used to produce transient simulations of the startup process for the University of Notre Dame’s planned large hypersonic quiet tunnel at a typical operating condition. In the present simulations, the shutter valve used to start the tunnel is modeled as instantaneously fully-open; this represents the ideal case, as well as nonideal cases for finite-time opening shutter valves of 30, 60, and 120 ms. Three shutter valve positions are simulated, and the position with the valve positioned nearest the throat, in the nozzle contraction, is found to have by far the most steady core flow as well as the shortest startup time. Flow uniformity is also predicted to be good and the expected design performance is obtained in terms of Mach number and pressure.

Fatigue & Fracture

Abstract ID: 42DCASS-021

Time-Dependent Validation of Finite Element Strain Distribution of a Plastically-Deformed Plate via Digital Image Correlation

Kevin Knapp (kevin.knapp@afit.edu)
Air Force Institute of Technology
Anthony Palazotto
Air Force Institute of Technology
Onome E. Scott-Emuakpor
Air Force Research Laboratory
Casey Holycross
Air Force Research Laboratory
Tommy George
Air Force Research Laboratory


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-063

Characterization of the Lode = -1 Meridian on the Al-2024 Failure Surface for *MAT_224 in LS-DYNA

Robert Lowe (rlowe1@udayton.edu)
University of Dayton
Jeremy Seidt
The Ohio State University
Amos Gilat
The Ohio State University


*MAT_224 is an elastic-viscoplastic material model for ductile metals in LS-DYNA that incorporates stress-state-dependent failure. A key ingredient in modeling damage accumulation and ductile fracture within *MAT_224 is the failure surface, a 3-D graphical representation of the equivalent plastic strain at fracture as a function of stress triaxiality and Lode parameter. Aerospace-related ballistic impact experiments used to validate the existing *MAT_224 constitutive and failure models for 2024 aluminum reveal a strong trend of ductile fractures along the Lode = -1 meridian, a region of the failure surface currently underpopulated with experimental data. Exploiting a novel physical interpretation of the Lode = -1 meridian, several new experiments to populate this critical region are proposed and numerically simulated in LS-DYNA, based on simple but clever adaptations of the standard quasi-static hemispherical punch test.

Abstract ID: 42DCASS-185

Fatigue Effects of Laser Shock Peening Over Minimally Detectable Partially-Through Thickness surface Crack

David Eisensmith (david.eisensmith@afit.edu)
Air Force Institute of Technology
Anthony Palazotto
Air Force Institute of Technology
Kristina Langer
Air Force Research Laboratory
Thomas Spradlin
Air Force Research Laboratory
Stefano Coratello
University of Dayton Research Institute


Awaiting Public Release Clearance.

Flight Dynamics & Controls

Abstract ID: 42DCASS-013

Model identification and control of a convertible aircraft

Tudor-Bogdan Airimitoaie (kelly.cohen@uc.edu)
Univ. Bordeaux, IMS-lab, UMR CNRS 5218, F-33405 Talence, France
Loïc Lavigne and Christophe Farges
Univ. Bordeaux, IMS-lab, UMR CNRS 5218, F-33405 Talence, France
Kelly Cohen
University of Cincinnati
Franck Cazaurang
Univ. Bordeaux, IMS-lab, UMR CNRS 5218, F-33405 Talence, France


This paper presents the main objectives of the MICA convertible aircraft project supported by a grant from the French National Research Agency (ANR). Integrating convertible aircraft in airspaces with high population density requires that a certain level of safety be guaranteed, in particular during the phases of take-off and landing. It is therefore necessary to implement trajectory control systems that are robust to disturbances, fault tolerant, and capable to generate last resort rescue trajectories ensuring the safety of people and habitations. This project aims to contribute to these expectations by addressing a current technological gap by providing the capability of operating and maintaining the convertible aircraft on security transition paths (in nominal and emergency paths) and thus meeting future national and international aviation regulations.

Abstract ID: 42DCASS-022

Position, Attitude and Fault Tolerant Control of Tilt Rotor Quadcopter

Rumit Kumar (kumarrt@mail.uc.edu)
University of Cincinnati
Manish Kumar
University of Cincinnati
Franck Cazaurang
Univ. Bordeaux, IMS-lab, UMR CNRS 5218, F-33405 Talence, France
Kelly Cohen
University of Cincinnati


In this paper, various flight capabilities of a tilting rotor quadcopter are presented. A tilting rotor quadcopter is a structural advancement in conventional quadcopter design and it provides additional actuated controls by actuation of propeller motors for tilt. It can considerably improve efficiency of the aerial vehicle during flight. A detailed PD control and differential flatness based control strategy is discussed to use the propeller tilts for position, orientation and fault tolerant control. This work incorporates use of propeller rotational and tilt movement for maneuvering the UAV. A comparative study of fault tolerant PD controller w.r.t. differential flatness based flight controller is presented by numerical simulations demonstrating the performance of the flight controller for various modes of flight.

Abstract ID: 42DCASS-028

Analyzing the Concept of Global Minima for the Travelling Salesman Problem.

Prateek Sahay (sahaypk@mail.uc.edu)
University of Cincinnati
Kelly Cohen
University of Cincinnati


The traveling salesman problem (TSP) is a popular archetype for a NP-Hard combinatorial optimization problem. In this effort, the primary focus is on providing a rationale for seeking an optimal solution or possibly a near global best in a system. As most of the real world problems are large in scale and complexity direct enumeration is not feasible given the extremely large number of possible combinations.. We suggest an alternative approach provide effective solutions which are computationally efficient. Simulation results demonstrate the applicability of the proposed strategy.

Abstract ID: 42DCASS-032

Adaptive Control for Rejection of Sinusoidal Disturbances Acting on an Unknown System: Theory and Application to Aerospace Systems

Mohammadreza Kamaldar (mkamaldar@uky.edu)
University of Kentucky
Jesse B. Hoagg
University of Kentucky


The rejection of sinusoidal disturbances is a fundamental control objective in many active noise and vibration control applications such as helicopter vibration reduction, active rotor balancing, and spacecraft vibration suppression. Existing disturbance-rejection control methods require model information regarding the dynamic system (e.g., helicopter or spacecraft) to be controlled. However, in practice, this required model information can be highly uncertain. Specifically, this information may be difficult to obtain accurately and can be subject to change. For example, a helicopter's structural dynamics can change depending on payload. We present a new adaptive harmonic steady-state (AHSS) controller, which is effective for rejecting sinusoids with known frequencies that act on a completely unknown multi-input multi-output linear time-invariant system. We analyze the closed-loop stability and performance, and show that AHSS asymptotically rejects sinusoidal disturbances. The new AHSS algorithm is a frequency-domain method, meaning that all computations are with discrete-Fourier-transform data.

Abstract ID: 42DCASS-033

A Discrete-Time Flocking Algorithm with Application to Autonomous Unmanned Air Vehicles

Brandon Wellman (bjwell3@g.uky.edu)
University of Kentucky
Jesse B. Hoagg
University of Kentucky


Multi-vehicle systems have applications ranging from distributed airborne sensing to cooperative surveillance. In these cooperative applications, each vehicle relies relative-position-and-velocity feedback data to accomplish tasks such as: collision avoidance, formation cohesion, and velocity matching. Formation control can be addressed using either position-formation or distance-formation methods. Position-formation methods force vehicles into a configuration using desired relative-position vectors between pairs of vehicles. In contrast, distance-formation methods induce a configuration using only a desired relative distance between adjacent vehicles. A distance-formation method that incorporates velocity matching is often called a flocking algorithm. Most formation-control approaches focus on vehicles with continuous-time dynamics and do not account for sample-data effects. We present a new multi-vehicle control method that addresses flocking in discrete time. The method is decentralized, that is, each vehicle’s controller relies on local sensing to determine the relative positions and velocities of nearby vehicles. The method can incorporate a centralized leader vehicle for guidance. Each vehicle is modeled with discrete-time double-integrator dynamics (which are obtained by sampling the continuous-time double integrator). We demonstrate with analysis and simulations that vehicles using the new discrete-time flocking method converge to flocking formations. This work is supported in part by the National Science Foundation under OIA-1539070, and the National Aeronautics and Space Administration through the NASA Kentucky Space Grant (NNX10AL96).

Abstract ID: 42DCASS-043

Stability Analysis of a Quadcopter Using Genetic Algorithm Tuned LQR Controller

Zachary Carlton (carlton.zach@gmail.com)
University of Cincinnati
Austin Ottaway
University of Cincinnati
Kelly Cohen
University of Cincinnati
Wei Wei
University of Cincinnati


Using a genetic algorithm, the state variable and control input weights of an LQR controller for a quadcopter were tuned in an effective, computational manner. The simulation environment was solely based around the pitch and roll stability of a quadcopter during flight. By comparing the performance of the GALQR controller to a standard PD, the results show that the GALQR was more efficient at driving the pitch and roll perturbations to zero. The results also show that by using a genetic algorithm, the cost function weights of an LQR controller can be more accurately tuned rather than using normal hand-tuning.

Abstract ID: 42DCASS-072

Autonomous Destination Seeking and Obstacle Avoidance for Rotorcraft: A Backstepping Model-Reference Controller

Thomas Kirven (thomasckirven@gmail.com)
University of Kentucky
Jesse Hoagg
University of Kentucky


Advances in sensing and computing for embedded systems have increased the capabilities of autonomous unmanned vehicles. For example, consider an autonomous unmanned air vehicle that uses onboard sensors to collect environment-describing data, which are passed to an algorithm that provides the vehicle with instructions for flight guidance. This type of autonomous guidance could allow the vehicle to travel through confined spaces and avoid obstacles. Applications include technologies in aerospace and defense as well as agriculture, communication, and disaster relief. We present a backstepping model-reference rotorcraft controller for autonomous destination seeking and obstacle avoidance. This controller uses orientation, rotational rate, and acceleration feedback from an IMU as well as velocity feedback from an optical-flow camera to follow velocity commands. These commands consist of two parts: a destination-seeking command and an obstacle-avoidance command. The obstacle-avoidance command is generated using an artificial potential field that relies on relative-position data of objects near the rotorcraft; this data is obtained from an onboard depth camera. In this presentation, we demonstrate the backstepping model-reference controller in both simulation and experiment. We also demonstrate destination seeking and obstacle avoidance using a rotorcraft simulation.

Abstract ID: 42DCASS-074

Evaluating the Flying Qualities of an Autonomous Variable Stability Aircraft

Ali Hamidani (ali.hamidani@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-081

Dual-Arm Manipulation Methods with a Humanoid Robot

Matthew Verbryke (verbrymr@mail.uc.edu)
University of Cincinnati
Catharine McGhan
University of Cincinnati


Dual-arm manipulation -- the ability of a robot to grab, pick-up, and manipulate objects using two arms -- has become increasingly important as robots are entrusted with tasks that are more suited for 'two-handed work', such as handling heavy or unwieldy items at the weight limits of single-arm manipulation, or interactions with the individual components of an object. However, the introduction of combined motions using two separate manipulators can drastically increase the complexity of the system; this requires adequate control and mitigation strategies to intelligently handle the force interactions, and mediation of the intended operations between them so that the separate actions present as a joint behavior. This is necessary in order to ensure that the combined system does not act in an unstable manner that would result in damage to the object being manipulated or the manipulator arms themselves. In this presentation, methods for performing dual-arm manipulation are investigated for a space scenario involving the accurate pointing of a communication antenna sitting on the surface of Mars, a task that requires dual-arm manipulation for efficient and successful goal completion. This scenario is derived from one of the tasks laid out in NASA's Space Robotics Challenge, originally involving the R5 robot. For this work, a humanoid platform -- the 3D-printed, open-source InMoov robot -- is used; this design is generic enough in its design that many of the algorithms and methods involved in this case should be directly applicable to the R5 and various other humanoid robots.

Abstract ID: 42DCASS-082

Noncommutative Attitude Control of Small Satellites

Shaoqian Wang (shaoqian.wang@gmail.com)
University of Kentucky
Jesse B. Hoagg
University of Kentucky
T. Michael Seigler
University of Kentucky


A noncommutative attitude control system (ACS) utilizes the fundamental principle of rigid body kinematics that rotation sequences do not in general commute. A conventional satellite ACS produces a change in attitude by spinning a set of momentum wheels. Alternatively, a noncommutative ACS can achieve an identical change in attitude by oscillating the momentum wheels. Additionally, the momentum wheels can be replaced by vibrating masses. Vibrational actuators have some advantages over rotational actuators for small satellites. Compared to rotational actuators, vibrational actuators can be very small and more energy efficient, and they do not require lubrication. However, a drawback of noncommutative ACS is that control system design is significantly more complex. We present steering control strategies for noncommutative attitude control systems. We first consider the kinematic-level control problem $\dot R = R \Omega$ on SO(3) in which $\Omega$ is kinematic-level control input. The kinematic system is constrained to vibrational actuation by restricting $\Omega$ to a class of sinusoidal functions. A new closed-form solution of the kinematic system on SO(3) is used to establish controllability. We then present a $R$-feedback controller that globally asymptotically stabilizes the identity, and a feedback tracking controller that yields bounded error. The dynamic level attitude control problem is also addressed. Numerical simulations and experimental results are presented to demonstrate the effectiveness of this actuation technique for small satellites. This work is funded by NASA KY EPSCoR under grant EPSCoR-15-002 and by the National Science Foundation under grant CMMI-1538782.

Flow Control

Abstract ID: 42DCASS-001

Effect of Compressibility on Plasma-Based Transition Control for a Wing with Leading-Edge Excrescence

Donald Rizzetta (donald.rizzetta@us.af.mil)
Air Force Research Laboratory
Miguel R. Visbal
Air Force Research Laboratory


Compressibility effects were investigated for the use of plasma-based flow control, that was applied to delay transition generated by excrescence on the leading edge of a wing. The wing airfoil section has a geometry that is representative of modern reconnaissance air vehicles, and has an appreciable region of laminar flow at design conditions. Modification of the leading edge can be caused by the accumulation of debris, insect impacts, microscopic ice crystal formation, damage, or structural fatigue, and may result in premature transition and an increase in drag. A dielectric barrier discharge plasma actuator, located downstream of the excrescence, was employed to delay transition, mitigate effects of turbulence, decrease drag, and increase energy efficiency. Numerical solutions were obtained to the Navier-Stokes equations, that were augmented by source terms used to represent the body force imparted by the plasma actuator on the fluid. A simple phenomenological model provided this force resulting from the electric field generated by the plasma. The numerical method is based upon a high-fidelity finite-difference scheme and an implicit time-marching approach. An overset mesh system was employed to represent excrescence in the leading-edge region. Solutions were obtained for several Mach numbers, up to the transonic range. The effect of compressibility on the transitional behavior was explored, and the effectiveness of plasma-based control to delay transition with increasing Mach number was determined.

Abstract ID: 42DCASS-030

Computational Optimization Under Uncertainty of Active Flow Control Jet

Luke Welch (lwelch526@gmail.com)
Air Force Institute of Technology
Jacob A. Freeman
Air Force Institute of Technology
Philip S. Beran
Air Force Research Laboratory


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-154

Stall Cell Formation in Post-Stall Flow over a Boeing Vertol VR-7 Airfoil

Ata Esfahani (ghasemiesfahani.1@osu.edu)
The Ohio State University
Nathan Webb, Mo Samimy
The Ohio State University


The results of experiments to investigate flow control using nanosecond DBD (NS-DBD) plasma actuators over a thin airfoil with an aspect ratio of 3 and leading edge separation at a post-stall angle of attack are presented. The Reynolds number based on the chord was fixed at 5.0·105 and the angle of attack was set to 19°. A single actuator was mounted near the leading edge of the airfoil. Fluorescent surface oil flow visualization and stereo Particle Image Velocimetry (PIV) were employed to investigate any spanwise non-uniformities on the airfoil surface as well as in the flow. The baseline showed some spanwise non-uniformity both on and off the surface. Excitation at low Strouhal numbers (0.3 < Ste < 0.78) led to the emergence of spiral nodes near the trailing edge of the airfoil surface and a three-dimensional surface topology (similar to an asymmetric stall cell pattern). Off-surface stereo PIV data suggests, however, that the flow field remains nearly two-dimensional. Excitation at a Strouhal number of 2.04 produces distinct 3D features in the stereo PIV data. Further increases in excitation Strouhal number result in slight acceleration of the flow near the leading edge and formation of two symmetric stall cells. Increasing the excitation Strouhal number even further results in more well-defined stall cells. This effect saturates around Ste = 6.0 and further increases in excitation frequency have minimal effects on the stall cells. This is surprising as the scientific community had hitherto believed stall cells to form only over thick airfoils. The results clearly indicate that the instabilities responsible for stall cell formation are present on thin airfoils as well as thick ones and perhaps the lack the appropriate disturbance environment (or appropriate diagnostic tool) is the reason stall cells have hitherto not been observed on thin airfoils.

