* Note that appearance on this list does not guarantee that the
abstract has been or will be accepted. All submitted abstracts
will be reviewed for suitability and technical content.
Acceptance will be confirmed via email with the submitting author.
Acoustics & Applied Aerodynamics
Abstract ID: 49DCASS-051
Internal Fluidic Injection for the Control of Supersonic Rectangular Jet Noise
Kaurab Gautam
University of Cincinnati
Aatresh Karnam
University of Cincinnati
Arshad Mohammad
University of Cincinnati
Mohammed Saleem
University of Cincinnati
Ephraim Gutmark
University of Cincinnati
The study focuses on the application of fluidic injection on supersonic jet noise reduction. A converging-diverging rectangular nozzle with an aspect ratio of 2:1 and a design Mach number of 1.5 was used. An embedded channel was made within the nozzle was to inject air into the core flow at a specific location in its diverging section. Injection was performed through rectangular ports with varying injection parameters like injection mass distribution and injection location at different operating conditions. Far-field acoustic results were utilized to assess the noise reduction in terms of Overall Sound Pressure Level (OASPL) and spectra. Schlieren imaging was employed to analyze the impact of fluidic injection on shock cell structure, while a Spatial Proper Orthogonal Decomposition (SPOD) analysis was carried out to identify instability modes of the jet. The results demonstrated that fluidic injection significantly mitigates screech and shock-associated noise, particularly in overexpanded conditions. At NPR 2.7, noise reduction up to 4dB was obtained in sideline and downstream observation angles. Schlieren results showed that fluidic injection had a significant effect on shock structure and jet shear layer that resulted in such noise reduction
Abstract ID: 49DCASS-097
Micro-Vortex Generator Exploration Using F414 Geometry
James Cramer
University of Cincinnati
Ephraim Gutmark
University of Cincinnati
Over the last 30 years jet noise and acoustics emissions from airplanes have moved to the forefront of engine research, fueled by VA claims from military operations and noise complaints from commercial aviation [1]. In recent years a new spin on an old technology, Micro-Vortex Generators (MVGs) have shown great promise using the F404 geometry, as an inaugural candidate [2] [3]. The results of the F404 MVGs had so much potential the work led to a full-scale F-18 test in just under 18 months from initial lab results. The question that remains open is how MVGs operate on a different geometry? Using the F414 geometry this series of experiments seeks to replicate the MVG results from the F404 on the F414, garnering more information on the performance of the MVGs and the various design sensitivities. Some of the variables for exploration are MVG location along the z-plane, angle from the flap to the MVG, angle between MVGs, number of MVGs, and MVG pattern. Experiments will take place at the Aeroacoustics Test Facility at the Gas Dynamics and Propulsion Laboratory located at the University of Cincinnati.
References
[1] Allan Aubert, and Richard McKinley, "Measurements of Jet Noise Aboard US Navy Aircraft Carriers," AIAA Centennial of Naval Aviation Forum @ Years of Achievement and Progress, American Institute of Aeronautics and Astronautics, 2011,
[2] Saleem, M., Karnam, A., Rodriguez, O., "Flow and acoustic fields investigation of noise reduction by micro vortex generators in supersonic nozzles," Physics of Fluids, Vol. 35, No. 10, 2023,
[3] Lopez Rodriguez, O., Saleem, M., Gutmark, E., "Implementation of vortex generators as a noise reduction system for a faceted supersonic nozzle," Aiaa Aviation 2021 Forum, 2021,
Abstract ID: 49DCASS-103
Experimental and Computational Analysis of Instability Suppression through Perforated Liners in Rotating Detonation Engines
Tyler Pritschau
University of Cincinnati
Jorge Betancourt
University of Cincinnati
Peter Glaubitz
University of Cincinnati
Ephraim Gutmark
University of Cincinnati
A perforated liner was installed on the outer wall of a Rotating Detonation Engine (RDE) on the backside of which is a single closed chamber. This liners geometry was designed based on empirical data in previous works. The liner was tested against a solid walled baseline in several RDE configurations including an annular RDE and a hollow RDE. In the annular experiments, the introduction of the liner suppressed some of the non-detonative instabilities but didn’t have a strong influence on the detonative mode achieved. In the hollow RDE the introduction of the liner often caused a change in detonative mode. In some cases, this resulted in complete suppression of the detonation wave, while in others the preferred operating mode transitioned from one wave to two co-rotating waves. Additional two-wave modes were also observed where previously no detonation formed. To further understand the unique influences the liner had on differing RDE configurations, acoustic eigenmode modelling was conducted on the combustion chamber alone, the combustion chamber and the air plenum together, and the full RDE including the combustion chamber and both the air and fuel plenum. Simulations were computed in COMSOL Multiphysics using average exhaust gas temperatures from computational literature to set the chamber conditions.
Abstract ID: 49DCASS-104
Experimental Investigation of a Skin-Actuated-Camber Morphing Wing Design
Julian Pabon
University of Dayton
Xinyu Gao
University of Dayton
Jielong Cai
University of Dayton
Sidaard Gunasekaran
University of Dayton
The aerodynamic performance of a novel Fishbone Skin-Actuated-Camber (SAC) morphing wing design, which actuates its skin to change its effective camber, was studied both experimentally and numerically. Force-based experiments were conducted at the University of Dayton Low Speed Wind Tunnel (UD-LSWT) to compare the performance of four morphing wing designs with different hinge locations, two ideal trailing edge flap wings, and one conventional trailing edge flap wing. All test articles have an Eppler 479 airfoil, an effective aspect ratio of four, and were tested within an angle of attack range of -15° and 15 °. The novel design achieved effective camber change without any buckling, maintaining comparable aerodynamic performance to ideal flap wings at a Reynolds number of 270,000. At a Reynolds number of 400,000, the morphing shows a lower drag than the ideal flap wing. Simulations from FlightStream®, a numerical solver correlated well with experimental lift data, with the morphing wing's pressure contours indicating reduced flow separation and gradual pressure change on the upper surface when deflected.
Abstract ID: 49DCASS-117
On the Linear Superposition of Wing and Propeller Performance in a Wing Embedded Propeller System
Jielong Cai
University of Dayton
Emma Schutter
University of Dayton
Sidaard Gunasekaran
University of Dayton
We continue our study on the aerodynamic performance of a Wing-Embedded Propeller (WEP) system by considering the effect of propeller installation location along the span and its overall size on the entire system. Six semi-span AR 2 wings with different spanwise opening locations and opening diameter-to-wing-chord ratios were investigated numerically using FlightStream® and experimentally in the University of Dayton Low-Speed Wind Tunnel without the installation of the propeller. The propeller is then installed and tested experimentally. Without the propeller, the wing with the opening closer to the wingtip experiences the least detriment to overall aerodynamic performance. Integrating the propeller into the WEP system and positioning it at the wing's mid-span, results in better overall performance. Increasing the propeller size has a more pronounced effect on the WEP system, increasing both lift and forward thrust generation. The linear superposition method is suitable for estimating the WEP system lift performance at a higher propeller incidence angle. However, it significantly underestimates the drag produced by the WEP system. This trend is consistent across all wing locations and propeller sizes tested in this study.