Abstract ID: 42DCASS-160

Control of Dynamic Stall over a NACA 0015 using NS-DBD Plasma Actuators

David Castaneda (castanedavergara.1@osu.edu)
The Ohio State University
Achal Singhal
The Ohio State University
Nathan Webb
The Ohio State University
Mo Samimy
The Ohio State University


Dynamic stall is the time-dependent flow separation and stall phenomenon that occurs due to unsteady motion of a lifting surface. When the motion is rapid enough, the flow remains attached well beyond the static stall angle of attack. The eventual stall and subsequent vortex formation, convection, and shedding processes introduce large unsteady aerodynamic loads (both lift and moment), which are undesirable. Dynamic stall occurs in many applications, including rotorcraft, MAVs, and wind turbines. This phenomenon typically occurs over the rotor at high forward flight speeds or during maneuvers with high load factors. The primary adverse characteristic of dynamic stall is the onset of high torsional loads and vibrations on the rotor due to the associated unsteady aerodynamic forces. Nanosecond Dielectric Barrier Discharge (NS-DBD) actuators are flow control devices that excite natural instabilities in the flow. These actuators have demonstrated the ability to delay or mitigate static stall. In this work, the study of the dynamic stall phenomenon and its control using NS- DBD plasma actuators is presented. Surface pressure measurements and PIV data of an oscillating airfoil were collected for the baseline and excitation cases. Pressure data was recorded for every combination of three Reynolds numbers (Re=167,000/300,000/500,000) and three non-dimensional pitch rates (k=0.025,0.05,0.075). For each case, twenty excitation Strouhal numbers (between Ste= 0.3 and 10.0) were tested. One representative case (Reynolds number and pitching rate) for baseline, low, and high Strouhal number excitation was selected for phase-locked PIV acquisition. The results lead to three important conclusions. First, low Strouhal number excitation (Ste

Abstract ID: 42DCASS-191

Shock-Trapping Capability of Various Controlled Cavities in a Supersonic Flow

Nathan Webb (webb.356@osu.edu)
The Ohio State University
Dennis Omari
The Ohio State University
Mo Samimy
The Ohio State University


Unstart prevention is a critical part of maintaining stable scramjet operation. An isolator is used to perform this function, among others. Augmenting an isolator by placing a cavity in it could increase the back-pressure margin, improving the overall engine stability and performance. A resonating cavity could provide the necessary increase in back-pressure margin by trapping the upstream travelling unstart shock; however, there would be an associated drag penalty. Localized Arc Filament Plasma Actuators (LAFPAs) have previously demonstrated the ability to enhance or suppress cavity resonance. Thus, a feedback control system could allow the cavity resonance to be suppressed during normal operation, thereby minimizing the drag penalty until added back-pressure margin is required. Previous work examined the back-pressure margin of a model isolator (straight duct) and the same isolator with a controlled cavity installed at the upstream end. The baseline resonating cavity was found to significantly increase the supportable downstream blockage. To expand on the previous work, several different cavity geometries (length and aft wall shape) were tested and the associated improvements in supportable downstream blockage and estimated back-pressure documented. The cavity geometry was found to have a strong effect on its shock-trapping ability, with cavity length being the dominant parameter. LAFPAs were used to modify the cavity resonance state. This was found to also have a significant effect on the achieved benefit, though the effect of geometry was still dominant. Finally, one cavity geometry was tested to observe the effect of the LAFPAs on the drag penalty. The excitation was found to significantly alter the cavity drag; however, rather than decreasing as the cavity resonance was suppressed, the drag increased. This is counter to what was expected, given the existing literature, and more work is required to understand this phenomenon.

Fluid Dynamics

Abstract ID: 42DCASS-015

Measurements of Crossflow Instability Modes for HIFiRE-5 at Angle of Attack

Matthew Borg (matthew.borg.3@us.af.mil)
Air Force Research Laboratory
Roger L. Kimmel
Air Force Research Laboratory


A 2:1 elliptic cone model was tested in a Mach-6 quiet wind tunnel with both low and elevated freestream noise levels. Simultaneous measurements were made using an infrared camera and 20 pressure sensors mounted flush with the model surface. In quiet flow, both stationary and traveling crossflow instabilities were observed. The transition Reynolds number increased by as much as 30% when the model pitch was increased from 0 to 2 degrees. Crossflow transition at a given spanwise location collapsed reasonably well with transition Reynolds number when the model was not pitched. Traveling crossflow instability waves were found to develop similarly over the instrumented portion of the model surface. In noisy flow, the boundary layer was observed to transition, however neither traveling or stationary crossflow waves were observed. For increasing angle of attack, transition in the region of high crossflow was observed to move downstream. Additionally, power spectra of the pressure signals normalized by the mean static pressure decreased in magnitude with increasing angle of attack. This behavior suggests that transition in noisy flow is at least affected by the crossflow instability.

Abstract ID: 42DCASS-027

Effects of Aerating Orifice Diameter on Two-Phase Flow Structures in Aerated-Liquid Injectors Using X-Ray Techniques

Travis Tidball (travis.tidball.ctr@us.af.mil)
Taitech Inc.
Kuo-Cheng Lin
Taitech Inc.
Alan Kastengren
Argonne National Laboratory
Campbell Carter
Air Force Research Laboratory


Structures of two-phase flows within an axisymmetric aerated-liquid injector were explored experimentally using the synchrotron x-ray source available at the Argonne National Laboratory. The aerated-liquid jet is a capable injection scheme to provide favorable plume and droplet properties for efficient combustion in several propulsion applications. This injection scheme features the addition of a small amount of gas into the liquid to create a two-phase mixture inside the injector. Structures of the two-phase flows inside the aerated-liquid injector dictates the structures of the discharged jets and, therefore, should be explored with an adequate experimental setup in order to provide guidance for injector design. The present experiments were carried out using a beryllium nozzle which has relatively high x-ray transmittance for x-ray diagnostics. This nozzle was mated with a stainless steel aerated-liquid injector featuring the inside-out aerating scheme for two-phase flow formation. Three perforated aluminum aerating tubes with variations in orifice diameter were used for aerating gas delivery. Each aerating tube has the same placement of aerating orifices in order to investigate the effects of aerating orifice diameter on two-phase flow structures. Water at room temperature was doped with potassium iodide at a low concentration to increase the x-ray absorption coefficient for enhanced contrast in x-ray imaging. Nitrogen at room temperature was used as the aerating gas and was metered at a relatively low flow rate. The two-phase flow was sprayed horizontally into a quiescent environment. High-speed imaging with a framing rate of 271,554 frames per second, a shutter speed of 0.25 s, and a spatial resolution of 2.88 × 5.12 mm were used to capture the temporal evolution of the two-phase mixture within the beryllium nozzle. A total of 7,500 images were recorded at each probing location for each injection condition. Each image was further processed for background noise removal and contrast enhancement. Average, standard deviation, and sequence of the obtained images were used to explore both time-averaged and temporal evolution of the two-phase mixtures. The preliminary results show that the aerating tube with small aerating orifices is capable of generating deeply-penetrating gas plumes to mix with the water crossflow for a given injection condition. Size and structure of the two-phase flow within the re-circulations zones at the corners of plenum chamber exhibit a dependence on aerating orifice diameter. Despite these differences, the time-averaged images show that the two-phase flow within the nozzle section is fairly insensitive to aerating orifice diameter. This observation agrees with previous studies using the x-ray fluorescence technique to quantitatively characterize the two-phase flows within the nozzle section.

Abstract ID: 42DCASS-039

Prediction of Hydrodynamic Ram Cavity Contraction using Orifice Mass Flow, Cavity Geometry, and Projectile Kinetic Energy Dissipation

Andrew Lingenfelter (andrew.lingenfelter@afit.edu)
Air Force Institute of Technology
David Liu
Air Force Life Cycle Management Center


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-042

Boiling and Evaporation Waves

Trace Kuberski (trku222@uky.edu)
University of Kentucky
Jose Grana-Otero
University of Kentucky


Evaporation waves are the typical boiling mode of liquids subject to moderate-to-high superheat degrees. These boiling fronts propagate into the liquid, with velocities that increase with the superheat degree, transforming the superheated liquid ahead into a two-phase flow behind composed of a mist vapor and small droplets. They play an important role in the basic understanding of boiling, but also in diverse fields such as in volcano’s eruptions. Despite their significance, they are still very poorly understood. We have conducted an experimental investigation aimed at uncovering the basic mechanisms controlling their formation and propagation velocity. High-speed visualizations, well-resolved in time and with high-spatial resolution have permitted to reveal that, contrary to the general belief, they are initiated and sustained, just as common boiling, by bubble formation and growth on microscopic cavities present on the container’s surface, rather than due to homogeneous boiling in the bulk liquid as previously speculated.

Abstract ID: 42DCASS-077

PIV Investigation into Cross-stream Behavior in Wing Wake Free Shear Layers: Challenges and Results

Muhammad Omar Memon (mmemon1@udayton.edu)
University of Dayton
Aaron Altman
University of Dayton


A fresh look at the cross-stream flow in the wake of a wing has revealed some surprising results and posed some experimental challenges. The wake behind a three-dimensional wing is formed through the interplay of different sources of drag. Lift induced drag is generated as a byproduct of downwash from wingtip vortices. Parasite drag results from form/pressure drag and a combination of the upper and lower surface boundary layers. These parasite effects amalgamate to create the free shear layer in the wake. Particle Image Velocimetry (PIV) experiments were performed in the cross-stream direction in the wake of an AR 6 wing with a Clark-Y airfoil in the UD Low Speed Wind Tunnel. The objective was to investigate flow topology of and around the wing wake free shear layer. Performing PIV in the cross-stream plane to capture and correctly resolve the entire wingtip vortex poses a number of challenges. Resolving cross-stream flow in the free shear layer in the wake behind a wing is more challenging as the cross stream flow is typically only small fraction of the freestream (out of plane) velocity. Reliable results can be systematically obtained by careful manipulation of test parameters, many beyond the range of existing commonly accepted heuristics. The resulting velocity components in and around the free shear layer in the wing wake showed counter flow in the cross-flow plane presumably corresponding to behavior associated with the flow over the upper and lower surfaces of the wing. An indication of the possibility of the transfer of momentum (from inboard to outboard of the wing) was identified through spanwise flow corresponding to the upper and lower surfaces through the free shear layer in the wake. A transition from minimal cross flow in the free shear layer to a well-established shear flow in the spanwise direction occurs in the vicinity of maximum lift-to-drag ratio (max L/D) angle of attack.

Abstract ID: 42DCASS-075

Multiphysics Modeling and Simulation of Multiphase Flows

Prashant Khare (Prashant.Khare@uc.edu)
University of Cincinnati


Liquid sprays and droplets play an important role in numerous applications of practical interest including, liquid-fueled combustion devices such as diesel, gas-turbine and rocket engines, cooling of turbine blades and microchips, and industrial processes such as spray painting and inkjet printing. Even after decades of research, because of the lack of appropriate diagnostic and simulation tools, the understanding of the atomization process remains limited. Additionally, no attempts were made in the past to conduct fundamental studies that led to the development of universal theories and models to predict statistics, such as, droplet/particles sizes and distributions, resulting from the breakup of liquid jets and droplets. Furthermore, the effect of multiphysics processes, such as acoustic, electro-static and electromagnetic excitation on the liquid sprays and droplets has not been widely explored. Therefore, this study focuses on three aspects of droplet dynamics, (1) investigation of fundamental physics underlying the deformation and fragmentation of liquid droplets; (2) development of generalized models to predict the behaviors of resulting child drops over a wide range of operating pressures and Weber numbers; and (3) the effect of electro-static fields on liquid jets and droplets. The theoretical and mathematical formulation to study these multiphase problems is based on incompressible Navier-Stokes equations with surface tension coupled with Maxwell's electromagnetic equations. A critical issue is the treatment of multi-scale liquid-liquid and gas-liquid interfaces. A state-of-the-art, high resolution, volume-of-fluid (VOF) interface capturing method is adopted to resolve the large-scale interfacial evolution. Surface tension is accommodated as a Dirac delta distribution function on the interface. The theoretical formulation outlined above is solved numerically using a finite volume method augmented by an adaptive mesh refinement (AMR) technique to improve the solution accuracy and efficiency. The adaptive quad/octree spatial discretization allows for interface refinement. For the purely hydrodynamic problem, general theories that quantitatively explain the dynamics of liquid droplets over a wide range of pressures, and Weber numbers are established. These theories are used to develop universal models that can predict the droplet behaviors, including size-distributions and drag coefficients with deformation and fragmentation. Next, two canonical configurations, liquid jet injection in quiescent environment and droplet collision, are used to demonstrate the effect of electro-static fields on multiphase flows. The ultimate goal of the effort is to enhance the fundamental understanding of multiphase flows, and to develop theories and algorithms for active and passive control using multiphysics processes.

Abstract ID: 42DCASS-095

Analysis of Windward Side Hypersonic Boundary Layer Transition on Blunted Cones at Angle of Attack

Matthew Tufts (matthew.tufts.1.ctr@us.af.mil)
Ohio Aerospace Institute
Roger L. Kimmel
Air Force Research Laboratory


The seven-degree half-angle cone is a canonical geometry for high-speed boundary-layer transition. This paper presents constant N-Factor transition predictions for the Mack 2nd Mode along the windward plane for this geometry at differing angles of attack and differing nose radii in preparation for an upcoming experimental series. The stability analyses are able to capture the experimental trends seen in previous experiments. Computationally derived entropy swallowing lengths are shown to largely match the trends derived analytically in previous studies. These entropy swallowing lengths appear to be applicable as a figure of merit for transition studies, even at angles of attack. Bluntness appears to delay the amplification of Mack 2nd mode instabilities along the attachment line, rather than increasing the critical amplitude leading to transition. As is the case for cones at zero angle of attack, when transition occurs within the swallowing distance along the attachment line, it does not appear to be well-correlated with PSE/LST N-Factors.

Abstract ID: 42DCASS-098

Influence of the External Aeroshell on the HIFiRE-6 using High-Fidelity Simulations

Nicholas Bisek (Nicholas.Bisek.1@us.af.mil)
Air Force Research Laboratory


High fidelity unsteady simulations were performed for the sixth vehicle in the Hypersonic International Flight Research Experimentation (HIFiRE-6) to investigate the complex, shock-dominated turbulent air as the vehicle flies along a nominal Mach 6 trajectory. The simulations were obtained using a fifth order WENO scheme and solved on a wall-resolved grid system with sufficient resolution to support the flow's unsteady behavior. The present study explores two versions of the geometry, including one where the external aeroshell has been significantly truncated by removing all appendages. A comparison between the two configurations shows how the external appendages change the development of the external flow field, which results in changes to the nozzle exit pressure profile in certain scenarios. Both time-mean and unsteady characteristics are investigated for several angles of attack and freestream conditions. In general, the external aeroshell does not influence the development or efficiency of the geometry's internal flow path, though excluding the external appendages does result in an under prediction in the captured mass flow rate for scenarios with the negative angle of attack or a lower freestream dynamic pressure.