Aircraft and UAS Design & Applications
Abstract ID: 49DCASS-088
Optimal Trajectory Solutions for Unmanned Pursuer/Evader Offensive Counterair Including Engagement Zone
Alexander Hansen
Air Force Institute of Technology
Michael Zollars
Air Force Institute of Technology
Isaac Weintraub
Air Force Research Laboratory
Alexander Von Moll
Air Force Research Laboratory
Awaiting public release.
Abstract ID: 49DCASS-093
MDAO for Airframe-Propulsion-PTMS-Integration – Engine Design Case Study
Joshua Lupi
Air Force Research Laboratory
Timothy E. Wontor
University of Dayton Research Institute
Joshua D. Deaton
Air Force Research Laboratory
Nathan A. Wukie
Air Force Research Laboratory
The design of military aerospace vehicles is becoming increasingly integrated and power-dense, requiring a more holistic approach to component technology development. The traditionally disconnected disciplinary teams that are characteristic of aerospace design are no longer sufficient for addressing the complex interactions between different components and disciplines that combine to determine system-level performance. As a result, the AFRL Aerospace Systems Directorate has established a cross-division program intended to better capture the system-interactions between conceptual air vehicle design, propulsion, and PTMS subsystems and their influence on mission utility. An important element of this program is establishing an integrated multi-fidelity multidisciplinary design analysis and optimization (MDAO) environment for airframe-propulsion-PTMS-integration (APPI). This work focuses on the initial MDAO integration and evaluation of two commercial MDAO frameworks, Altair HyperStudy and OmniQuest/Altair Iliad and will emphasize the integration of propulsion system design models including turbine engine cycle and low fidelity part design. The results of MDAO studies demonstrate the impact of engine-only design and multidisciplinary engine and conceptual air vehicle design on mission performance.
Abstract ID: 49DCASS-098
Flight Test Validation of Tandem Propeller Performance with Vertical and Horizontal Offset
Michael Foster
University of Dayton
Jessica DeMoor
University of Dayton
Jielong Cai
University of Dayton
Sidaard Gunasekaran
University of Dayton
Tandem propellers in forward flight experience an increase in power consumption when compared to the combined output of two standalone propellers. The increment in power is a function of horizontal and vertical displacement between the propellers (including overlap), the advance ratio based on the front rotor, and the inclination angle of the rotors. This functional relationship was quantified in our previous study through experimental investigations in the University of Dayton Low Speed Wind Tunnel using two KDE propellers. All tests were conducted under trim conditions, where the pitching moment of the two propellers was balanced by increasing the RPM of the rear rotor. To validate some of the functional dependencies identified from the wind tunnel investigations, a custom quad-rotor platform was designed and fabricated to conduct a series of flight tests with various propeller configurations that replicate the parameter space explored in the earlier experimental campaign. The quad-rotor platform will utilize an 8-inch propeller to assess the flight performance at three different horizontal and vertical distances between the propellers. For each test-flight, global positioning data, motor rpm, and motor power consumption will be recorded and compared against each propeller configuration. Comparisons between the flight test data and the wind tunnel experiment results will be made.
Abstract ID: 49DCASS-099
Deep Reinforcement Learning based pursuit of a ground target in a grid with local, partial and erroneous information
Srikanth Elkoori Ghantala Karnam
University of Cincinnati
Rajnikant Sharma
University of Cincinnati
Autonomous navigation has been on the rise the past few years due to its versatile capabilities like delivery, search, tracking, etc. The pursuit-evasion differential game from game theory can be used to model the search and capture of a target moving through an environment. This study investigates an optimal strategy for capturing a randomly moving ground target in a grid network. When the evader takes no evasive motion, then it is referred to as a target. The network is embedded with unattended ground sensors (UGS) to record the target’s timestamped information. A high-flying UAV is chosen as the pursuer to perform the search and capture task. When the pursuer visits an UGS, the timestamped information of the target, if available, is instantaneously transmitted to it. Hence, the pursuer only gets information of the target locally through the sensors it visits. We propose a deep reinforcement learning based solution to achieve capture of the target before it escapes. The pursuer, the learning agent, is trained to make its own decision based on the reward policy defined on the pursuit-evasion game with partial information. The purser learns the optimal strategy through both exploration and exploitation in several iterations of the game. We show 100\% capture rate for a target moving in a random path with a constant speed in presence of error-free sensors and a capture rate of 97\% in presence of a sensor with a 10\% change of recording erroneous information.
Abstract ID: 49DCASS-127
MDAO for Airframe-Propulsion-PTMS-Integration – Air Vehicle Design Case Study
Timothy Wontor
University of Dayton Research Institute
Joshua M. Lupi
Air Force Research Laboratory
Joshua D. Deaton
Air Force Research Laboratory
Nathan A. Wukie
Air Force Research Laboratory
Within aircraft design, siloing of efforts for an air vehicle’s airframe, engine/propulsion, and PTMS systems has proven to be a major hinderance to identifying the system-level optimal performance of the final vehicle. To address this, a collaborative, cross-division program within AFRL Aerospace Systems Directorate has been initiated to create a capability for integrated air vehicle, propulsion, and PTMS subsystem development. A key component of this capability is multidisciplinary design analysis and optimization (MDAO), which enables the coupled analysis and design of the disparate engineering disciplines. This work demonstrates the initial integration of models for airframe design components (including aerodynamics, mass properties, stability and control, mission analysis) with propulsion models (engine cycle, low fidelity turbine part design) into an integrated MDAO framework. Two commercial MDAO frameworks, Altair Hyperstudy and OmniQuest Iliad, are explored to accomplish the integration. This talk will emphasize the airframe and conceptual air vehicle design model integration and results of MDAO studies comparing isolated conceptual air vehicle design with simultaneous air vehicle and propulsion engine cycle design.
Abstract ID: 49DCASS-128
Developing RF Ranging-Based Multi-Rotor Test-Bed for Cooperative Localization
Jayanth Ammapalli
University of Cincinnati
Rohith Boyinine
University of Cincinnati
Rajnikant Sharma
University of Cincinnati
The primary objective of this paper is to provide an in-depth discussion into cooperative localization of Unmanned Aerial Vehicles (UAVs) while also presenting the development of a generalized hardware test-bed to support multi-UAV flight experiments in outdoor environments. This test-bed utilizes inter-vehicle cooperative measurements (Radio Frequency based range measurements) to enhance the precision and efficiency of multi-UAV localization and navigation, specially in areas where Global Navigation Satellite Systems (GNSS) may not be reliable or available. The paper describes the critical elements of cooperative localization algorithms currently existing in literature and provides necessary modifications (like addition of bias states) to increase the accuracy on hardware implementation. The paper provides the necessary details for these changes and presents initial results to demonstrate the capabilities of the test bed created that can assist in the rapid development and transition of multi-vehicle simulations onto hardware platforms.
Applications & Facilities
Abstract ID: 49DCASS-006
Draco Ventus Clean Air Hypersonic Conditions Test Tunnel
Joseph Herdy
CFD Research Corporation
Garry Freeman
Army Research Office
Ben Kerstiens
Army Research Office
Jonathan Coleman
Army Research Office
Doug Engle
None
Ground testing to create hypersonic flight conditions is of significant interest to the Department of Defense and the Army. The U.S. Army’s Space and Missile Defense Command (SMDC) Tech Center desires to continue Research and Development (R&D) on ground test infrastructure to generate the data necessary to execute low-risk Test and Evaluation (T&E) programs and increase the Technology Readiness Levels (TRLs), that can provide ground test High Speed / Hypersonic Mission (HS/HM) simulations. This R&DT&E effort, named Draco Ventus, will develop physical and virtual prototypes to support life cycle production decisions for advanced offensive and defensive weapons solutions. The goal is to develop and deliver the specific ground test infrastructure TRL improvements, and supporting analysis, test and evaluation technologies.
This R&DT&E Science and Technology program, commissioned in August of 2023, is focused on finding solutions to three requirements in order to provide the greatest impact on enhancing existing materials and tunnel Research, Design, Test and Evaluation (R&DT&E) capabilities. These requirements include 1) abilities to supply clean air at hypersonic flight enthalpies with real-time changes corresponding to mission scenarios, 2) continuously vary Mach number within a single test at appropriate scales, and 3) execute test processes over mission length durations. While daunting, these requirements take advantage of a new approach to create clean, non-vitiated hot air. Significant effort on this program has yielded the promise to completely decompose nitrous oxide into pure oxygen and nitrogen while taking advantage of the generated enthalpy and simultaneous matching temperatures from the mixing of nitrogen to create the correct air species to meet free stream similitude conditions.
Creating the conditions that are capable of the dynamic temperatures and pressures associated with HS/HM will yield a paradigm shift in R&DT&E because it will push the capabilities and reduce limitations in simulating the most important HS/HM flow conditions. Goals are to create test attributes including: 1) Mach enthalpy (via temperatures on the order of 3000°F and higher to meet Mach 8 or greater); 2) altitude (“sea-level” to 130Kft altitudes); 3) alignment with real-time mission scenarios (i.e. provide condition transition time to realistically mirror flight engagements); and 4) permit run times of greater than 1000 seconds.
SMDC is interested in defining practical and actionable products and solutions, reducing program risk, and guiding the technology maturation that will leverage previous, successful efforts. The new Draco Ventus initiative will first see demonstration environments with actionable test customer usefulness. There will be parallel and informing activities for looking at revolutionary test HS/HM infrastructure. Early test configurations possibilities are intended to support propulsion systems, materials testing, and aerodynamics in relation to long-range precision-strike missions. Specifically, Draco Ventus will be configured to test an air-breathing (ramjet) propulsion system, as well as conduct hand-off and retrofit efforts necessary to incorporate the air-breathing technology into current and future weapon systems. Other HS/HM programs are also slated for benefitting from the generated R&DT&E infrastructure environment in the near and far term futures.
Abstract ID: 49DCASS-074
Human Aware Navigation for Mobile Robots in Airport Environments: A Framework Integrating Enhanced Potential Field and Fuzzy Inference System
Shurendher Kumar Sampathkumar
University of Cincinnati
Daegyun Choi
University of Cincinnati
Donghoon Kim
University of Cincinnati
The deployment of Autonomous Mobile Robots (AMRs) is transitioning from traditional spaces, such as factory floors and warehouses, to more public environments like airports and shopping malls. This shift reflects a recent trend in the integration of robotics and automation technologies into everyday public spaces due to the demand for efficiency and automating repetitive tasks. Unlike their earlier deployment in controlled industrial settings, this change exposes AMRs more to the highly dynamic, human-rich ambiance. In such environments, the primary focus of the AMRs should be navigation, taking into account human safety and comfort in a socially acceptable manner. This work discusses a Human Aware Navigation (HAN) framework for AMRs deployed in airport environments. The framework utilizes the Enhanced Potential Field (EPF) method, known for its collision avoidance capabilities without encountering local minima and obstacles near the goal. The coefficients of the EPF heavily influence the motion of AMRs. To strictly adhere to human factors associated with safety and comfort, it is necessary to actively adapt these coefficients based on information about humans, AMRs, and obstacles. Thus, a Fuzzy Inference System (FIS) is employed to assist the EPF in the navigation of AMRs. FIS is chosen for its robustness to uncertainty and scalability, while a Genetic Algorithm (GA) is incorporated to optimize the internal parameters of the FISs. The use of GA addresses the reduction of the AMRs’ path length while maintaining alignment with human factors. The training scenarios are carefully selected to encounter local minima and obstacles near the goal, considering the human-rich environment. Finally, the effectiveness of the proposed approach is validated and analyzed via Monte-Carlo simulations across various airport-based scenarios like transportation and cleaning.
Combustion & Fuels
Abstract ID: 49DCASS-066
Combustion Dynamics within Type-II Intermittent States in a Multi-Nozzle Lean Direct Injection Combustor
Yuvi Nanda
University of Cincinnati
Aditya Saurabh
None
Lipika Kabiraj
None
Ephraim Gutmark
University of Cincinnati
Intermittent combustion oscillations are observed in a fuel-staged Multi Nozzle Lean
Direct Injection combustor. The unstable periodic and aperiodic dynamics of two type-II
intermittency states have been identified and investigated to establish a trend and
quantify the variations existing between the two states. Through phase space
reconstruction of scalar pressure measurements, recurrence plots and recurrence
quantification is implemented for analyzing the behavior of the combustion instability.
The three independently controlled fuel stages, i.e., the pilot, intermediate, and outer
stage of the combustor continuously interact with each other and contribute towards
the instability. Time-resolved OH* chemiluminescence images and spectral proper
orthogonal decomposition are used to study the changes in the flame behavior and
interaction between the pilot, intermediate, and outer stages corresponding to the two
type-II intermittent states.