Abstract ID: 42DCASS-136

HIFiRE-5b Flight Experiment Results

Roger Kimmel (roger.kimmel@us.af.mil)
Air Force Research Laboratory
David Adamczak
Air Force Research Laboratory
DSTG AVD Brisbane Team
- Other -


The Hypersonic International Flight Research Experimentation (HIFiRE) program is a hypersonic flight test program executed by the Air Force Research Laboratory (AFRL) and Australian Defence Science and Technology Group (DSTG). The HIFiRE-5b flight flew in May 2016. The principle goal of this flight was to measure hypersonic boundary-layer transition on a three-dimensional body. The flight was fully successful, and measured leading edge, acreage and centerline transition under supersonic and hypersonic conditions. This paper describes the HIFiRE-5b mission and some general conclusions derived from the experiment.

Abstract ID: 42DCASS-139

Local Dissipation Scales in External Wall-bounded Flows with and without Free-Stream Turbulence

Sabah Alhamdi (sfal222@g.uky.edu)
University of Kentucky
Sean C.C. Bailey
University of Kentucky


There has been a particular interest in the scaling of the distribution of the probability density functions (PDFs) of the length scales at which dissipation occurs. This is in contrast to the traditional approach of treating dissipation as occurring at a single (Kolmogorov) length scale. Such PDFs have been previously reported in pipe and channel flows. In both cases, the PDFs suggested that the distribution of local dissipation scales is universal when normalized by a properly chosen length scale. In homogeneous isotropic turbulence, this scale is analogous to the Kolmogorov scale, but in wall-bounded flows this scale is proportional to the distance from the wall. The objective of the current research is to examine the scaling of PDFs within boundary layers developing within both laminar and turbulent free streams. The objective is to assess whether the channel flow scaling suggested by Bailey & Witte (J. Fluid Mech. (2015), vol. 786, pp. 234-252) using a wall-distance dependent length scale can be applied to these flows. Experiments have been performed to measure the properties of a nominally two-dimensional, zero-pressure-gradient boundary layer developing along a smooth plate mounted in a wind tunnel. Streamwise velocity profiles for these two flow regimes were measured using a hot-wire anemometer traversed normal to the plate surface. Results suggest the existence of a universal distribution which scales differently for inner and outer regions, and for the two flow regimes in turbulent boundary layers. Future work is focusing on improving the scaling by accounting the effect of the intermittency in the edge of the outer region of turbulent boundary layer flow.

Abstract ID: 42DCASS-147

A Comparison of Liquid Jets in a Gaseous Crossflow

Beau Stegemann (beau.stegemann@afit.edu)
Air Force Institute of Technology
Marc D. Polanka
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-179

Refined Insight Into The Wingtip Vortex - Free Shear Layer Interaction

Sidaard Gunasekaran (gsidaard@gmail.com)
University of Dayton
Aaron Altman
University of Dayton


The interaction between the free shear layer (FSL), the wingtip vortex and the aerodynamic efficiency was investigated behind a lift-generating wing using PIV to better understand the balance between the parasite drag and lift-induced drag. After a baseline clean wing configuration was tested, the boundary layer on the upper surface of the wing was tripped using a spanwise boundary layer trip. The resultant changes in the wingtip vortex and the FSL were quantified. The profiles of the velocity components reconstructed through multiple different PIV planes indicated strong FSL interaction with the evolution of the wingtip vortex in all three axes. This interaction exerted its greatest influence around the maximum (L/D) angle of attack.

Abstract ID: 42DCASS-193

Gust Interaction and Aeroelastic Response of an Airfoil at Transitional Reynolds Numbers

Caleb Barnes (caleb.barnes.1@us.af.mil)
Air Force Research Laboratory


This presentation explores the interaction of a parallel vortical gust with rigid and elastically mounted NACA0012 airfoils at a transitional Reynolds number. Gust interaction was conducted at a Reynolds number of 150,000 where previously a strong artificial disturbance was required to excite a sustained flutter response. Both pitch-only and pitch-heave configurations yielded laminar separation flutter--a recently discovered phenomenon for airfoils in the transitional flow regime distinct from the more common stall flutter and transonic flutter phenomena. The transient stages of unsteady flow structure during and after gust interaction exhibited features reminiscent of those observed during gust-free flutter; breakup of the lower surface boundary layer into a series of spanwise coherent vortices followed by transition to turbulence accompanied by a moving laminar separation bubble. The combination of these two effects provided a long-lasting pitch-up bias in the moment coefficient responsible for exciting the subsequent flutter response.

Heat Transfer

Abstract ID: 42DCASS-055

Multi-Dimensional Surface Chemistry Comparison of Iso-Q Arcjet Cases

Justin Cooper (Justin.Cooper@uky.edu)
University of Kentucky
Haoyue Weng
University of Kentucky
Alexandre Martin
University of Kentucky


An in-situ equilibrium surface chemistry solver for the calculation of recession in carbon-phenolic type in-depth ablators is coupled with a material response code modeling pyrolysis gas momentum. Verification cases are presented for the recession module and a validation based on NASA Ames Research Center (ARC) arcjet data, as well as an intercode comparison of surface temperature and recession. Further, the geometric effects of pyrolysis gas movement through the body of the ablator and its subsequent effect on the calculation of surface chemistry is examined. The multidimensional approach highlights the concept of varying concentrations of pyrolysis gas species and their effect on the gas equilibrium reactions computed locally across the body surface.

Abstract ID: 42DCASS-078

Computational Analysis of an Ultra Compact Combustor with Conjugate Heat Transfer

Brian Bohan (brian.bohan@afit.edu)
Air Force Institute of Technology
Marc D. Polanka
Air Force Institute of Technology
James L. Rutledge
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-080

Relating Film Cooling at Low and High Temperatures

Christopher Vorgert (christopher.vorgert@afit.edu)
Air Force Institute of Technology
Marc D. Polanka and James L. Rutledge
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-083

Validation of the KATS Material Response Solver with Arc-Jet Experiments

Olivia Schroeder (olivia.schroeder1@uky.edu)
University of Kentucky
Haoyue Weng
University of Kentucky
Alexandre Martin
University of Kentucky


Modeling the atmospheric entry of spacecraft is challenging because of the vast number of physical phenomena that occur during the process. In order to size thermal protection systems, engineers rely on high fidelity solvers to provide accurate predictions of both the thermochemical environment surrounding the heat shield, and its material response. Therefore, it is necessary to guarantee that the numerical models are correctly implemented and thoroughly validated. The current work provides a framework for the validation of the Kentucky Aerothermodynamics and Thermal Response Solver (KATS) [1] using an array of arc-jet experiments [2]. The experiments were performed on a comprehensive range of pressures, from 2.3 kPa to 84.4k Pa, and stagnation-point cold-wall heat fluxes, from 107 W/cm^2 to 1102 W/cm^2. Furthermore, different environment conditions are tested by varying concentrations of Argon in air. The predicted recession and surface temperature of the Phenolic Impregnated Carbon Ablator (PICA) samples are compared to experimental results as well as to an extensively validated material response code, the Fully Implicit Ablation and Thermal Analysis Program (FIAT) [3]. References: [1] Weng, H. and Martin, A., “Multi-dimensional modeling of pyrolysis gas transport inside charring ablative materials,” Journal of Thermophysics and Heat Transfer, 2014. [2] Milos, F. S. and Chen, Y.-K., “Ablation and Thermal Response Property Model Validation for Phenolic Impregnated Carbon Ablator,” Journal of Spacecraft and Rockets, Vol. 47, No. 5, September-October 2010, pp. 786–805. [3] Chen, Y.-K. and Milos, F. S., “Fully Implicit Ablation and Thermal Analysis Program (FIAT)” Journal of Spacecraft and Rockets, Vol. 36, No. 3, May-June 1999, pp. 475– 483.

Abstract ID: 42DCASS-119

Synthetic Vortex Interaction with Film Cooling Effectiveness

Aaron Wheeler (wheelar@mail.uc.edu)
University of Cincinnati
Paul Aghasi
University of Cincinnati
Ephraim Gutmark
University of Cincinnati


Film cooling has been a primary form of turbine blade cooling for decades. Film cooling holes act essentially as jets in cross flow, creating a counter-rotating vortex pair (or CVP). This CVP encourages mixing of the coolant flow and the hot crossflow, which overall reduces the film effectiveness. Reducing the intensity of these CVPs would in theory also reduce mixing of coolant and crossflow, in turn increasing the film cooling effectiveness. This work will employ the addition of vortex generators upstream of a row of film holes on a flat plate wind tunnel to observe the effect of these synthetically created vortices on the film cooling effectiveness. A flat plate film cooling facility will be used which implements a mass transfer analogy (introduction of a foreign gas to represent the coolant) along with measurements of gas sampling and oxygen sensitive paint to measure the effects of the vortex generators on the film cooling effectiveness. Vortex generators will be placed at varying distances upstream, downstream, and cross stream of the film holes to observe their differing effects.

Abstract ID: 42DCASS-125

Relative Contributions to Overall Effectiveness in Turbine Blade Leading Edge Cooling

Carol Bryant (carol.bryant@afit.edu)
Air Force Institute of Technology
James L. Rutledge
Air Force Institute of Technology
Marc D. Polanka
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-196

Cooling System for 0.1 kN Thrust Micro-Engines: Concept Design Using Additive Manufacturing

Matteo Ugolotti (ugolotmo@mail.uc.edu)
University of Cincinnati
Mayank Sharma
University of Cincinnati
Zachary Williams
University of Cincinnati
Justin Ouwerkerk
University of Cincinnati
Siddharth S. Balachandar
University of Cincinnati


This project is the outcome of the 2015/2016 Aerospace Outreach Propulsion Program (APOP) and is aimed to add an innovative cooling system to a small turbojet engine of 0.1 kN of thrust. The interesting and innovative aspect of this project is the application and adaption of conventional cooling system solutions to such a small single-stage turbine (11mm of blade height), employed on the JetCat P90-Rxi. The ultimate goal is to increase the maximum temperature of the engine cycle by 150 K, in order to get more thrust. To avoid the failure of the engine due to the higher temperatures, an internal cooling system is implemented to cool down both the stator and rotor of the turbine. The stator blades are provided with internal channels and trailing edge slots to maintain the temperatures below the Inconel 718 thermal limit, while a pattern of bleed holes on the shroud combined with an annular slot at the hub supply cooling to the rotor blades. The rotor blade surface is covered with an innovative thermal barrier coating which should guarantee a sufficient temperature drop. The original stator and rotor components were modi ed, while the overall structure of the engine was maintained. The new parts were rst 3D printed in ABS to verify the matings and, eventually, 3D printed in stainless steel 316L. The final step was the 3D printing in Inconel 718.

Human Factors

Abstract ID: 42DCASS-026

Development of an exoskeleton for sit-to-stand (STS) transition support based on multimodal action intent recognition

Gaurav Patil (patilgs@mail.uc.edu)
University of Cincinnati
Lilian Rigoli, Michael J. Richardson
University of Cincinnati
Anca Ralescu, Adam Kiefer
University of Cincinnati
Manish Kumar
University of Cincinnati
Tamara Lorenz
University of Cincinnati


1.5 million Senior citizens live under supervision and most require assistance with at least one or more Activities of Daily Living (ADL), including transferring in and out of chairs, beds and toilets which requires the ability to perform sit-to-stand transition. This sit-to-stand transition is a complex full-body activity that requires the synergistic coordination of the upper and lower limbs and trunk. The goal of this research is to come up with a working prototype of an active, assistive exoskeleton which can be controlled based on behavioral models of user’s intent. The research plan includes synchronized multimodal data-collection of sit-to-stand transitions across various environmental situations and action intent contexts and development of intelligent control algorithms to actuate and operate the exoskeleton. This work can be expanded to control robots in any environment which requires human-robot coordination to complete the same task. This presentation will give a brief overview of the research plan, adaptive control system design and its performance in simulations and intent recognition from the multimodal data.

Abstract ID: 42DCASS-054

On the Effects of System Characteristics and Reference Command as Humans Learn to Control Aerospace Systems

Seyyedalireza Seyyedmousavi (sse243@g.uky.edu)
University of Kentucky
Faina Matveeva, Xingye Zhang, T. M. Seigler, and Jesse B. Hoagg
University of Kentucky


Humans learn to interact with many complex physical systems. For example, humans learn to ride bicycles, drive cars and fly airplanes. This research seeks to address questions of human learning: What control strategies do humans use? What characteristics make a system difficult for humans to control? What are the effects of changing reference command? We present results from experiments in which human subjects learn to control different dynamic systems. In one set of experiments, 55 subjects (5 groups of 11) each interact with a dynamic system 40 times over a one-week period. Each group interacts with a different dynamic system. We use the experimental results to determine characteristics (e.g., system zeros, relative degree, phase shift) that can make a system difficult for a human to control. We also use subsystem identification to model the control strategies (feedback and feedforward) that each subject uses on each trial. For each group, the average identified feedforward controller used on the last trial approximates the inverse of the dynamic system. In another set of experiments, 44 subjects (4 groups of 11) each interact with a dynamic system 40 times over a one-week period. For each interaction, a subject is asked to perform a command-following task. For each subject and each interaction, the dynamic system is the same; however, the task (i.e., reference command) is not necessarily the same. We use the experimental results to examine the effect of changing task. Experimental results suggest that the subjects are able to generalize a control strategy learned on one task to a different task. Results also suggest that subjects are able to learn without relying on prediction, but their ability without prediction is more limited. This work is supported in part by the National Science Foundation under CMMI-1405257, and the Kentucky Science and Engineering Foundation under KSEF 148-502-15-364.

Imaging & Diagnostics

Abstract ID: 42DCASS-009

Spatially localized, gas-phase temperature measurements through the ceramic walls of a flow reactor and jet-stirred reactor using Radar REMPI

Joseph Miller (joseph.miller.35@us.af.mil)
Air Force Research Laboratory
Mark Gragston, Yue Wu, Zhili Zhang
University of Tennessee-Knoxville
Robert Stachler and Joshua Heyne
University of Dayton
Scott Stouffer
University of Dayton Research Institute


“See-through-wall” coherent microwave scattering from Resonance Enhanced Multiphoton Ionization (REMPI) has been developed for rotational temperature measurements of molecular oxygen and demonstrated in a flow reactor and jet-stirred reactor at atmospheric pressure. Through limited, single-ended optical access, a laser beam was focused to generate local ionization of molecular oxygen in a heated quartz flow reactor enclosed by ceramic heating elements. Coherent microwaves were transmitted, and the subsequent scattering off the laser-induced plasma was received, through the optically opaque ceramic walls and used to acquire rotational spectra of molecular oxygen and determine temperature. The diagnostic was characterized in a non-reacting flow reactor with well-defined boundary conditions. Both axial and radial air-temperature profiles were obtained with an accuracy of ±20 K (±5%), showing good agreement with a steady-state computational heat transfer model. The diagnostic was applied in a well-stirred reactor to measure spatial inhomogeneity of temperature, yielding data useful for characterizing the reactor and enabling quantitative assessment of deviations from perfectly stirred reactor operation.

Abstract ID: 42DCASS-112

Four-dimensional laser-induced fluorescence measurements of OH

Ben Halls (hallsbenjamin@gmail.com)
Spectral Energies LLC
Paul Hsu, Naibo Jiang, Mikhail N. Slipchenko, Ethan Legge, and Sukesh Roy
Spectral Energies LLC
Terrence R. Meyer
Purdue University
James R. Gord
Air Force Research Laboratory


Four-dimensional OH concentration fields were measured in a lifted jet flame. The base of the flame was volumetrically illuminated by a 284 nm laser pulse. The 10 kHz pulse train was generated by a burst-mode-laser-pumped optical parametric oscillator. Eight perspectives were imaged using 2 high-speed intensified cameras coupled to quadscopes. The 8 unique views were processed in volumes using LaVision DaVis 8.3 tomography software. Calibration data were collected in a Hencken burner to determine the absolute OH concentration. The accuracy and precision of the measurements are quantified and the limitations of the technique are discussed. This manuscript has been cleared for public release by the Air Force Research Laboratory (No. 88ABW-2016-6581).