Abstract ID: 49DCASS-077
Investigating the Use of Low-Voltage Nanosecond-Pulsed Discharges for Cavity Ignition in Supersonic Flow
Katherine Opacich
University of Dayton
Joshua S. Heyne
Other - please contact webmaster
Erik L. Braun
National Research Council
Timothy M. Ombrello
Air Force Research Laboratory
Nanosecond-pulsed high-frequency discharges (NPHFDs) are an ignition technology of interest due to their efficient production of active radicals and excited species with low overall energy input. However, they produce significant electromagnetic interference (EMI) due to their high peaks in voltage and current that can be damaging to other electronic components in the surrounding region. Therefore, methods to reduce the high voltages needed for NPHFDs while not sacrificing ignition performance are of interest. The present work experimentally investigated the ability of a burst of low-voltage nanosecond discharges to ignite a cavity-based flameholder. Results were compared against the ignition performance of single, high-voltage nanosecond discharges. All experiments were conducted within a Mach-2 flow with stagnation temperature and pressure of 589 K and 483 kPa, respectively. The cavity was directly fueled with ethylene to test ignition at two fueling rates of 50 and 95 slpm. Results showed that low-voltage bursts of nanosecond pulses produced low- and high-current discharges. The low-current pulses were found to pre-condition the discharge region and establish breakdown, whereas the high-current arcing discharges governed ignition. Pulse repetition frequencies of 300 kHz with peak voltages of 1.6 kV successfully ignited the cavity with up to 89% lower voltage and at least 57% lower energy compared to single pulse cases while not impacting the time to ignition. Ultimately, utilizing a burst of low-voltage nanosecond pulses is a promising method of reducing the EMI related to NPHFDs without sacrificing ignition performance.
Abstract ID: 49DCASS-109
Performance of Glow Plug Ignition Systems for Gas Turbine Engines in Quiescent Atmospheric Conditions
Bryce Ullman
Wright State University
Brent Rankin
Air Force Research Laboratory
Mitch Wolff
Wright State University
Ignition systems play a critical role in starting combustion within gas turbine engines, especially in challenging environmental conditions. The research presented investigates the performance of glow plug ignition systems across quiescent atmospheric conditions in a gas turbine engine, aiming to enhance their effectiveness and reliability in various operational environments. The study will entail a standalone glow plug ignition system that has metered fuel flow and power. The experimental setup will use a Commercial-Off-The-Shelf (COTS) ignition system that will be used to characterize the ignition parameters at quiescent atmospheric conditions. The ignition system will have a constant current D.C. power supply and a volumetric fuel flow rate that is regulated by pressurized fuel using a needle valve and pressure gauge. A camera will capture videos and images to provide data for the flame length. This study will measure the inputs to the system such as the volumetric fuel flow rate, current, and voltage to the glow plug. The study findings will be whether the specified conditions ignite into a steady flame as well as the flame length. The results will provide further insights and direction into future endeavors that will test sub-atmospheric conditions. By addressing the critical issue of glow plug ignition performance under quiescent atmospheric conditions, this study contributes to the advancement of gas turbine engine technology, facilitating more reliable and efficient operation.
Abstract ID: 49DCASS-111
Comparison of Thermoacoustic Instability Behavior in an LDI combustor When Using Propane and Methane Fuel
Shyam Muralidharan
University of Cincinnati
Yuvi Nanda
University of Cincinnati
Ephraim Gutmark
University of Cincinnati
This paper compares the cross-frequency interaction that occurs during thermoacoustic instability in a lean direct injection combustor. Methane and Propane fuels are used to study the effect of fuel used. The setup had an end plate at the exit of the combustor with a hole the same diameter as the nozzle. The combustor was operated at atmospheric back pressure conditions and inlet air was preheated to 465K. The synchronized sound pressure (P) and OH* chemiluminescence images were obtained at 25kHz. The oscillation frequencies are discerned as pressure fluctuations at F0 and corresponding OH* fluctuations at F1=2F0, signaling the presence of nonlinear cross-frequency interactions. Pressure and OH* signals filtered at the dominant frequency were used to study the interactions. Analytical techniques like FFT, spectrogram and synchronization theory were used to study the dynamic behavior of the combustor across varying equivalence ratios (1.15, 0.99, and 0.88). The quantification of instability strength was calculated using the synchronization index providing an alternative to Rayleigh’s index.
Abstract ID: 49DCASS-122
PLIF Imaging of Combustion Instabilities in a TARS Combustor
Grace Fischer
University of Cincinnati
Ephraim J. Gutmark
University of Cincinnati
Flow experiments are conducted in a swirl-stabilized combustor with a lean-direct injection fuel nozzle with several configurations of a Triple Annular Research Swirler (TARS). Planar Laser Induced Fluorescence (PLIF) imaging using a high-speed laser to visualize the hydroxyl radical (OH) shows where the combustion reaction is occurring as it is produced at an intermediate stage in the combustion process and can map the flame structure and show combustion completeness. TARS consists of three air passages that each have their own swirler that can be changed independent of the others. This allows differing configurations of TARS by changing the swirlers either to different swirling angles or different rotating directions in order to form differing swirling flow fields. By having two separate fuel circuits (pilot and main) and the ability to interchange inlet swirl number TARS is designed for testing combustion instabilities with different types of flows and injection patterns. Additionally, lean-direct injection has the potential to have equivalent nitrous oxide (NOx) emission reduction and exit temperatures as lean premixed combustion without the drawbacks like flashback and autoignition that are seen with premised combustors. The TARS nozzle was run with six different cases captured with PLIF images to compare stable flow to unstable flow in lean combustion. This was done to investigate the pilot fuel and main fuel split and its effect on combustor flame stabilization and behavior for methane fuel. POD performed on these findings can demonstrate how combustion dynamics during combustion instability are affected by the pilot and main fuel flow rates. Combustion instabilities affect both power and propulsive technologies and so it is important to study the characteristics and dynamics of it.
Computational Fluid Dynamics
Abstract ID: 49DCASS-001
A Multi-fidelity Gradient-based Constrained Optimization Method Applied to Benchmark Problems
Markus Rumpfkeil
University of Dayton
Phil Beran
Air Force Research Laboratory
The traditional design process relies upon low-fidelity models for expedience and resource savings. However, the reduced accuracy and reliability of low-fidelity tools often lead to the discovery of design defects or inadequacies late in the design process. These deficiencies result either in costly changes or the acceptance of a configurations that do not meet expectations. Multi-fidelity methods attempt to blend the increased accuracy and reliability of high-fidelity models with the reduced cost of low-fidelity models.
In this presentation, a gradient-based multi-fidelity constrained optimization approach is presented and exercised on two analytical benchmark problems, as well as a constrained drag minimization problem for the RAE-2822 airfoil. The results show promise in overall computational savings compared to using high-fidelity information alone.
Abstract ID: 49DCASS-007
Simulations of Nose Curvature Effects on Hypersonic Boundary Layer Transition
Tim Leger
Ohio Aerospace Institute
Matthew Tufts
Air Force Research Laboratory
Nicholas Bisek
Air Force Research Laboratory
A set of high fidelity simulations were performed using Overflow in support of an experimental test series. The test series was conducted in the AFRL Mach 6 Ludwieg Tube facility, aimed at better understanding the effects of leading-edge curvature on hypersonic boundary layer transition. The test article geometry was derived from the AFOSR BoLT flight experiment, at half scale, with flattened test surfaces and interchangeable nose tips of different curvature. Grids for the simulations were initially generated using Chimera Grid Tools, and then iteratively refined and shock aligned to the base flow field solution. To stimulate boundary layer transition, a novel boundary condition was employed to introduce white noise outside and along the bow shock. The simulations show the same general trends of boundary layer transition and surface heating as observed in the experiments, but also provide more detail of the involved flow physics. Distribution Statement A: Approved for Public Release; Distribution is Unlimited. PA# AFRL-2023-6458
Abstract ID: 49DCASS-009
Turbulence Anisotropy and Mean Flow Correlations in the Periodic Hills Case
James Wnek
Wright State University
Mitch Wolff
Wright State University
Christopher Schrock
Air Force Research Laboratory
The accurate modeling of turbulence is critical to CFD applications, but standard eddy viscosity models are known to be deficient in predicting separated flows. Data-driven corrections or predictions of the turbulence anisotropy provide a potential way to improve the accuracy of RANS simulations, and feature selection is an important component of these efforts. In this work, the turbulence anisotropy tensor was obtained for the parametric periodic hills case from DNS data and corresponding k-epsilon RANS simulations. A barycentric mapping was used to visualize the anisotropy tensor as a convex combination of one-component, two-component, and isotropic states of turbulence. A random forest model was then used to correlate mean flow features with the barycentric coefficients. The relative importance of the input features was evaluated using the Gini importance and permutation importance measures. Additional analysis was conducted to distinguish features whose importance is obfuscated by strong correlation with other features. This analysis indicated very high importance of the wall distance Reynolds number and scalars formed from the velocity gradients, consistent with traditional turbulence modeling assumptions. The results of this research can help inform the choice of input features for data-driven turbulence modeling.
Distribution Statement A: Approved for Public Release; Distribution is Unlimited. PA# AFRL-2024-0096.
Abstract ID: 49DCASS-037
An application of the Lattice-Boltzmann method for ice particle melting
Carlos Eduardo Americo
University of Kentucky
Savio J. Poovathingal
University of Kentucky
The presence of ice in the atmosphere is common within a certain altitude range. High-velocity vehicles can fly through this ice-rich environment and have their surface damaged and the flowfield near its surface perturbated by the presence of solid particle aggregation. When a vehicle's flight creates either weak or strong shockwaves, right downstream the shockwave, the flowfield temperature increases significantly, and the ice particles suspended in the air that goes through the shock wave receive this heat transfer, now susceptible to heat conduction and phase change. Understanding how the ice particle melts in this environment is relevant to predicting the flowfield more accurately and understanding surface damage due to the impingement of ice particles, its fragments, and any combination of phases due to the particle's temperature increase.
The main topic of this work is to comprehend how the ice particle behaves under those circumstances. This process is achieved by numerical investigations using in-house software under development. This computational code uses the Lattice-Boltzmann method for heat conduction and phase change with BGK approximations, applying the Chapman-Enkog expansion. The main particle distribution function is numerically implemented as a one-dimensional lattice domain (D1Q2).
The boundary conditions are implemented with prescribed temperatures. The initial domain is created as ice, and one of the boundaries is water. The phase change front will propagate through the domain, and an intermediate mushy region will develop between ice and water. The physical properties of all three areas are not equal, and a fixed space in the domain will experience different temperature distributions based on the current phase change front and the status of ice, water, or mushy.
The temperature profile of the ice particles is investigated based on the simulations that run under small time steps (e-11 seconds) and a small total time (e-3 seconds) due to the nature of the environment under analysis. The temperature profiles are also compared against another in-house program based on the finite volume method. The temperature distribution for heat conduction when the domain is either water or ice has a good agreement between both codes.
Abstract ID: 49DCASS-039
Gas-kinetic Simulations of Turbulent flow over a Porous TPS Surface
Ahilan Appar
University of Kentucky
Savio Poovathingal
University of Kentucky
The present work aims to investigate turbulent flow over a Thermal Protection System (TPS) porous surface, introducing a scenario in which gas particles may permeate the porous medium. This complex configuration gives rise to a coexistence of continuum and rarefied flow regimes, making gas-kinetic molecular methods such as Direct Simulation Monte Carlo (DSMC) an ideal technique for the study. Our initial focus involves assessing the efficacy of the DSMC technique in capturing the coherent structures that characterize turbulent flow. The potential of gas-kinetic methods to support the development, propagation, decay, and regeneration of essential long-range correlations for the establishment of coherent structures crucial to sustaining turbulence is an ongoing area of research. The primary focus of our research is to understand the ability of gas-kinetic methods like DSMC, with molecular chaos assumptions, to capture sustained turbulence. This method indirectly determines the molecular velocity distribution by monitoring a significant number of computational molecules as they undergo motion and collisions, as opposed to solving the Boltzmann Equation (BE) directly to obtain the molecular velocity distribution. While the precision of DSMC in simulating flow instabilities and the decay of homogeneous isotropic turbulence has been demonstrated, more understanding is needed to ascertain the capacity of the DSMC method to simulate turbulence that is sustained over time. Sustained turbulence requires a continuous renewal of long-range correlations and the associated coherent structures, contrasting with decaying turbulence, which does not demand such regeneration. Consequently, our investigation addresses the pivotal inquiry of whether gas-kinetic methods like DSMC, which enforce molecular chaos exclusively during collisions, allow for the presence of long-range correlations associated with coherent structures essential for sustaining turbulent flow. To investigate this matter, we simulate turbulent flow over a curved wall and meticulously compare our results against experimental data and continuum simulations. This study aims to contribute valuable insights into the potential of gas-kinetic methods to accurately represent and sustain turbulence in challenging flow scenarios involving porous surfaces.