Abstract ID: 42DCASS-199

Measurement Stability Analysis of Fuel-Air Ratio using Correlation between Laser-Induced-Breakdown Signal and Electron Generation

Anil Patnaik (anil.hbar@gmail.com)
Spectral Energies LLC
Paul S. Hsu
Spectral Energies LLC
Yue Wu, Mark Gragston, Zhili Zang
University of Tennessee-Knoxville
James R. Gord
Air Force Research Laboratory
Sukesh Roy
Spectral Energies LLC


In a recent study, nanosecond-laser–based LIBS was employed for quantitative local fuel to air (F/A) ratio measurements in well-characterized methane-air flames at pressures of 1 – 11 bar. Nitrogen and hydrogen atomic emission lines at 568 nm and 656 nm, respectively, were selected to establish a correlation between the line intensities and the F/A ratio. From the detailed parametric study, it was observed that ns-LIBS shows very high instability in F/A ratio measurement at high pressures. For this presentation, simultaneous and correlated measurement of time-resolved laser-induced-breakdown spectroscopy (LIBS) signal and electron number is proposed for diagnostics of hydrocarbon flames at elevated pressure. The proposed measurement enables developing an understanding for the source of the aforesaid measurement instability. The results obtained from experiment clearly show that both LIBS and N are very highly correlated at low pressure but the correlation is almost lost at high pressures. Also, from current observations, it is clearly observed that the high instability in the LIBS signal and hence instability in F/A ratio measurement at high pressures are caused by the highly unstable avalanche ionization, in addition to the presence of high level of soot. The planned correlated measurements of LIBS and electron number with short-ps to long-fs excitations are expected to reduce the instability in F/A ratio measurements at high pressure.

Materials

Abstract ID: 42DCASS-041

Investigating the effectiveness of TiAlN coating on carbide tools during flood coolant and dry machining of aerospace material Ti-6Al-4V

Muhammad Jahan (jahanmp@miamiOH.edu)
Miami University
G.K. Arbuckle
Western Kentucky University


Ti-6Al-4V, commonly known as grade 5 titanium alloy, is a widely used material in aerospace applications due to its lightweight, high specific strength, and corrosion resistance. Ti-6Al-4V is commonly known as a difficult-to-cut material due to its poor thermal conductance, strong alloying tendency and work-hardening characteristics. Among various approaches to overcome the difficulty of machining titanium alloys, use of coated cutting tools have been found to be effective in recent studies. In the current study, the effectiveness of titanium aluminum nitride (TiAlN) coating on carbide tools for machining of Ti-6Al-4V was evaluated in both conventional flood coolant and sustainable dry machining conditions. The performance of TiAlN-coated and uncoated carbide tools was evaluated based on the resulting cutting forces, tool tip temperature, surface roughness and tool wear. It was found that the cutting forces and tool tip temperature were significantly higher for dry machining compared to coolant machining, when uncoated carbide tools were used. The TiAlN coating was found to reduce both cutting forces and tool tip temperature, when used in dry machining conditions. However, the effect of tool coating was not very significant when used in flood coolant condition. The cutting forces and tool tip temperature was lower at coolant machining condition compared to dry machining, but did not further reduce when TiAlN-coated tools were used in coolant condition. The tool wear was also significantly reduced in dry machining with TiAlN-coated tools due to the reduction of cutting forces and tool tip temperature. The average surface roughness was reduced significantly in dry machining with TiAlN-coated tools, although there was no significant improvement in the surface roughness in coolant machining when TiAlN-coated tools were used. Therefore, it was concluded that TiAlN coating was effective in dry machining of Ti-6Al-4V, and the performance of dry machining with TiAlN-coated tools were comparable to that of coolant machining with uncoated tools.

Abstract ID: 42DCASS-065

Process Modeling of Additively Manufactured High Temperature Thermosets

Victoria Hutten (victoria.hutten.1.ctr@us.af.mil)
University of Dayton Research Institute
Brent Volk
Air Force Research Laboratory
Andrew Abbott
University of Dayton Research Institute


In this work, we create a finite element method for predicting the processing of printing – via fused deposition modeling (FDM) – polymer resins and polymer matrix composites. With increasing development in extrusion-based additive manufacturing, there is a great need for understanding the process parameters that result in optimal final part performance. Process models capable of predicting the thermal history and residual stresses during extrusion and layup would provide insight into road-to-road bonding, the main contributor to part strength. A framework for predicting the history of the printed part, while leveraging the process modeling advancements for traditional polymer matrix composites, is developed. Specifically, the print path is coupled with sequential thermo-mechanical analyses in which the state variables are calculated as a function of time and temperature. We apply this framework to thermoplastic and thermosetting polymers materials that are printed using FDM.

Abstract ID: 42DCASS-067

Coupled Multi-Scale Approach for Improved Design of FDM Printed Parts

Solomon Duning (solomon.duning@udri.udayton.edu)
University of Dayton Research Institute
Brian Czapor and Gyaneshwar P. Tandon
University of Dayton Research Institute
Cody Godines and Frank Abdi
Alpha STAR Corporation


Since its inception, additive manufacturing has primarily been utilized for the rapid generation of prototype parts. Recently however, there has been a growing desire to use additive manufacturing processes to create flight hardware. For optimal flight hardware design, parts are often improved through the utilization of computational models and finite element approaches. One process that may benefit from finite element analysis is Fused Deposition Modeling (FDM)—a process in which beads of material are extruded along a complex path to build layers of the part. This work presents a novel coupled multi-scale approach for improved design of FDM printed parts. Specimens and parts were printed on a FORTUS 400mc machine using both a commercial-grade ULTEM 1010 filament and an experimental carbon fiber filled ULTEM 1010 material. The printed parts were then examined at mesostructure, ply, and part scales using image analysis techniques and mechanical stress analysis. Data from the material examination was used to characterize the model material in MCQ and then stress-strain data was validated using GENOA 3D. After model verification for both materials at each of the three scales, a multi-material part consisting of a conductive silver paste printed on a polymer substrate was generated and modeled. This approach proved to be a valid technique for verifying and improving the design of FDM printed parts.

Abstract ID: 42DCASS-068

Influence of Matrix Conversion on Development of Fiber-Matrix Interfacial Strength in Polymer Matrix Composites

Ray Coomer (richard.coomer@udri.udayton.edu)
University of Dayton Research Institute
G. P. Tandon
University of Dayton Research Institute
Nicholas J. Pagano
UES Inc.
Tara M. Storage
Air Force Research Laboratory


Residual stresses affect the dimensional accuracy and structural integrity of complex composite parts often leading to overdesigned components and excessive manufacturing cost. The polymer matrix cure process involves complex interactions among thermal inputs, and evolving constituent properties. Numerical simulations are helpful, but state-of-the-art (SOA) process models do not currently incorporate the effect of the fiber-matrix interface; a perfect interface is assumed at all times during processing. It is not known when the interface begins to develop strength and participate in load transfer. Through the development of novel experimental methods, this work aims to provide insight into the quantitative development of fiber-matrix interfacial strength, as a function of DoC, and its effect on accumulated residual stresses during the processing of polymer matrix composites. Distribution A. Approved for public release: distribution unlimited. Case#: 88ABW-2017-0277

Abstract ID: 42DCASS-086

Development of VISTA, an open-source Avcoat material database

Ali Omidy (adomidy@gmail.com)
University of Kentucky
Justin M. Cooper
University of Kentucky
Haoyue Weng
University of Kentucky
Alexandre Martin
University of Kentucky


The re-emergence of the Avcoat material, used on NASA's current Multi-Purpose Crew Vehicle, has brought the need to reassess its material properties and engineering performance. In order to further understand the phenomenological physics material presents, an open source material database has been developed. The study assembles historical data available from literature to develop a theoretical framework for modeling purposes with the intent of providing a standard to the material response community. The presented material model, dubbed VISTA, allows for direct comparison between codes while allowing further investigation to Avcoat specific behaviors. The selection process of the properties is presented, and a preliminary assessment of its performance is provided, using historical Apollo data.

Abstract ID: 42DCASS-106

Analysis of the Effects of Additive Manufacturing on the Material Properties of 15-5PH Stainless Steel

Eric Lum (eric.lum@afit.edu)
Air Force Institute of Technology
Anthony Palazotto
Air Force Institute of Technology
Allison Dempsey
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-115

Fatigue Behaviour of an Advanced Melt Infiltrated Sic/Sic Composite at 1200˚C in Air and in Steam

Nicholas Boucher (nicholas.boucher.4@us.af.mil)
Air Force Institute of Technology
Marina B. Ruggles-Wrenn
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-132

Creep and oxidation of Hafnium Diboride in Air at 1500°C

Thomas Bowen (thomas.bowen@afit.edu)
Air Force Institute of Technology
Marina-Ruggles Wrenn
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-135

Studying the Effect of Boundary Conditions on a Micromechanics Based Material Model in Determining the Elastic Properties of Unidirectional Composites

Sandeep Medikonda (medikosp@mail.uc.edu)
University of Cincinnati
Ala Tabiei
University of Cincinnati
Rich Hamm
Procter & Gamble


Boundary conditions between constituent sub-cells effectively define the response of a RVC in composite material models, hence, there is a need to accurately define these boundary conditions in order to achieve an accurate response of the lamina. A micromechanical model based on the physically viable sub-cell boundary conditions is developed and implemented for use with uni-directional composite laminates in the explicit finite element method. Stress-strain relations have been presented in a three-dimensional context and hence can be used with solid elements. The objective of this work is to study the effect of boundary conditions in accurately estimating the elastic properties of a uni-directional composite lamina. In order to achieve this, the developed micro-model has been studied alongside 2 other models with different boundary conditions specified in the literature. Numerical results are generated for engineering constants by the considered models and compared against each other for different laminas. In particular, transverse and shear modulus have been analyzed in detail alongside popular analytical methods and verified against available experimental results for various volume fractions. Good agreements have been observed for the presented model in comparison with the experimental results.

Abstract ID: 42DCASS-137

The Quantification of Porosity Distributions in Direct Metal Laser Sintered Inconel 718

Luke Sheridan (luke.sheridan.1@us.af.mil)
Air Force Research Laboratory
Sonya Sokhey
Wright State University
Joy Gockel
Wright State University
Onome Scott-Emuakpor
Air Force Research Laboratory
Tommy George
Air Force Research Laboratory


Additive manufacturing (AM), a developing approach to component fabrication, offers flexibility in design and reductions in cost that make it a promising alternative to traditional manufacturing techniques for many diverse applications. In AM components, non-optimal processing parameters may result in lack of fusion and gas-induced porosity which are common sites of fatigue crack initiation in aerospace components. Some of the AM input parameters that have been noted to influence the amount of porosity throughout the build are beam power, scan speed, and hatch spacing. Previous observations have noted that pore size and location have a significant influence on fatigue life, but no work has attempted to quantify the pore distributions in order to draw relationships with the life of the component. This initial investigation uses a design of experiments approach to quantify the porosity developed in a part due to varying process parameter combinations. Porosity within AM specimens will be characterized by multiple methods, and a summary of the collected data will be presented. This investigation is a subset of a larger effort tracing the influence of AM process variables to mechanical properties of a component, which will aid in qualifying AM components for aerospace applications.

Abstract ID: 42DCASS-162

Modeling of Additive Manufacturing Process to Determine the Feasibility of Using Low Coefficient of Thermal Expansion Materials

Chigozie Obidigbo (obidigbo.2@wright.edu)
Wright State University
Joy Gockel
Wright State University


The use of additive manufacturing (AM) in tooling enables low cost production of components to be fabricated by reduction in waste of raw material, increased design flexibility and a reduced lead time. Composite materials are used to produce aerospace components and require tooling that does not expand when subjected to thermal loads. Invar 36 is a popular metal tooling material known for its low coefficient of thermal expansion and is made of Fe-36wt.% Ni alloy. Because of these properties, it is a candidate for composite tooling where highly reliable and high precision materials are required. This work uses thermal finite element modeling of powder bed fusion AM processes as a tool to determine the feasibility of using Invar 36 powders for building components using AM. Results show that Invar 36 is a potential material for use in AM which will enable rapid tooling for composite structures.

Abstract ID: 42DCASS-174

Creep of Hi-Nicalon Fiber Tows at 700 ˚C in air and in Silicic Acid-Saturated Steam

Ronald Mitchell (jottemtrotter@yahoo.com)
Air Force Institute of Technology
Marina Ruggles-Wrenn
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-175

The influence of SLM process parameters and aging on shape memory response of Ni 50.8 Ti 49.2 alloy

Soheil Saedi (soheil.saedi@uky.edu)
University of Kentucky
Mohammad Elahinia
Univeristy of Toledo
Haluk Karaca
University of Kentucky


Additive manufacturing makes fabrication of components with complex geometries feasible. The combination of such fabrication flexibility with unique NiTi properties such as superelasticity and shape memory effect, makes selective laser melting (SLM) fabricated NiTi parts promising for enormous applications. The involving SLM process parameters such as laser power, powder layer thickness, scanning speed, spacing, and strategy play an important role in the evolution of grain formation, texture, etc of the fabricated parts. It is well-known that microstructural features can greatly affect the thermomechanical response of NiTi alloys. This study is a systematic work, concerning the effect of each fabrication parameter individually on microstructure and mechanical response of SLM Ni 50.8 Ti 49.2 . It is concluded that the right process window should be targeted to achieve desired and optimum performance. It will be shown that up to 5.5 % stabilized strain recovery is achievable without going through any additional treatment only with the selection of right process parameters. Additionally, a comprehensive aging study has been conducted to monitor the effect of aging time and temperature on functional properties of SLM NiTi. It will be shown that heat treatment techniques, can be employed effectively to enhance the shape memory effect and superelasticity of dense and porous SLM NiTi significantly.

Abstract ID: 42DCASS-183

Temperature Dependent Indentation Response of High Temperature NiTiHf Shape Memory Alloys

Peizhen Li (pnli222@uky.edu)
University of Kentucky
Haluk E. Karaca
University of Kentucky
Yang-Tse Cheng
University of Kentucky


In this study, Ni50.3Ti29.7Hf20 shape memory alloy was subjected to spherical indentation over a range of temperature (25 °C to 340 °C). A spherical indenter (tip radius, r = 25 µm) was used to indent the sample with selected maximum load level of 500 mN. Indentation hardness, contact modulus, and work recoverable ratio were investigated as a function of temperature. It is concluded that the indentation response can be used to determine transformation temperatures and whether the material is capable of superelastic deformation. Thermal cycling indentation experiments were conducted on aged Ni50.3Ti29.7Hf20 alloy. The results show that the indentation response was analogous to the thermal cycling under stress experiments but provide more information about the shape memory response. The changes in indentation response will be discussed in details and related to the mechanical behavior. It will be shown that, indentation can be used to provide rapid and very useful temperature-dependent information about the shape memory responses of alloys by using the energy-based analysis. NiTi and Ti alloys were also conducted with the high temperature indentation experiments. Indentation response of hardness, contact modulus and work recovery were also compared with NiTiHf alloy.

Abstract ID: 42DCASS-195

RF Impedance Sensing of DNA based on Graphene

Sheena Hussaini (sheena.hussaini@nokia.com)
Wright State University


Graphene has attracted much attention recently due its unique properties and potential applications. Biological sensing systems can utilize grapheme because of its unique 2D structure and electrical properties. The dynamic processes occurring in microscopic, mesoscopic, and macroscopic organisms play key roles in device sensing and can be effectively monitored by impedance characterization. In this work, on-chip integrated impedance bio-sensors are demonstrated using coplanar waveguides (CPWs) as the sensing platform. Absorption of chemicals such as Chitoson and DNA on graphene /graphene derivatives lead to remarkable shift in resonant frequencies. Substrate complex permittivity has also been extracted from momentum simulations. The imaginary part of the permittivity indicated significant leakage currents in the grapheme and its derivatives, Chitoson, and DNA.

Optimization

Abstract ID: 42DCASS-003

Persistent Intelligence, Surveillance, and Reconnaissance: A New Approach to the Persistent Monitoring Problem

Christopher Olsen (christopher.c.olsen@gmail.com)
Air Force Institute of Technology
Donald L. Kunz
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-007

CFD Analysis and Shape Optimization of the Internal Convective Cooling System of a Hypersonic Vehicle

Kan Liu (kanliu.1891@gmail.com)
Air Force Institute of Technology
Dylan Stelzer
The Ohio State University
Carl Hartsfield
Air Force Institute of Technology
David Liu
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-008

Comparison of Unified and Sequential-Approximate Approaches to Multifidelity Optimization

Dean Bryson (brysond1@udayton.edu)
University of Dayton
Markus Rumpfkeil
University of Dayton


Multifidelity approaches are frequently used in engineering design when high-fidelity models are too expensive to use directly and lower fidelity models of reasonable accu- racy exist. In optimization, corrected low-fidelity data is typically used in a series of sequential optimizations bounded by trust regions around the approximate model. Each sub-optimization is independent of the previous one, except for the starting design and trust region size. A new unified multifidelity quasi-Newton approach is presented that preserves an estimate of the inverse Hessian between iterations, determining search di- rections from high-fidelity data and using approximate models for single line searches. The proposed algorithm produces better search directions, maintains larger step sizes, and requires significantly fewer low-fidelity function evaluations than Trust Region Model Management. The multifidelity quasi-Newton method also provides an expected optimal point which is forward looking and is useful in building superior low-fidelity corrections. As part of the multifidelity framework, a warm starting technique is demonstrated to ini- tialize high-fidelity optimization when transition away from approximate models is deemed fruitful. The new approach is compared to Trust Region Model Management on a variety of analytic test problems using both polynomial and kriging correction functions. In sum- mary, the unified multifidelity quasi-Newton approach required fewer high-fidelity function evaluations than Trust Region Model Management in about two-thirds of the test cases.

Abstract ID: 42DCASS-011

Application of a Genetic Algorithm and K-Means Clustering for Planetary Exploration Using a UAV Swarm

Chase Hartman (hartmac4@mail.uc.edu)
University of Cincinnati
Seth Holsinger
University of Cincinnati
Kelly Cohen
University of Cincinnati


The exploration of planets is a task space agencies have been trying to tackle for decades. NASA is currently investigating the usage of unmanned aerial vehicles (UAVs) referred to as “Extreme Access Flyers” which are able to withstand the hostile conditions of space and “hunt and gather” samples from other worlds in places that are inaccessible to rovers. In this effort, an effective method is introduced when incorporating a swarm of collaborating UAV's equipped with instruments to analyze attributes of points-of-interest. The goal is to develop computationally efficient path planning algorithms for a UAV swarm to meet a minimum time cost function, where the problem is posed as a multiple traveling salesman problem. The solution approach is to cluster first by allocating the targets to individual UAVs and then solving the TSP per UAV. In order to achieve this, a hybrid strategy consisting of the combination of k-means clustering with a genetic algorithm implemented in MATLAB. Simulation results demonstrate the effectiveness of the developed methodology.

Abstract ID: 42DCASS-017

Controller Design and Energy Optimization for Lower Limb Exoskeleton

Ameya Chamnikar (chamnias@mail.uc.edu)
University of Cincinnati
Gaurav Patil, Manish Kumar, and Tamara Lorenz
University of Cincinnati


The population of the United States, ages 65 and older will increase from 40.2 million to 88.5 million in 2050. Therefore, it is necessary to develop the means to promote a high quality of life for those with vanishing abilities for independent performance of Activities of Daily Living (ADL). In this paper, a lower limb exoskeleton is proposed to assist a person in Sit-to-Stand (STS) transition. Besides applications in ADL, the proposed exoskeleton can also be used in space-suits for astronauts for potential aerospace applications. The lower limb exoskeleton is modeled as an inverse 3-DOF pendulum. For simplicity, we have assumed that the center of mass of each link lies at the respective geometric center. Dynamics of the system are derived using the Euler-Lagrange equation. The highly non-linear dynamics are linearized using an input-output feedback linearization technique. The controller for this linearized dynamic is designed using LQR and PD techniques. Simulation studies for the developed non-linear model of the exoskeleton shows that the LQR controller follows the desired trajectory more closely, i.e., with less error compared to a typical PD controller. Also, the total energy requirement of the controller is less in case of the LQR controller. Subsequently, a two-layer optimization technique to determine the trajectory that minimizes the energy requirement for the STS transition is discussed. A Genetic Algorithm (GA) generates and evaluates possible trajectories which demonstrates the stable sit-to- stand transition of the entire body. The LQR controller is then used to track the trajectory with low energy requirement. The GA then determines the trajectory that requires least overall energy.

Abstract ID: 42DCASS-035

A Robust Solution to Distributed Multi-Agent Path Planning Using Genetic Fuzzy Pareto and Game Theory

Mohammadreza Radmanesh (radmanma@mail.uc.edu)
University of Cincinnati
Rumit Kumar
University of Cincinnati
Manish Kumar
University of Cincinnati
Kelly Cohen
University of Cincinnati
Donald French
University of Cincinnati


In this paper, a new concept for solving multi-agent optimization problem using the Genetic Fuzzy pareto approach has been introduced. The solution is generated by genetic fuzzy pareto and Stackelberg game theoretic approach which uses a leader-follower concept to solve the distributed problem. Further, the changeability valuation is offered to such class of problems. The approach has been extended to multi-agent problems with uncertainties for demonstrating the robustness of the proposed decision making process. The stability of the system is evaluated by introducing perturbations. The process of decision making has been introduced theoretically and the claims have by verified via simulation of a simple multi-agent trajectory planning problem. During simulation, different scenarios with uncertainties present in the environment have been considered.

Abstract ID: 42DCASS-036

Area Coverage Based On Lévy Foraging Hypothesis Applied to Robot Swarm Emulating Ant Foraging Behavior

Aditya Deshpande (deshpaad@mail.uc.edu)
University of Cincinnati
Manish Kumar
University of Cincinnati
William Nafstad
University of Minnesota Duluth
Subramanian Ramakrishnan
University of Minnesota Duluth


With the advancements in computing, sensing, and communication technology, there is huge potential of application of autonomous UAV swarms in a number of domains that necessitates development of solutions to coordinated working of the swarm agents in uncertain environments. In this study, a swarm control law for area exploration inspired by the dynamical behavior of ant foraging is investigated with the inclusion of Lévy foraging strategy. Lévy flight, as opposed to traditional Brownian motion as source of randomness, has been used to model searching behaviors of several biological entities that have shown superior search performance in presence of scarce targets. Influence of different pheromone (the chemotactic agent that drives the foraging) threshold values for switching the swarm behavior between Lévy flights and Brownian motion is studied using a number of performance metrics such as area coverage, frequency of visit, entropy, and detection of dynamic targets. Results highlight the benefits of this control framework for search in areas of scarce resources.

Abstract ID: 42DCASS-064

Solving State Constrained Pursuit-Evasion Minimax Problems Using Semi-Direct Collocation

Ryan Carr (barakaldonights@gmail.com)
Air Force Institute of Technology
Richard Cobb
Air Force Institute of Technology
Meir Pacther
Air Force Institute of Technology
Scott Pierce
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-107

Topology Optimized Perforators Using Additive Manufacturing Techniques

Zachariah Provchy (zaprovchy@gmail.com)
Air Force Institute of Technology
Anthony Palazotto
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-134

Bayesian Inspired Multi-Fidelity Optimization - Transcending Traditional Bi-Fidelity Optimization

Christopher Fischer (fischer.36@wright.edu)
Wright State University
Ramana Grandhi
Wright State University


This work presents an optimal design technique which allows for the use of multiple simulation models of varying physics and complexity (fidelity levels). An advancement of traditional Surrogate Based Optimization (SBO) techniques is intended to alleviate the computational cost associated with structural and multidisciplinary design optimization while maintaining a high degree of accuracy typically associated with fully coupled, nonlinear, complex physics-based models. This methodology, termed Bayesian Influenced Low-Fidelity Correction Approach to Multi-Fidelity Optimization, utilizes a combination of surrogate modeling, Bayesian statistics, and Trust Region Model Management (TRMM) techniques. A novel Bayesian Hybrid Bridge Function (BHBF) was developed to serve as the low-fidelity correction technique. This BHBF is a Bayesian weighted average of two standard bridge functions, additive and multiplicative. The correction technique is implemented in parallel with a modified Trust Region Model Management (TRMM) optimization scheme. It is shown that optimization on the corrected low-fidelity model converges to the same local optimum as optimization on the high-fidelity model in fewer high-fidelity function evaluations and ultimately lower computational cost. This work also extends the low-fidelity correction optimization beyond the traditional bi-fidelity (limited to 2 fidelity levels) optimization to that of a novel approach to handling optimization with multiple (2 or more) fidelity objective and constraint functions with commercial optimization solvers. It is shown that implementation of this Bayesian low-fidelity correction optimization approach results in high-fidelity results at a reduced computational cost. This is demonstrated on computationally different engineering design problems.

Abstract ID: 42DCASS-141

Topology Optimization of Thermoelastic Structures via Level-set Methods

David Neiferd (david.neiferd@wright.edu)
Wright State University
Ramana V. Grandhi
Wright State University


Today, there exists a need for high performing, reliable thermal structures for use in the extreme thermal environments found on hypersonic air vehicles, atmospheric reentry platforms, and low-observable embedded engine aircraft. To date, structures designed to operate in these environments represent non-optimal solutions. They have frequently failed due to a lack of understanding of the fundamental interactions of different physics in this complex design domain. By using multiphysics topology optimization, we propose to tailor thermoelastic behavior of thermal structures where traditional minimum compliance objectives result in suboptimal structures. In this work, a stress-based explicit level-set approach to topology optimization, as opposed to the traditional density-based method is used. A hole nucleation technique is added to the explicit level-set formulation to increase the design freedom. The optimization problem is solved using the Method of Moving Asymptotes (MMA) allowing faster convergence over traditional Hamilton-Jacobi level-set methods. By considering stresses in the optimization, either through the objective function or constraints, the failure criterion, including thermoelastic effects, is more directly controlled resulting in more optimal designs versus minimum compliance.

Abstract ID: 42DCASS-149

Atmospheric Impacts on Optimal Range of Hypersonic Boost-Glide Vehicles

Melissa Dunkel (melissa.dunkel@afit.edu)
Air Force Institute of Technology
Richard Cobb
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-165

Genetic Algorithm Optimization of Geosynchronous Earth Orbit Space Situational Awareness Systems via Parallel Evaluation of Executable Architectures

Steven Wachtel (steventwachtel@gmail.com)
Air Force Institute of Technology
Jordan Stern
Air Force Institute of Technology
John Colombi
Air Force Institute of Technology
David Meyer
Air Force Institute of Technology
Richard Cobb
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-166

Simplex Methods for Optimal Control of Unmanned Aircraft Flight Trajectories

Michael Zollars (michael.zollars@afit.edu)
Air Force Institute of Technology
Richard G. Cobb
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-169

Level Set Based Cellular Division Method for Structure Conceptual Design

Hao Li (li.138@wright.edu)
Wright State University
Ramana Grandhi
Wright State University


In the typical process of aircraft design, there is a big gap between conceptual design phase and design refinement. In order to bridge the gap, this research proposes a new framework performing simultaneous sizing, shape and topology optimization integrating cellular division method with level set method. Cellular-division based design methodology is an innovative biologically-inspired layout and topology optimization method aiming to generate unconventional aircraft structural configurations. Multi-objective Genetic Algorithm is used to mimic the process of how the nature develops the best design. A typical drawback of such method is that the shape changes are based on the physics of the cellular dynamics in living organisms, but not the physical behavior of engineering system, which results in the low convergence rate. In this research, conventional level-set based shape optimization method is integrated to replace the cellular dynamic process by driving the boundaries of the structure implicitly based on the system behavior. The methodology is demonstrated on benchmark problems of structural stiffness designs.

Abstract ID: 42DCASS-172

Optimal Finite-Thrust Guidance for Insertion into and Maintenance of Relative Motion Teardrop Trajectories

Eric Prince (eric.prince@afit.edu)
Air Force Institute of Technology
Richard Cobb
Air Force Institute of Technology


Awaiting Public Release Clearance.

Space

Abstract ID: 42DCASS-006

BAEROCATS: University of Cincinnati Spacecraft Design Capstone

Chirau Patel (patelcj@mail.uc.edu)
University of Cincinnati
Chase Hartman
University of Cincinnati
Justas Jodele
University of Cincinnati
Gaurav Patel
University of Cincinnati


NASA’s Student Launch Initiative (SLI) competition is an annual research-based, competitive, experiential exploration project that provides the opportunity to gain experience in relevant, cost-effective research & development, fabrication & integration, testing & evaluation, and systems engineering practices through the development of a rocket and payload designed to meet competition requirements. In order to successfully complete the mission, the launch vehicle must reach an altitude of 5,280 feet above ground level and release a payload capable of identifying three tarps placed near the launch pad in real time. The payload must also be capable of remaining upright by mission completion. The University of Cincinnati’s Baerocats team will present their approach to this year’s NASA SLI competition, surveying alternatives explored and detailing the development program and initial testing, including full-scale rocket launches. Additionally, the team will share important lessons learned while working on this two-semester undergraduate capstone project.

Abstract ID: 42DCASS-018

Formation Flight of Earth Satelites on KAM Torus using Classical Orbital Elements

Marissa Reabe (marissa.reabe@us.af.mil)
Air Force Institute of Technology
William Wiesel
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-019

Hall Effect Thruster Characterization Through Potential, Magnetic, and Optical Measurements

Nick Hyatt (nick.hyatt12@gmail.com)
Air Force Institute of Technology
Carl Hartsfield
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-020

Thrust Coefficient Losses in Additively Manufactured Low Thrust Nozzles

Christopher Tommila (christopher.tommila@afit.edu)
Air Force Institute of Technology
Carl Hartsfield
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-050

Utilizing Future Launches of Centaur Rocket Bodies to Deorbit Large Debris in Polar and Sun-synchronous Orbits.