Abstract ID: 49DCASS-043
Developing a Benchmarked Pathway to Quantify Patient-Specific Boundary Conditions for Hemodynamic Modeling in Cerebral Aneurysms
Hang Yi
Wright State University
Zifeng Yang
Wright State University
Luke Bramlage
Miami Valley Hospital
Bryan Ludwig
Miami Valley Hospital
Boundary condition (BC) is one pivotal factor affecting the accuracy of hemodynamic predictions of cerebral aneurysms (CAs) using hemodynamic modeling. However, a benchmarked pathway to secure accurate BCs for hemodynamic modeling has not been developed. Previous investigations used generalized rather than patient-specific BCs for hemodynamic modeling in CAs, which can induce significant errors in the pathophysiology diagnosis and then rarely be employed in the clinic. To develop a benchmarked pathway to secure the precise BC for hemodynamic modeling in CAs and quantify the hemodynamic differences under various BC strategies, this study conducted a comprehensive investigation based on transcranial Doppler (TCD) ultrasonography measurements and the discrete Fourier transform (DFT) simulation. Specifically, two representative patient-specific CA models were rebuilt and their blood flow information in the internal carotid artery was tested by TCD techniques and DFT modeling. Then, systematic numerical investigations were conducted to explore the appropriate number of samples (N) for DFT modeling to secure the prime BC by comparing hemodynamic parameters using an in-vitro validated CFD model. Subsequently, a comprehensive comparison of hemodynamic characteristics under patient-specific BCs and a generalized BC was conducted to reinforce the understanding that a patient-specific BC is pivotal for accurate hemodynamic risk evaluations on CA pathophysiology. The results showed that N? 16 for the DFT model is a decent choice to secure the prime BC profile to calculate time-averaged hemodynamic parameters. At the same time, more data points such as N?36 can ensure the accuracy of instantaneous hemodynamic predictions. Additionally, results revealed the generalized BC could overestimate or underestimate the hemodynamic risks on CAs significantly; thus, patient-specific BCs are highly recommended for hemodynamic modeling for CA risk evaluation.
Abstract ID: 49DCASS-049
Investigation of Spallation in TPS Materials
Kate Rhoads
University of Kentucky
Alexandre Martin
University of Kentucky
Kristen Price
University of Kentucky
Ablative thermal protection systems (TPS) are used to shield internal components of space vehicles from intense heating during atmospheric descent. Mass loss of ablators due to the ejection of particles from the bulk material is known as spallation. During this process, particles of the TPS break off and join the flow, resulting in increased surface recession. Recent work has been completed on the development of a probability density function for particle ejection location based on arc-jet tests conducted at the NASA Aerodynamic Heating Facility (AHF). Previously, various tests were conducted at NASA’s Hypersonic Materials Environmental Test System (HyMETS) facility. High-speed images, particle tracking velocimetry analysis, facility measurements, and the Kentucky Aerothermodynamics and Thermal-response System (KATS) have been used to develop models of the HyMETS samples. This data has been analyzed for both comparison to AHF results and the ultimate goal of developing an empirical model to more accurately predict the spallation phenomenon.
Abstract ID: 49DCASS-072
Automatic Geometry Generation, Meshing, and Aerodynamic Modeling
Christopher Humphrey
University of Cincinnati
Jose Camberos
Air Force Institute of Technology
With the resurgence of interest in hypersonics, a need to fully utilize simulation tools has
arisen. Assessing the accuracy of these software tools and techniques against accepted results
requires careful consideration. NASA’s low fidelity tool Configuration Based Aerodynamics
(CBAero) provides accurate and efficient results at hypersonic speeds. Benefits include the
increased ability to explore a (design) trade space and reduce the time needed to test new
configurations. Executing the process for design exploration with simulations requires the
automation of a parametric modeling tool and a meshing tool. The research performed included
two parts: First, create and link the geometry and the meshing. Second, input and simulate
the aerodynamics with CBAero. The first part used Engineering Sketch Pad (ESP) to generate
the geometry of the X-43 hypersonic vehicle. Part two utilized a python script using GMSH to
generate a two dimensional surface mesh, creating a link between ESP and the meshing tool.
The python script controlled the automatic unstructured meshing and has several options to
define the fidelity of the mesh. This mesh then goes into CBAero to simulate a variety of flight
conditions. The final result generates a streamlined process from geometry generation and
meshing to simulation data in a simple and easy to use package.
Abstract ID: 49DCASS-084
Simulations of Mixing Interactions between Fuel Injection and Cavity Flameholding in Supersonic Crossflow
Benjamin Millard
University of Cincinnati
Daniel Cuppoletti
University of Cincinnati
This research presents simulations of gaseous fuel injection mixing in a cavity flameholder of a high-speed combustor when fueled from an upstream injector. We will be exploring this problem as the intersection of two fundamental fluid dynamics problems, a jet in supersonic cross flow and supersonic flow over a cavity. This was done by simulating a single ethylene injector in Mach 2 crossflow using 3D RANS. A realistic cavity flameholding geometry was then introduced downstream of the injector and the resultant interaction of the injector plume and the cavity recirculation zone was captured. The injector plume's evolution is compared to the baseline no-cavity behavior with emphasis placed on fuel jet trajectory and mass exchange within the cavity recirculation zone. Three different momentum flux conditions were simulated to capture how different injector momentum changes the interaction with the cavity, with these simulations being able to be validated against existing experimental data in literature.
Abstract ID: 49DCASS-089
Simulating Fluid Flows Using Quantum Computing
Marek Brodke
University of Cincinnati
Prashant Khare
University of Cincinnati
Recent advances in quantum computing hardware and its increased availability has made it possible for researchers to explore its potential. While this emerging area has been investigated by researchers in several different fields of inquiry, there is a lack of literature demonstrating its viability in simulating classical fluid flows. In this research effort, we build on previous work that showed the viability of solving the Navier-Stokes equations using classical-quantum algorithms. Specifically, we implemented a quantum amplitude amplification and estimation circuit and successfully demonstrated that it can be combined with classical algorithms to simulate fluid dynamic flows. This hybrid quantum-classical algorithm is then applied to investigate quasi-1D flow through a converging-diverging nozzle. Our results are in excellent agreement with the theoretical calculations for the flow through the nozzle. This evidence-based assessment is an important step to demonstrate the viability of applications of quantum computers to simulate fluid dynamic phenomena.
Abstract ID: 49DCASS-095
CFD Analysis of Separate Flow Mixing of a Flow-Through Rotating Detonation Engine
Bret Lane
University of Cincinnati
Ephraim J. Gutmark
University of Cincinnati
A flow-through rotating detonation engine (FTRDE) is a rotating detonation combustor (RDC) with a flow through the center, or core, in place of the usual center body that creates the annular chamber featured in annular RDCs. This configuration of an RDE removes the inner wall and improves on losses due to the wall, however the affects of the core flow on the RDC are less known. The following work was a CFD model which describes the interaction and mixing of flow from the axial core flow and the radially-outward injected RDC air flow over multiple combinations of mass flow rates. Fuel injected into the core and RDC was also modeled to determine what steady-state conditions can be expected before ignition would be attempted, and how fuel and air mixing is affected by air mass flow rates and equivalence ratios. The core fuel is injected upstream through four spray bars in a cross-like shape with multiple holes to cover the area of the core. The RDC fuel is injected axially near the air inlet through many small holes.