Krista Roth (krista.roth@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-053

Analysis of Model Fidelity Impact on Light Curve Prediction for Geostationary Satellites

Rachel Oliver (racheloliver93@gmail.com)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-088

Dynamic Modeling of Spacecraft Behaviors

Justin Sadowski (justin.sadowski.2@us.af.mil)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-092

Sensing Satellite Articulation using Computer Vision

David Curtis (david.curtis@afit.edu)
Air Force Institute of Technology
Richard G. Cobb
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-113

Maneuver Detection and Characterization Using Wavelets for Geosynchronous Spacecraft

Brian Pitman (bwpitman@gmail.com)
Air Force Institute of Technology
Richard G. Cobb
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-126

Satellite Signal Classification using RF-DNA Techniques

Patrick Dunkel (patrick.dunkel.1@us.af.mil)
Air Force Institute of Technology
Eric Swenson
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-127

Constellation Architecture Design for Persistent Space Situational Awareness of Direct Ascent to Geosynchronous Orbit

Laura Broch (laura.broch@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-157

Space Mission Design in a Multi-Body Environment: An Investigation of High-Altitude Orbits for Transfers, Reconstitution, and Unconventional Missions

John Brick (john.brick@afit.edu)
Air Force Institute of Technology
Christopher D. Geisel
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-164

An Analysis of Radio-Frequency Geolocation Techniques for Satellite Design

Daniel Barnes (daniel.barnes@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-167

Testing and Evaluating Deployment Profiles of the Canisterized Satellite Dispenser (CSD)

Stephen Tullino (stephen.tullino@afit.edu)
Air Force Institute of Technology
Eric D. Swenson
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-170

Evaluation of Networked Satellite Command & Control Via Internet Conduit

Cameron Cunningham (cameron.cunningham@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Structures

Abstract ID: 42DCASS-031

Analysis of Additively Manufactured Metal Lattice Structures by Linear Finite Element Analysis

Christopher Box (christopher.box@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-058

Optimal Design of a Hexakis Icosahedron Vacuum Based Lighter than Air Vehicle

Joseph Schwemmer (jrschwemmer@gmail.com)
Air Force Institute of Technology
James Chrissis
Air Force Institute of Technology
Anthony Palazotto
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-094

A Study of Quasi-Static and Dynamic Analyses of a Hexakis Icosahedron Frame for Use in a Vacuum Lighter Than Air Vehicle

Jordan Snyder (jordan.snyder@afit.edu)
Air Force Institute of Technology
Anthony Palazotto
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-100

Structural Analysis of a Composite Truss 3D Printed in Space

Jacob Downey (jacob.downey@afit.edu)
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-101

The Evaluation of a ROM for a Piezoelectric Shaker Table

Anthony Palazotto (palazotto@sbcglobal.net)
Air Force Institute of Technology
Randall Hodkin
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-128

Evaluating the Inclusion of an Optimized and Deformable Internal Structure on an Airfoil’s Aerodynamic Response

Joshua Lee (joshua.lee@afit.edu)
Air Force Institute of Technology
Anthony Palazotto
Air Force Institute of Technology
Donald Kunz
Air Force Institute of Technology


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-144

Longitudinal Damage Detection in a Beam Using Lamb Wave Based Finite Element Analysis

Chad Hale (chad.hale@afit.edu)
Air Force Institute of Technology
Chan Yik Park
Agency for Defence Development
Anthony N. Palazotto
Air Force Institute of Technology


Awaiting Public Release Clearance.

Thermal Management

Abstract ID: 42DCASS-079

Development of a Thermal Testing Apparatus for Rigid, High Porosity, Materials using Comparative Cut-Bar Methodology

Marcus Irvan (mlir222@g.uky.edu)
University of Kentucky
Christopher T. Barrow Ashley N. Keen
University of Kentucky
John F. Maddox
University of Kentucky


The insulation materials used in thermal protection systems are crucial for protecting payloads during the reentry of space probes and hypersonic vehicles. The insulation that is used has high porosity and is composed of multiple fibers. Because of this, the three primary modes of heat transfer through the material during reentry are solid conduction, gaseous conduction through the fibers, and radiation from fiber to fiber. The conduction through the fibers varies with temperature due to the varying thermal conductivity of the solid; the contribution from gas conduction is dependent on temperature, pressure, and gaseous composition; and the contribution of radiation through the material has a non-linear dependence on the temperature difference across the material. This dependence on environmental conditions is the main reason that accurate modeling is so hard to achieve when dealing with these materials. By using physics based modeling instead of the traditional route, it is possible to more efficiently measure these modes of heat transfer without the need to replicate the conditions that the insulation material will undergo when in use on hypersonic and reentry vehicles. With the use of a vacuum chamber and a comparative cut-bar apparatus it is possible to isolate each of these different modes and analyze the heat transfer effects of each individually on the sample and this data allows for the derivation for a physics based model for the insulation material.

Turbomachinery

Abstract ID: 42DCASS-005

Fillet Influence on Transonic Axial Compressor Performance Predictions at Multiple Tip Clearance Heights

Rebecca Howard (rebecca.howard.1@us.af.mil)
Air Force Research Laboratory
Michael List
Air Force Research Laboratory
Steven Puterbaugh
Universal Technology Corp.


Inclusion of fillet geometry in compressor simulations was demonstrated to have a substantial influence on the choked massflow prediction for a low aspect ratio, transonic axial compressor stage. To demonstrate fillet influence, CFD simulations were performed with and without fillets, each at three different tip clearance heights, and compared with measured data from compressor tests performed in the Compressor Aero Research Lab (CARL) in the Turbine Engine Division at the Air Force Research Laboratory. The computational results for the stage simulations without fillet geometry over-predicted the experimentally measured choked massflow by an average of 3% for the three tip clearance configurations. When the fillet geometry was included in the computational model for the rotor blades and stator vanes, the error in massflow was reduced, resulting in calculated values that averaged 1% above the measured value for the three tip clearances. Circumferentially averaged profiles at the rotor and stator exit were compared for the simulations with and without fillets, showing differences in flow behavior due to the presence of the fillets. The computational investigation demonstrated the influence of fillets on the choked massflow prediction and determined that the inclusion of fillets has a significant impact for low aspect ratio blade rows.

Abstract ID: 42DCASS-012

A Computational Study of a Through-Flow Wave Rotor’s Compression Efficiency

Michael McClearn (michael.mcclearn.1@us.af.mil)
Air Force Research Laboratory
Fred R. Schauer
Air Force Research Laboratory
John Hoke
Innovative Scientific Solutions Inc.


Recently, a renewed interest has developed in wave rotor technology due to the advent of additive manufacturing with high temperature metals. This renewed interest stems from the potential benefits of wave rotors at low mass flow rates when compared to conventional turbomachinery. Wave rotors operate on fundamentally different operating principles and are not subject to the same sorts of efficiency drops that plague turbomachinery at these low flow rates. A through-flow wave rotor was previously designed for study at these flow rates, however, due to the large amounts of exhaust gas recirculation, quantifying the compressive efficiency experimentally is difficult at best, therefore; computational modeling is used to determine the efficiency. The previously designed wave rotor was modeled at a variety of off-design conditions spanning a 10,000 RPM rotor speed change and approximately 400 K combustor exit temperature over mass flow rates between 1.5 and 1.65 kg/min of air. Although many of the efficiencies are only comparable to turbomachinery, the methodology for analyzing and plotting these efficiencies varies greatly and will be the focus of this presentation. Compressor maps for typical turbomachinery generally utilizes axes plotting an overall pressure ratio on the y-axis and corrected mass flow on the x-axis. For a wave rotor, this does not work as they do not ingest more mass flow simply because rotor speed increases. Instead they must be plotted against rotor speed to see something resembling efficiency islands; this alone does not adequately describe the complex factors dominating the compressive efficiency. This work proposes that wave rotor cycle efficiencies be plotted with respect to sonic velocity in order to compare different rotors while individual rotors are mapped with rotor speed for operation.

Abstract ID: 42DCASS-073

Review of Turbine Endwall Losses

Mary Jennerjohn Christianer (mjennerj@slu.edu)
St. Louis University
Mark McQuilling
St. Louis University


In efforts to reduce fuel consumption and weight of a gas turbine engine, the gas turbine industry is continually pushing design boundaries for these reasons. For the turbine section, this translates to fewer blades leading to increased loadings and Mach numbers towards low supersonic ranges. These design alterations cause an aerodynamic concern: losses. An aerodynamic loss is known as the reduction in total pressure, or the production of entropy. There exist three different characterizations of aerodynamic losses in the turbine: tip-leakage loss, endwall loss, and profile loss. Endwall losses have seen to accrue up to 50% of the total pressure losses in the turbine section, which is the focus of this presentation. This review aims to educate the audience on the endwall flow structure while shedding light on the gaps of understanding. The author will present first the current understanding of the endwall flow structures, then discuss the different design parameters which can impact such flow structures including both experimental and computational results from the literature. The final section will briefly touch on the current mitigation strategies.

Abstract ID: 42DCASS-099

Surface Mounted Thin Film Sensors for Unsteady Flow Measurements in a Low Speed Wind Tunnel

Emma Veley (emma.veley@us.af.mil)
Air Force Research Laboratory
Christopher Marks
Air Force Research Laboratory
Rolf Sondergaard
Air Force Research Laboratory
Richard Anthony
Air Force Research Laboratory
Mitch Wolff
Wright State University


The flow field at the junction of a highly loaded low pressure turbine (LPT) blade and endwall is comprised of complex three dimensional flow structures. Time resolved measurement of these near wall flow features in a turbine passage is challenging because the data-taking instrumentation alters the flow field. This study verifies the use of thin-film heat flux gauges adapted to take unsteady flow data in a low-speed wind tunnel (LSWT). The flexible thin-film conforms to contours, thereby unobtrusive to boundary layer flow. The sensor is connected to a constant temperature anemometer system and uses similar calibration to a hotwire. Several experimental arrangements were tested in a small LSWT to validate the ability of the sensor in this configuration: mainly flow behind a cylinder. The sensor produced shedding frequency data comparable to an adjacent hotwire. When non-dimensionalized as Strouhal and Reynolds numbers, the data was comparable to values found in literature. The intent of the initial research is to conduct validation experiments in a small LSWT, working towards taking measurements in a low speed linear cascade of LPT blades. The new application of the thin film sensor enables time resolved near wall flow measurements on LPT passage surfaces providing new insight into the unsteady behavior of endwall flow structures and interaction between the blade suction surface boundary layer and secondary flow.

Abstract ID: 42DCASS-114

Performance of Partitioned Pulse Detonation Combustors-Axial Turbine System

Ethan Knight (knighten@mail.uc.edu)
University of Cincinnati
Andrew St. George
University of Cincinnati
Robert Driscoll
University of Cincinnati
Vijay Anand
University of Cincinnati
Ephraim Gutmark
University of Cincinnati


Over the past 20 years there have been a considerable amount of studies focused on power generation of pressure gain combustors. Pulse detonation combustors (PDC) have been investigated greatly over this period but there remains little information about power generation of these devices. In previous studies, pulsed-air was fed through a turbine at the frequency and flow rate that matched that of a pulsed detonation engine. However, turbine performance for pulsed-air conditions is a function of corrected flow rate and the range of flow angles experienced by the rotor. The frequency of the pulsed-air has little effect on the power generation through this method. However, for the case of the detonating flow, there is a maximum volumetric flow rate for a given pulse frequency, equal to the product of the combustor volume and the firing frequency (exceeding this volumetric flow rate will result in detonation propagating beyond the combustor and through the turbine). Previous studies at the University of Cincinnati that aimed to analyze power generate of detonative waves achieved significant power extraction despite the considerable high amplitude unsteadiness of the detonation process. Nevertheless, these studies used a considerable amount of bypass air to dampen fluctuations in the detonative flow. This study characterizes power extraction from multiple PDCs to an axial turbine system for wide range of operating conditions without a plenum based inflow. For this study, six PDCs were attached to the valve system from an in-line, 6 cylinder internal combustion engine. These valves allowed for the equivalence ratio and fill fraction to be carefully controlled throughout the experiment, as well as providing ignition timing for optimal power generation. The gases were then fed into a pipe, long enough to ensure the deflagration to detonation transition (DDT) required for the system to be considered a PDE. Once the detonation formed, the pressure of the exhaust gasses would reach between 7 and 8 bar. High pressure flow from the combustion was fed into a turbine through a series of shallow curved tubing to ensure little to no fluidic losses. Since the ignition timing was staggered, only one PDC exhausted into the turbine at a time which provided the turbine with a intake flow of 6 times the operating frequency of each PDC causing a more uniform flow. Kulite and PCB pressure sensors and ion probes were positions along the length of the PDCs and a torque cell was used to obtain the power generation from the turbine’s output. For every test the ion probes were used to verify the existence of detonating flow within the PDCs. Pressure sensors provided data that conveyed a pressure gain of 6-7 bar although the spinning rotor of the turbine did have an effect on their overall profile. Regardless, there is no evidence indicating that rotor speed has an effect on the flow behavior upstream of the turbine. Changes in operating frequency of the tests resulted in greater overall pressures for higher frequencies but the peak pressures remained the same. During operation, the power output of typical tests would take 5-10s to reach steady state. With the data obtained, thermal Brayton and Fickett-Jacobs (detonation) cycle efficiencies were found for operation at 5 and 10 Hz with the 5 Hz runs proving to be more efficient by a small margin. Moreover, the thermal efficiencies of the runs were greater than the turbine efficiency stated for a Brayton cycle. For these cases, the detonation cycle provides a tangible overall improvement to the cycle by utilizing detonative combustion. Brayton cycle efficiencies are included to highlight that a relatively inefficient turbine with a detonation combustor can outperform a relatively efficient turbine with a conventional combustor at these low cycle pressure ratios.

Abstract ID: 42DCASS-184

T-Blade3: A Novel Curvature-Based Design Tool for High Performance Rotating and Fixed-Wing Devices

Karthik Balasubramanian (balasukt@mail.uc.edu)
University of Cincinnati
Mark G. Turner, Kiran Siddappaji
University of Cincinnati


Parameterization and generation of highly differentiable shapes is necessary for design and optimization of high performance rotating devices and wings. Curvature and slope-of-curvature continuity of the aerodynamic surface is crucial to achieve smooth surface pressure distributions which can minimize losses and unintentional flow separations. T-Blade3 is a novel, open-source 3D blade and wing geometry generator which implements a parsimonious parametrization scheme while enabling sufficient local control through the use of B-spline control points. Slope-of-curvature continuous airfoil sections are created by superimposition of a thickness distribution over a camber line. These 2D sections are then stacked to create space curves which can then be lofted in CAD to produce smooth surfaces. Novelty of the tool is highlighted by camber line parameterization which requires user specification of second derivative control points, which are then splined and twice integrated to obtain the camber line. The chord-wise control along with B-spline based span-wise control of the 3D shape results in a large design space, encompassing complex shapes with inherent slope-of-curvature continuity. Various design and optimization studies have used T-Blade3, which is now available as an open-source code through GitHub, to explore a wide spectrum of both conventional and novel shapes for rotating devices. The ability to link T-Blade3 to an optimizer and the ease of implementation of such a tool-chain for both 2D and 3D single and multidisciplinary optimization have been demonstrated in past studies. The capabilities of T-Blade3 have been further expanded to incorporate isolated airfoil and wing design including specification of a twist distribution as opposed to section metal angle specifications which are crucial for most rotating devices. The thickness parameterization scheme has been redefined to allow treatment of leading, trailing edge definitions and rest of the blade as a single entity. This allows for local droop and hook-like shape while maintaining slope-of-curvature continuity of the airfoil curve, further expanding the design space and enabling to achieve cutting edge improvements in efficiency. CFD based optimization using T-Blade3 has been carried out and presented to demonstrate the capabilities of the parameterization. The usage of the tool and examples of various novel geometries for both rotating and non-rotating devices are presented to highlight the generality of application.

Uncertainty Quantification

Abstract ID: 42DCASS-010

Construction of Multi-Fidelity Surrogate Models for Uncertainty Quantification

Markus Rumpfkeil (Markus.Rumpfkeil@udayton.edu)
University of Dayton
Philip Beran
Air Force Research Laboratory


In order to enable technological advancements in aircraft designs the conceptual design phase should include both low- and high-fidelity engineering analyses to enable reasonable computational cost and accuracy trade-offs for design space exploration and uncertainty quantification. In this presentation it will be demonstrated that accurate multi-fidelity locally optimized surrogate (MFLOS) models can be obtained at lower computational cost than global single-fidelity ones. Once an accurate surrogate model is constructed it can be used for evaluating a large number of conceptual designs and/or Monte-Carlo samples for uncertainty quantification since one can capitalize on the computationally cheap surrogate function evaluations. To demonstrate the power of the MFLOS models analytic test functions as well as a transonic NACA 00xx airfoil case will be employed.