Abstract ID: 49DCASS-118
The Effect of Temperature and Pressure on Radar Cross-Section of High-Speed Vehicles
Nathan Bebinger
University of Cincinnati
Prashant Khare
University of Cincinnati
At speeds exceeding Mach 5, also known as the hypersonic regime, there are significant changes in the pressure, temperature, and species composition in the near-field of a vehicle. Designing communication systems for such vehicles requires the understanding of how radio waves travel through high-temperature and pressure environments with varying species concentrations. Despite its importance, apart from a few research investigations focused on the communication blackout and its mitigation, the literature is lean with regard to the analysis of the RF spectrum created around hypersonic vehicles that incorporates detailed fluid dynamics. This presentation discusses the effect of inhomogeneous pressure and temperature fields on the radar cross-section of objects in hypersonic flight. The effect of species concentration will be discussed in a future study. Modeling this phenomenon involves integrating the electric field magnitudes across a domain contained by a defined far-field boundary. COMSOL Multiphysics is a useful tool for computing the estimated radar cross-section of vehicles but restricts analyses to homogeneous domains only. To address this issue, we implemented a methodology in COMSOL with the ability to import experimental or simulated data – including the inhomogeneous pressure and temperature fields – and compute the electric field magnitude at the far-field boundary. In this presentation, we will discuss preliminary results from this framework and show the differences between the RF signature obtained from homogeneous and inhomogeneous pressure and temperature fields around a canonical object flying at Mach 10.
Data Analysis & Uncertainty Quantification
Abstract ID: 49DCASS-012
Architectural Optimization of Emulator Embedded Neural Networks for Aerospace Vehicle Design
James Schmitz
Wright State University
Harok Bae
Wright State University
An approach for the architecture optimization of emulator embedded neural networks is proposed. While the emulator embedded neural network has been shown to provide accurate predictions with suitable emulators, there is still a challenge regarding how to select the hyperparameters of network architectures, i.e., the number of neurons, layers, types of activation functions, etc. The selection of hyperparameters greatly affects the performance of the neural network model both in terms of accuracy and efficiency. To address this challenge, this study proposes an algorithm to test a range of hyperparameters and select the best performing set. The algorithm compares network architectures using average cross-validation error. Additionally, the algorithm implements Bayesian optimization to accelerate the hyperparameter selection process. The proposed method is demonstrated using analytical examples. A representative aerospace vehicle design study will be included as a practical example in the final paper of this ongoing project.
Abstract ID: 49DCASS-029
Interpretable AI for Aerospace Applications
Bharadwaj Dogga
University of Cincinnati
Anoop Sathyan, Kelly Cohen
University of Cincinnati
In this presentation, we make use of explainable AI-based modules - LIME and SHAP to understand the results from a Deep Neural Net (DNN) based regression model applied to a NASA C-MAPPS dataset. These interpretations are to help with decision making for the stake holder in the field of prognosis and health management of jet engines.
Abstract ID: 49DCASS-034
Quantification of TPS material permeability
Donghyun Kim
University of Kentucky
Luis Chacon
University of Kentucky
Savio J. Poovathingal
University of Kentucky
Thermal protection system (TPS) provides robust protection on hypersonic vehicles against scorching-temperature environment caused by the atmosphere. Having a thorough understanding of TPS is crucial for successful aerospace missions. The materials used to compose TPS are manufactured using various techniques. Any methodology inherently introduces a variability to the material structure and properties, and this must be accounted for experimental or computational studies and modeling on the TPS materials. This study attempts to quantify and analyze the stochasticity of a material property, such as permeability, of FiberForm, one of the TPS materials. The samples are prepared to be computationally compatible by utilizing X-ray computed tomography (XRCT), and large numbers of flow simulation using the direct simulation Monte Carlo (DSMC) method are conducted on the samples to acquire data on material characteristics which is then processed to yield probability distribution.
Abstract ID: 49DCASS-036
Parametric and uncertainty quantification study of a fluid-structure interaction model
Michael Belair
University of Kentucky
Ethan Huff
University of Kentucky
Savio J. Poovathingal
University of Kentucky
A model of simulation code on a high-performance computer is developed using Direct Simulation Monte Carlo (DMSC) and Lattice Particle Method (LPM) to simulate supersonic and hypersonic flows over an ice particle which causes it to fracture. The DSMC and LPM are coupled together to produce iterative sequences of the fluid contacting the particle, developing a fracture, and subsequently looping back to the next iteration of fluid flow. Over several iterations, the fracture of the ice particle becomes apparent. However, the limitations of the model include cost and power. Thus, a study to understand the parametric and uncertainty quantification determines the fluid-structure interaction’s computational efficiency for the optimization of the model.
Abstract ID: 49DCASS-045
Uncertainty Quantification by Probabilistic Analysis of Fluid/Solid Interaction
Rama Gorla
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-130
Uncertainty Quantificaiton of Hypersonic Aerodynamic Heating
Laura Holifield
Air Force Research Laboratory
Matthew Tufts
Air Force Research Laboratory
In this study, uncertainty quantification (UQ) analysis was performed on an axisymmetric cone. Training data were generated using three different sampling methods: Latin Hypercube Sampling (LHS), Centroidal Voronoi Tesellation (CVT), and Sobol Sequence. Two different types of surrogate models, a polynomial model and a Proper Orthogonal Decomposition (POD) model with Kriging, were created from the training data in order to perform the UQ analysis with wall heat flux being the quantity of interest. The POD model with Kriging, generated from the training data with the Sobol sequence, was the most agreeable model and was used for the UQ analysis. Three input parameters (P0, T0, and Tw) were varied using two different variations, a 5% variation and the variation from experiment. The UQ analysis showed that a 5% variation in T0 is the most sensitive, followed by a 5% change in Tw, with a 5% change in P0 having the lowest sensitivity. For the experimental variation, a 1.02% change in T0 is the most sensitive, followed by a 3.42% change in P0 , with a 0.17% change in Tw being the lowest.
Abstract ID: 49DCASS-125
Convergence of PIV for Electric Rotor Applications
Peter Sorensen
University of Cincinnati
Daniel R. Cuppoletti
University of Cincinnati
Electric aircraft propulsion is poised to revolutionize the field of aeronautics. To realize this vision, understanding of eVTOL and electric rotorcraft noise is crucial for adoption and acceptance of these novel vehicles. Analysis must be done to understand the source mechanisms of their noise sources. Time-resolved planar image velocimetry (PIV) is still a relatively unexplored technique for electric rotor applications. This presentation aims to look at convergence of PIV data for electric rotors to better understand the amount of data necessary to be confident in data sets. Understanding a reasonable range of data to be measured is necessary to efficiently and accurately obtain PIV measurements. Cases that will be discussed that include single rotor applications of various sizes and two rotors. The focus will be on various methods of assessing this data and convergence of data for periodic PIV for velocity, vorticity, and TKE measurements.