Abstract ID: 42DCASS-105

Investigation of Aerothermoelastic Model Sensitivity under Transitional Fluid Loading in High Speed Flow

Zachary Riley (zachary.riley.ctr@us.af.mil)
Universal Technology Corp.
Benjamin Smarslok
Air Force Research Laboratory


Laminar-turbulent transition significantly increases the aerothermodynamic loads on a hypersonic vehicle. Thus, accurate prediction of boundary-layer transition is essential to reducing structural weight and avoiding overly conservative designs. This already challenging problem is exacerbated by the coupling between the aircraft’s structural response and the fluid loading. Specifically, the surface of a hypersonic aircraft is expected to deform during flight, augmenting the aerodynamic heat flux and pressure. Boundary-layer stability is dependent on the surface pressure gradient and temperature which both vary with structural deformation. Therefore, during flight the region of transitional flow is coupled to the evolving structural state. In a previous study, one of the authors examined how prescribed transitional fluid loading impacts structural response using a fundamental 2-D aerothermoelastic model under the assumption of an aluminum panel with simple supports. Specifically, an existing model was extended to examine the impact of transition onset location, transition length and transitional overshoot in heat flux and fluctuating pressure on panel response. This study found that transitional flow can lead to more severe aerothermal loading and structural responses that exceed those predicted assuming turbulent conditions. The objective of this paper is to more rigorously quantify the uncertainty in the predicted structural response associated with transitional boundary-layer effects. Specifically, probability distributions will be specified for the input parameters of the transitional models. These include the transitional N factor (i.e., the transition onset location), the length of transition, and the potential for overshoot in the aerothermodynamic loads. A sensitivity analysis will be performed to assess the importance of the above variables on the predicted structural response. Completion of this study will provide insight into aspects of boundary-layer transition that predominantly affect structural response prediction and help to identify areas that require improved model accuracy.

Abstract ID: 42DCASS-121

Calibration and Uncertainty Quantification of Open Source AVCOAT Similar VISTA Material Model Using Bayesian Inference

Przemyslaw Rostkowski (rstkwsk2@illinois.edu)
University of Illinois at Urbana-Champaign
Marco Panesi
University of Illinois at Urbana-Champaign


Typical model calibration schemes rely on finding parameter values through the solution of the inverse problem. This task is commonly achieved through deterministic methods that rely on minimization of an error metric between computed model output and observed data. However, these approaches fail to give insight into the uncertainty of the calibrated parameter values and inadequacy of the model in replicating physical phenomena being studied. In the presented material we use Bayesian inference to calibrate model parameters of VISTA, the open source AVCOAT similar material database formulated for use alongside the KATS ablation toolbox, with open source Apollo-era temperature data. In addition, a global sensitivity analysis is performed in order to identify most important and negligible parameters; thermal conductivities of the material’s virgin and char states are shown to have the largest influence on the computed temperature output. The statistical inverse problem is then carried out which concurrently quantifies the uncertainty due to physical model inaccuracies as well as each calibrated parameter. In order to quantify the uncertainty of the final model output, uncertainties associated with calibrated parameters and model shortcomings are propagated through the statistical forward problem. It is seen that calibration and forward propagation procedures yield a most probable temperature output that is an improvement upon the results obtained prior to calibration of VISTA.

Abstract ID: 42DCASS-146

A Non-Deterministic Metamodeling Framework for Aircraft Design Under Mixed Uncertainty

Daniel Clark (clark.333@wright.edu)
Wright State University
Ha-Rok Bae
Wright State University


As the aerospace industry moves closer to the “Digital Twin” era, Uncertainty Quantification will become a large component of total computational cost. In this talk, a metamodeling technique is proposed to capture distinguishable epistemic (reducible) and aleatory (irreducible) uncertainties. The proposed Non-Deterministic Kriging (NDK) framework utilizes a kernel process to first characterize the aleatory uncertainty of the design space, followed by a maximum likelihood estimate of the aleatory uncertainty. Comparisons will be made to deterministic kriging, stochastic kriging, and regression kriging. The power of NDK will be discussed though fundamental examples with varying degrees of dimensionality.

Unmanned Aerial Systems

Abstract ID: 42DCASS-014

Air Transport of Rescue Pod using Collaborating UAVs during Emergency Evacuations/Disaster Management

Shraddha Barawkar (barawksd@mail.uc.edu)
University of Cincinnati
Manish Kumar and Kelly Cohen
University of Cincinnati


Abstract—Natural disaster is a cause for substantial loss of life and property. Recent advances in unmanned aerial technology provides tremendous promise for emergency/rescue operations during disasters. However, many technological gaps exist that prevent them to be deployed in real world. These gaps consist of unavailability of efficient control methodology, ability to handle sensor uncertainties, and unavailability of low-cost high-payload aerial platform. The main objective of this research is to develop a deployable, safe and reliable collaborating multi-UAV system that can provide autonomous airlift during disasters for rescue evacuations. Multi-UAV lift provides benefits such as low-cost functionality, quicker response, and more stability as opposed to a conventional single high payload aerial vehicle. A novel force feedback controller (FFC), based on leader-follower approach, is proposed in this research for multi-UAV collaboration. The FFC overcomes the limitation of traditional position based leader-follower controller which is difficult to be implemented practically due to high inherent errors in position data obtained using GPS. In the proposed control scheme, the leader follows a trajectory using a PID controller while a fuzzy logic FFC controller is implemented for the follower UAV that controls the follower based on minimization of the contact forces and torques acting on it.

Abstract ID: 42DCASS-024

Iterative Genetic Fuzzy Logic System for Solving the Aircraft Conflict Resolution Problem

Anoop Sathyan (sathyaap@mail.uc.edu)
University of Cincinnati
Nicholas D. Ernest
Psibernetix Inc
Loïc Lavigne and Franck Cazaurang
Univ. Bordeaux, IMS-lab, UMR CNRS 5218, F-33405 Talence, France
Manish Kumar and Kelly Cohen
University of Cincinnati


It is extremely important to have an efficient mechanism to determine optimum conflict-free paths for aircraft in order to improve airspace safety. This research focuses on developing a genetic fuzzy logic based approach for solving the aircraft conflict resolution problem where the objective is to obtain conflict-free trajectories for aircraft in a circular airspace while minimizing the cost of maneuver. Uncertainties in the velocity and the maneuver parameters are also considered which causes each aircraft's position at any instant to be within a region of uncertainty represented by a convex hull. A new and unique architecture for Fuzzy Logic System (FLS) is used that consists of a hidden layer of neurons and a layer of decoupled Fuzzy Inference Systems (FISs) which is capable of iteratively traversing the search space to find a near optimal solution. For this purpose, an artificial intelligence (AI) called EVE is used to tune the parameters of the system and once it is trained, its capability is evaluated on a set of test scenarios. The results obtained for the five and ten aircraft problem for different levels of uncertainty are compared to those obtained by directly applying Genetic Algorithm (GA). The FLS is able to obtain near-optimal solutions comparable to those of GA at a fraction of the computational cost.

Abstract ID: 42DCASS-023

Offline Evolutionary Learning for Vision-based Landing of a Quadcopter using Fuzzy Logic

Nicklas Stockton (stocktno@mail.uc.edu)
University of Cincinnati
Manish Kumar
University of Cincinnati
Kelly Cohen
University of Cincinnati


The purpose of this work is to demonstrate the efficacy of using a fuzzy logic based approach to successfully perform the landing maneuver of a small multirotor aircraft on a dynamic target at low speeds. The corrective control of the vehicle is exerted based on visual input data from an on-board camera system. Given the constraint of the vehicle size, the computational complexity of image processing and control correction must be low in order to be performed on an on-board computer system-on-a-chip. The design of the controller around visual sensing reduces the dependency of special sensors on-board either the vehicle or target for rendezvous. The use of fuzzy logic enables the controller to adapt to different plant behaviors dynamically without the need for gains scheduling. Fuzzy systems can approximate highly nonlinear functions of arbitrary complexity; fuzzy control, therefore, is a apt fit for highly nonlinear systems which are difficult to control using linear techniques. In addition to being adaptable to nonlinearities, fuzzy logic requires only very basic mathematical operations, making it ideal for platforms which either have limited computational capabilities or need to respond quickly to dynamic conditions. In order to properly tune the fuzzy controller, a simulation has been developed which is used to evaluate the efficacy of the controller. This simulation is the basis of a genetic algorithm which searches for a solution. The research is being performed using Robot Operating System to integrate sensing with control motor actuation. All control is performed with Python using OpenCV for image processing. This paper details the simulation setup and image processing algorithms; it also concerns the design and tuning of a simple PID controller. Once the genetic algorithm finds an acceptable solution to the control problem, the control structure can be loaded onto physical hardware for flight testing.

Abstract ID: 42DCASS-025

Route Planning for Multiple Land Surveying Drones

Brandon Kunkel (kunkelbm@mail.uc.edu)
University of Cincinnati
Kelly Cohen
University of Cincinnati
Manish Kumar
University of Cincinnati


The unmanned aerial vehicle (UAV) is a versatile platform that is breaking out as a disruptive technology. One of the many civilian applications for UAVs is to automate land monitoring and surveying. Land surveying is typically performed by teams of two to four surveyors, and current surveying techniques leave room for improvement. Using a swarm of 5 collaborating UAV’s to automate the land surveying process will save surveyors’ time and money, as well as reducing potential risks to equipment and personnel. Optimizing the flight path for a UAV to survey an area of land can be done using a variety of methods. This paper investigates using genetic algorithm based three-dimensional path planning of a surveying UAV, given topographical elevation data, a prescribed area of operations as well as additional constraints such as no fly zones and other air traffic/FAA restrictions. A genetic algorithm traveling salesman solver is used to calculate the constrained time-optimal path to survey designated area of interest for two cases, one UAV and up to 5 UAVs. Figures of merit and the cost function that drive the optimal path-planning problem are explored. Route planning methodology is developed and simulation results are presented.

Abstract ID: 42DCASS-038

Localization and Navigation of Aerial Vehicles in Indoor Environments using Actively Deployed Radio Beacons

Suyash Kulkarni (kulkarsy@mail.uc.edu)
University of Cincinnati
Manish Kumar
University of Cincinnati
Kelly Cohen
University of Cincinnati


The localization of aerial vehicles in an indoor environment is difficult due to absence of GPS signals. The localization problem in indoor environments is usually solved using a technique called Simultaneous Localization And Mapping (SLAM) that utilizes on-board sensors such as Inertial Measurement Units, Lidar, proximity sensors, or vision sensors. However, these techniques often prove to be insufficient in complex and dynamic environments. An example of such environment is a tunnel which does not provide distinguishing environmental features for the SLAM algorithms to work properly. This paper proposes the use of actively deployed radio beacon to localize the aerial vehicle, using the beacons as landmarks. The beacons are actively deployed by the vehicle as the vehicle navigates through the environment. Using an Extended Kalman Filter, the data from the on-board sensors is combined to that of localization information obtained from the beacons to reduce the error in localization. The vehicle monitors the estimate of its localization error which is then used to make decisions to deploy successive beacons.

Abstract ID: 42DCASS-029

Vision based target geolocation and tracking using a quadrotor platform

Sarthak Kukreti (kukretsr@mail.uc.edu)
University of Cincinnati
Manish Kumar
University of Cincinnati
Kelly Cohen
University of Cincinnati


In this research, we propose a comprehensive, efficient, and accurate framework for the geolocation of targets. We investigate algorithmic development for the use of UAV to detect and track ground based objects. In order to achieve this goal, we divided this problem into two sub problems: 1) object tracking, and 2) target geolocalization. Using the pixel location of the target in an image, with measurements of UAV position and attitude, and camera pose angles, the target is localized in world coordinate systems. The geolocation and tracking methods have been implemented and simulation results are presented demonstrating the localization and tracking of a target as close to its actual location. Unmanned vehicles are prime candidates for tasks involving risk and repetition. The simplified goal of many of these tasks is to image and/or locate a target for tracking, reconnaissance, or delivery purposes. Therefore the ability to accurately determine the location of a ground-based object using aerial images would contribute to the success of these tasks. This research presents a method of determining the location of an object in world/inertial coordinates using a camera on-board a small, multirotor UAV. Due to low-altitude and low-velocity flight capabilities, these rotorcrafts allow significant advantage in solving the problem. It is therefore reasonable to explore localization methods involving more robust and less-expensive UAV platforms. In addition, this research will use the video stream as means of localizing objects found in its field of view. Vision-based object localization represents as effective use of on-board resources and provide information for many practical applications. We present a general approach for target localization and tracking, provide an analysis of possible error sources, and demonstrate the effectiveness of the approach.

Abstract ID: 42DCASS-037

PDE Based Trajectory Planning for Unmanned Air Vehicles

Mohammadreza Radmanesh (radmanma@mail.uc.edu)
University of Cincinnati
Manish Kumar
University of Cincinnati
Kelly Cohen
University of Cincinnati
Donald French
University of Cincinnati


This paper proposes a novel physics-based method for three-dimensional UAV trajectory planning in dynamic hostile environments using Partial Differential Equations (PDEs) . The proposed method exploits the dynamical property of fluid flowing through a porous medium. This method evaluates risk to generate porosity values through-out the computational domain. The path which encounters the highest porosity values determines the path from the point of origin to the goal position. The best trajectory is found by determining the resultant motion of fluid in porous media by obtaining analytical solution of the PDEs representing the fluid flow over the streamlines generated. Constraints due to Unmanned Aerial Vehicle (UAV) dynamics, obstacles, and predefined way points are applied to the problem after solving for the best trajectory to find the optimal path. This method shows near-optimality and much reduced computational effort when compared to other typical analytical optimization methods.

Abstract ID: 42DCASS-046

Cooperative Localization for Unmanned Vehicles: Centralized and Distributed Approach

Anusna Chakraborty (chakraaa@mail.uc.edu)
University of Cincinnati
Rajnikant Sharma
University of Cincinnati


Multiple Unmanned Aerial Vehicle (UAV) missions have gained precedence over single vehicle applications in the past decade due to increased sensor coverage, efficiency and reliability. Most multi-UAV missions are heavily reliant on the availability of Global Positioning System (GPS) signals. Mission completion can get severely compromised in areas with intermittent or unreliable GPS signals where the Inertial Measurement Unit (IMU) can develop drift. In order to limit such dead reckoning, multiple vehicles can share their interoceptive sensor information (IMU, wheel encoders etc) and exteroceptive sensor data (camera, LIDAR etc) amongst each other to cooperatively estimate their individual states. This technique is referred to as Cooperative Localization in literature. Cooperative localization can be centralized or distributed. In the centralized case, a central node receives the data from all the other vehicles and estimates the states for each vehicle. On the other hand, in the distributed case, each vehicle estimates it's own states based on the information shared with it's immediate neighbors. Such localization techniques have widespread applications. In this presentation, we will highlight the major differences between centralized and distributed localization and discuss the usability of each of the approaches.

Abstract ID: 42DCASS-047

Implementation of Cooperative Timing Mission for a Moving Target using Cooperative Localization

Sohum Misra (misrasm@mail.uc.edu)
University of Cincinnati
Rajnikant Sharma
University of Cincinnati


Time synchronization of autonomous systems (robots and unmanned vehicles) to collaboratively accomplish a given task is crucial. One such case of time synchronization problem is for individual systems working in groups, to reach one or multiple targets at the same time. An improvement to this would be for the autonomous systems being able to reach (or hit) a moving target at the same time. In this presentation, we demonstrate the cooperative homing problem of smart munitions for a moving target. Further, cooperative localization of the munitions will improve the robustness of the system, helping them to navigate in GPS-restricted environment with limited sensing capabilities. The time synchronization problem is addressed using PN-guidance laws. The application to this problem can further be extended to drone surveillance and capture of moving objects such as hostile drones, where time synchronization plays a crucial factor. Simulation of the above problem is provided to demonstrate the effectiveness of the system.

Abstract ID: 42DCASS-048

Bridge Inspection using Simultaneous Localization and Mapping

Srijanee Biswas (biswasse@mail.uc.edu)
University of Cincinnati
Rajnikant Sharma
University of Cincinnati


Bridge safety inspection can involve several logistical challenges in terms of efficient visual assessment of wide variety of structure types in challenging locations. Also, it might pose safety risks for personnel and high cost implications in the form of under bridge inspection snoopers and lifts. Unmanned Aerial Vehicles (UAVs) are gaining prominence as a tool to achieve safer and cost-effective bridge inspection. Simultaneous Localization and Mapping (SLAM) is concerned with creating a map of an unknown environment and using the same to navigate a robot in this surrounding. This presentation deals with the implementation of this algorithm in a robot-centered frame using appropriate sensors and human-in-loop approach.

Abstract ID: 42DCASS-060

UAVs for Emergency Sample Collection in Hazardous and Inhospitable Environments

Lydia Smoot (smootlg@mail.uc.edu)
University of Cincinnati
Vince Dechellis
University of Cincinnati
Kelly Cohen
University of Cincinnati


In 2011 the Great East Japan earthquake off the Pacific coast of Tōhoku became the fourth most powerful earthquake in the world; yet, the real devastation came with the tsunami triggered by this natural disaster. Due to electrical loss and cooling system failure at the Fukushima Daiichi Nuclear Power Plant, hydrogen gas build up within the outer containment building caused level 7 meltdowns and multiple nuclear reactor explosions. These explosions released radioactive material into the air, soil, and sea inciting a large scale evacuation and over 1,000 deaths. The purpose of this project is to investigate the adaptation of Unmanned Ariel Vehicles (UAVs) for missions involving sample collection, such as contaminated soil, in hazardous or inhospitable environments resulting from natural and unnatural disasters. For a community already devastated by a major disaster, the implementation of a UAV in place of a human life could have incredible financial and safety benefits. This project aims to create an easy-to-operate UAV using low cost off-the-shelf components, simple system design, and a light weight frame. Through this research the team hopes to further prove the unlimited application opportunities for UAVs, and prevent the harm of human life in dangerous environments.

Abstract ID: 42DCASS-052

Implementing Path Planning and Obstacle Avoidance for AUVSI SUAS Competition

Nathaniel Richards (richarnt@mail.uc.edu)
University of Cincinnati
Nicklas O. Stockton, Kelly Cohen, Manish Kumar
University of Cincinnati


The University of Cincinnati Autonomous Vehicle Team, or UCAV, will be competing in the AUVSI SUAS international competition for the first time in 2017. The competition involves the design, fabrication, and demonstration of an unmanned aerial system. The vehicle must autonomously perform a set of tasks, such as waypoint navigation and search patterns, all while remaining within the flight boundaries and avoiding simulated obstacles. The algorithms used for these tasks must be computationally light, as much of the onboard computational resources will be dedicated to image processing and communication. First, a series of waypoints are generated to search areas of interest. Then, the obstacle avoidance algorithm generates intermediate waypoints to mitigate potential collisions with the simulated objects. Rather than discretized grid methods, such as A*, the path planning and obstacle avoidance algorithms used are based on simple geometry to resolve the path from the vehicle’s current position to the next waypoint. This research discusses the algorithm development, implementation, and simulation results in preparation for the competition.

Abstract ID: 42DCASS-061

Small Robotic Manipulator Integration with UAV

Jishu Medhi (medhijk@mail.uc.edu)
University of Cincinnati
Catharine McGhan
University of Cincinnati


The development of Unmanned Aerial Vehicles with manipulating capabilities is an emerging field of research study which has gained a lot of popularity in recent times due to the wide range of possibilities that it can offer in many fields. This presentation discusses research in-progress to design, build, and test an unmanned aerial vehicle integrated with a robotic manipulator arm having multiple degrees of freedom in simulation, and then flight tests of a physical system in a structured indoor test space. This integration helps the aerial vehicle to interact and manipulate its environment with the help of its robotic arm while remaining airborne in addition to carrying out surveillance operations. A framework of the model will be developed using the kinematic and dynamic equations of the system and simulation will be performed in order to check the stability of such a system. The proposed aerial manipulator can be used to carry out aerial grasping, including such tasks as: lifting and picking up objects on flight, the transportation of objects between locations, and operations in human-shared spaces where the risk of collision must be a factor in both manipulation and trajectory planning.

Abstract ID: 42DCASS-066

Key factors for drone design in implementation of 4G/LTE for control of small UAS

Jim Herner (herner.5@wright.edu)
Wright State University
Jason Gottweis, Josh Childers, Thomas Easton
Wright State University


4G/LTE imposes a wide variety of constraints on the development of small commercial UAS, requiring flight-control software capable of safely controlling the UAS while dealing with the unique challenges imposed by the 4G/LTE network infrastructure. Utilizing the same communication channel in order to support a First Person View (FPV) video feed for real-time video streaming further complicates network usage. In order to mitigate common small UAS safety hazards associated with degraded or lost communication between the UAS and the pilot, several flight-software features are investigated and tested. These include error handling associated with 4G/LTE handovers, dynamic data prioritization based on available network speeds, and automated backtracking to recover network connectivity.

Abstract ID: 42DCASS-089

Vision-based Indoor Navigation of an Autonomous and Interactive Unmanned Aerial System

Hongyun (Elliot) Lee (hylee101001@gmail.com)
The Ohio State University


This paper considers the development and implementation of a vision-based onboard flight management computer (FMC) and inertial navigation system (INS) that can estimate and control the attitude and relative position of the vehicle while interacting with an operator and managing simple flight tasks in an indoor environment. In this project, a hexacopter is developed which uses an onboard inertial measurement unit (IMU), a small-sized computer, a Microsoft Kinect sensor, and a miniature ground control station and the software. Since directional maneuver of the hexacopter can be achieved by changing the angular position of the vehicle, the project consists of inertial and vision sensor fusion. To accomplish stable control of a multicopter's orientation, a complementary filter is implemented on the IMU. Since the raw sensors do not provide an accurate angular position of the vehicle per se, the complementary filter is used to fuse the raw sensor data to provide low-noise and low-drift estimation of Euler angles. Autonomous indoor flight requires position sensors other than GPS, since indoor reception of GPS is poor and accuracy is approximately 2 to 3 meters at best. In this project, a computer vision algorithm is used to provide position estimation. The vision algorithm uses an onboard Microsoft Kinect and a computer to execute EmguCV and Kinect SDK libraries. Since Kinect can provide color and depth data, it allows for object detection and provides its real-time local coordinates which can be used for an autonomous indoor flight. The coordinates are used to control the position of the vehicle. Also, for human-robot interaction, input commands such as takeoff, landing, proceed, or retreat during the flight can be perceived using an artificial neural network that classifies human gestures with specific implications. The system is expected to maintain vehicle position within 1 meter, for better accuracy when compared to low-cost GPS receivers, and to control corresponding angular orientation for the position control. Also, the gesture classifier and the flight task decision algorithm are expected to work in a consistent manner.

Abstract ID: 42DCASS-093

Derived Angle of Attack Algorithm Characterization for sUAS

Matthew McCrink (mccrink.2@osu.edu)
The Ohio State University
James W. Gregory
The Ohio State University


Knowledge of an aircraft’s angle-of-attack (AoA) is a critical need for reducing loss-of-control incidents occurring in the general aviation fleet. Numerous methods exist for directly measuring AoA using specialized pressure sensors mounted to the aircraft. However, product certification and the costs associated with hardware installation and calibration is likely prohibitive to wide-scale adoption. Indirect AoA estimation techniques using inertial sensors and fusion algorithms are a novel way of providing similar information. Indirect estimation techniques can be used with data available from most attitude and heading reference system (AHRS) and typically require minimal external modifications or additional hardware. However, this represents a trade-off as the reduction in hardware needs is somewhat offset by the need for high-fidelity aerodynamic parameters. These parameters are generally developed during initial flight-testing and certification of the aircraft and may not be available for all airframes. Numerous methods for estimating the derived AoA have been presented in the open literature. Loosely speaking, indirect estimation methods fuse measured aircraft angular rates and accelerations with some a priori knowledge of the vehicle’s performance characteristics. In this work, we will present an advanced estimation method utilizing learning algorithms and training feedback to remove some of the requirements for high-fidelity aerodynamic data. Learning-based methods typically require training data sets and calibration spanning the operating envelope of the aircraft, but adapt well to changes in vehicle performance due to loading, modification, or damage. While these methods have proven successful in a simulated environment, rigorous quantitative comparison of indirect AoA estimation techniques on real flight vehicles is lacking.

Abstract ID: 42DCASS-123

System Design of an Autonomous Quadcopter for Indoor Exploration

Hans Guentert (guenteph@mail.uc.edu)
University of Cincinnati
Bryan Brown, Nick Stockton, Justin Ouwerkerk
University of Cincinnati
Manish Kumar and Kelly Cohen
University of Connecticut


This presentation covers details of a Self-Navigating UAS Indoor Surveillance System built and designed at the University of Cincinnati’s UAV Master Lab. The purpose of the vehicle is to fly itself through a GPS-denied environments, charting waypoints for self-guidance, and using the same data to generate a 2D map alongside video of the enclosed environment for later retrieval. It was further tested and demonstrated actual indoor scenarios such as exploring hallways and steam tunnels. The UAS is a completely custom built quadcopter designed to meet operational requirements of indoor flight. With quadrotors ability to negotiate tight spaces and autonomous functionality, it was sought as a competitive solution to ground robots. The major contribution of this team has been in the frame design, software and hardware integration of various kinds of sensors, on-board computing boards and communication devices. Specifically, the team contributed by developing its own software interfaces to enable data communication and processing, navigation and mapping algorithms that utilizes on-board sensors, control algorithms to stabilize and navigate the quadrotor to desired locations, and online target recognition algorithms. This presentation provides details of these contributions.

Other

Abstract ID: 42DCASS-004

Ground-Based Sense and Avoid for the Air Force Research Laboratory and State of Ohio

Arthur Huber (afhuber@bellsouth.net)
Air Force Research Laboratory


The Air Force Research Laboratory and state of Ohio have embarked on a cooperative effort to develop and field a Ground-Based Sense and Avoid (GBSAA) system at Springfield-Beckley Municipal Airport (KSGH) which will enable Beyond Line of Sight (BLOS) flight operations of small Unmanned Aerial Systems (sUAS). The system is scheduled for installation and checkout in 2QCY17 followed by petition to the Federal Aviation Administration (FAA) for a Certificate of Waiver or Approval (COA) to operate in the unrestricted airspace around KSGH. Once approved and in operation, AFRL researchers or its sponsored partners as well as state entities and commercial firms will be able to use the GBSAA system to conduct sUAS flights in support of R&D, operator training, or commercial activities. This presentation will give an overview of the OH-AFRL partnership, the capabilities and concept of operation of the GBSAA system, and status of fielding and FAA approval pursuits.

Abstract ID: 42DCASS-044

Satellite Propulsion Spectral Signature Detection and Analysis

Pamela Wheeler (Pamela.Wheeler@afit.edu)
Air Force Institute of Technology
Richard Cobb
Air Force Institute of Technology
Carl Hartsfield
Air Force Institute of Technology
Benjamin Prince
Air Force Research Laboratory


Awaiting Public Release Clearance.

Abstract ID: 42DCASS-084

Use of a Mobile Robotic Manipulator as a Testbed for Scalable Software Architecture Supporting Intelligent Operations

Divya Ravichandran (ravichda@mail.uc.edu)
University of Cincinnati
Catharine McGhan
University of Cincinnati


For many reasons, it would be useful to have a robot that can work beside us, helping us as an assistant -- for instance, using a manipulator arm to retrieve objects for us with just a single voice command (“bring me the phone”) without any further user intervention. Though this is an easily understandable request for a human, it is nontrivial for a robot to perform. In order to do this, a robot would need several advanced capabilities, such as the ability to be able to socially interact with the human they are helping (speech recognition, natural language processing), as well as the ability to autonomously manipulate or fetch objects and move around within unstructured and cluttered environments like a home or office (machine vision, learning, localization and mapping, trajectory planning, combined robotic-base and manipulator operations, etc.). All of the underlying techniques needed to allow a robot to do this currently exist today -- many of them readily-available in open-source Robot Operating System (ROS) packages -- but in general it is still difficult to combine these techniques together for a particular use case, let alone make any guarantees as to safety or performance. This presentation will discuss ongoing work being done to develop an easily-extendable, easily-modified, scalable open-source software framework intended for use by a large developer community, meant to allow users to reprogram or add different behaviors without issue, and with some initial basic functionality implemented (object retrieval and cleaning tasks) that demonstrates the system's capability to support intelligent real-time response.

Abstract ID: 42DCASS-097

HIFiRE 5b Launch Campaign and Flight

David Adamczak (david.adamczak@us.af.mil)
Air Force Research Laboratory
Roger Kimmel
Air Force Research Laboratory
Douglas Dolvin
Air Force Research Laboratory


The Hypersonic International Flight Research Experimentation (HIFiRE) program is a hypersonic flight test program executed by the Air Force Research Laboratory (AFRL) and Australian Defence Science and Technology Group (DSTG). HIFiRE flight five b flew on May 18, 2016 from the Woomera Test Range, Australia. Principle goals of this flight were to measure hypersonic boundary-layer transition on a three-dimensional body with a secondary goad of measuring fin aerothermal response for a ceramic matrix composite fin. The flight went as planned for this mission and was able to collect hypersonic boundary layer transition data on ascent up to Mach ~5.5 and on reentry at Mach ~7.5. The launch campaign was conducted from the Woomera Test Range located at Woomera, South Australia. by a multinational team from AFRL, DSTG, Deutsches Zentrum für Luft und Raumfart (DLR) – Mobile Rocket Base (MORABA), Germany, the Royal Australian Airforce – Air Warfare Center and the Royal Australian Navy – Ranges and Assessment unit. The payload and booster preparation steps leading up to launch took place over a two week period and was ready for launch with a few minor delays. From the multitude of sensors flown on the vehicle it was able to reconstruct the flight trajectory and vehicle dynamics to enable further interpretation of the scientific experiments flown. Approved for Public Release per case # 88ABW-2017-0129

Abstract ID: 42DCASS-142

Modeling a Dynamic Variable Cycle Engine in Simulink

Robert Buettner (buettner.10@wright.edu)
Wright State University
Rory Roberts
Wright State University
Mitch Wolff
Wright State University
Alireza Behbahani
Air Force Research Laboratory


Variable cycle turbine engines provide many changes in the way propulsion can affect the air vehicle. The efficiency improvement is the most obvious as the turbine cycle is changed from a high performance turbojet to a high efficiency turbofan engine over the mission. An adaptive turbine engine has the potential to provide thermal management capability with a heat exchanger placed in the third stream flow path. The amount of power that is pulled from the engine is also increasing by orders of magnitude. Changes in engine loading can take place at any time during the mission. To properly model these effects a transient engine model is required. The engine model must be computationally stable and efficient while being easily integrated with other subsystem models. A Simulink based model, which is stable and efficient, is being developed which will enable system integration and will provide an ideal environment for controls development.

Abstract ID: 42DCASS-145

Power-Thermal Emulation of Subcomponents for Full-Body Aircraft Simulation

Jack Chalker (jackchalker5@gmail.com)
Wright State University
Jack Brendlinger
Wright State University
Dominic Dierker
Wright State University
Mitch Wolff
Wright State University
Tommy Baudendistel
PC Krause & Associates


Modern day aircraft are made up of many subcomponents that ensure safe flight. Each of these subcomponents performs its own desired task, utilizing power from the system while producing some thermal load in the process. The focus of the power-thermal emulator is to simulate the side effects caused by one of these subcomponents found in an aircraft. The effects being simulated include retaining a certain power load from the power generating source and converting that power into a thermal load on the overall system. Many loading cases can be created for a single subsystem in an aircraft, depending on how many models were created and used. Creating these loads can be performed by using a controlled circuit, consisting of a microcontroller, solid state relay (SSR), and a resistor. The use of the SSR allows the microcontroller to vary the amount of power being dissipated and heat generated. Additional sensors are incorporated for monitoring the temperature gradients and flow characteristics of the subsystem. An external box is designed to house these components, matching the size, thermal mass, and flow path restraints set by the original subcomponent. Through testing, the Cp (coefficient of pressure) and thermal capacitance values for the design can be found and matched to those values of the original subsystem. Through the use of a power source, temperature variations of the circuit’s components are seen, confirming successful emulation. Similar methods are capable of being used to recreate and simulate loading effects for all components within the system, effectively creating a full-body aircraft simulation.