Digital Engineering
Abstract ID: 49DCASS-061
Digital Curation for Aerospace System Product Development
Rick Graves
Air Force Research Laboratory
Kaitlin Henderson
Radiance Technologies
The value proposition of implementing an active digital curation capability within an
aerospace system product development life cycle is qualitatively examined. This first evaluation is necessary to incentivize methodology demonstration and software development tailored for a science and technology ecosystem. For complex digital ecosystems, active
digital curation and resourcing of the appropriate infrastructure is necessary to accelerate secure sharing of digital materiel across diverse stakeholder groups.
Digital curation use cases are examined for specific domains involving modeling and simulation, ground testing, and flight testing activities. Experience generating and curating artifacts for these use cases suggests that active digital curation is useful for informing near-term digital product development and enabling longer-term digital materiel management. Existing examples of digital curation life cycles are examined and discussed. As the time horizon increases, active digital curation may be viewed as an enabling capability for digital materiel management, and becomes a critical-path activity for accelerating digital concept maturation and transition to stakeholders. Active digital curation functionality should be considered as a fundamental requirement for an organization’s digital transformation to augment passive knowledge management best practices.
Abstract ID: 49DCASS-065
Autonomous Navigation of Simulated Unknown Environments using CNN-DNN Network Fusion
Liam Mckenna
University of Cincinnati
Andrew Gerstenslager
University of Cincinnati
Jomol Lewis
University of Cincinnati
Poorva Patel
University of Cincinnati
This paper explores the application of CNN-DNN network fusion to construct a robot navigation controller within a simulated environment. The simulated environment is constructed to model a subterranean rescue situation, such that an autonomous agent is tasked with finding a goal within an unknown cavernous system. Imitation learning is used to train the control algorithm to use LiDAR and camera data to navigate the space and find the goal. The trained model is then tested for robustness using Monte-Carlo.
Abstract ID: 49DCASS-076
Application of Reinforcement Learning in High-Degree-of-Freedom Robotic Arms in Space
Kuan Wen
University of Cincinnati
Anirudh Chhabra
University of Cincinnati
Donghoon Kim
University of Cincinnati
Due to the significant growth of the space industry in the past decade, numerous satellites and their debris have emerged in space. This has led to a substantial demand for spacecraft equipped with robotic arms to perform servicing and debris removal missions. In addition, the scope for space manufacturing using autonomous robot has increased. In space operations, for the precise maneuvering of the robotic arm to reach on-orbit objects, it is necessary to calculate the kinematic and dynamic models of both the spacecraft and the target. However, this process poses challenges due to the complex dynamic coupling between the free-floating base and the robotic arm, as well as the relative movement between the orbiting object and the spacecraft. Furthermore, robot arms with redundant degrees-of-freedom (DOF) require more complex system modeling, thereby affecting the performance due to modeling errors.
This study aims to simplify the complex modeling of robotic systems through reinforcement learning (RL) algorithm. The RL algorithms are incorporated into the control system and trained using the feedback from the environment. Through trial and error, the RL model generates an optimized strategy for path planning and joint control without establishing the kinematic and dynamic models of the robot. In this study, the Mitsubishi PA10 7CE robotic arm, featuring 7 DOF, is constructed in a virtual environment using the MATLAB Robotics Systems Toolbox. Various RL models are trained and tested to investigate how different algorithms and reward functions impact robotic arm operations. Furthermore, energy conservation is crucial on a power-limited spacecraft. This study not only measures the error of the end-effector of the robotic arm but also monitors the energy consumption. In future, developers can refer to the results of this study to choose the most suitable algorithm for application to robot arms in space.
Abstract ID: 49DCASS-096
Hybrid Camera-LiDAR Trilateration with YOLO Based Landmark Detection
Travis Moleski
Ohio University
Jay P. Wilhelm
Ohio University
GNSS denied navigation is notoriously difficult for UAVs due to reduced visibility of satellites and multi-path interference in urban canyons, caves, or forested areas. Current leading methods for navigating in a constrained environment often require additional specific sensing hardware for a localization solution or only provide local frame navigation. Autonomous systems often include LiDAR and RGB cameras for mapping, sensing, or obstacle avoidance that can be leveraged for navigation by providing only or complimentary localization solutions in a global or local frame. Known locations and ranges to landmarks can be leveraged to compute a position estimate using trilateration on a global or local scale. Information from scanning LiDAR has been correlated with camera pixel coordinates and used to range unique visual landmarks with known locations but has only been investigated for color-based detection of glowing spheres. The present work included landmark detection using the You Only Look Once (YOLO) algorithm for object segmentation and localization in an image frame. The benefit of extending the process to utilize YOLO is that the trilateration process can be expanded and adapted to more environments, such as an urban canyon that may have unique signage or street intersections. The modified camera-LiDAR trilateration process was evaluated using signs with known positions, 32 laser scanning LiDAR, and a camera in a simulation environment. Landmark ranging error was evaluated at varying locations within an image while projecting LiDAR to camera points and identifying the point that maps nearest to a single landmark's centroid within a frame. The process was demonstrated in a flight demonstration to showcase the ability to compute a position while moving through an urban canyon environment. Microsoft's Airsim plugin for Unreal Engine was utilized to simulate a multi-rotor aerial vehicle's camera, LiDAR, and flight mechanics in an urban canyon-like corridor for the flight demonstration, but static testing was performed on a ground vehicle. YOLO-based landmark detection was utilized on a trained network of common business signage for landmark identification. The impact is that the process can be applied to more systems with a need for GNSS-constrained positioning on a global or local scale by leveraging the adaptability of YOLO for detection. Hardware testing of the localization process on an indoor ground vehicle is presented as a proof of concept that is compared to Vicon motion capture as a baseline.
Abstract ID: 49DCASS-119
Aerial Vehicle Detection Using Ground Based LiDAR
Jack Kirschler
Ohio University
Jay Wilhelm
Ohio University
Aerial vehicle classification and position estimation are important to maintaining safe and secure airspace which is typically achieved using RADAR systems. With the emergence of advanced air mobility and non-metallic vehicle frames, an alternative method for detecting aerial vehicles is needed, which can be accomplished using ground-based LiDAR object detection. Autonomous vehicles use LiDAR sensors for collision avoidance and can be used to classify and estimate aerial vehicle position. Using a simulated environment built in Gazebo, a Complex-You Only Look Once (Complex-YOLO) algorithm was implemented utilizing various LiDAR sensors and classes of aerial vehicles. The algorithm consists of a Convolutional Neural Network (CNN) to detect and classify objects which were then fed into a Euler-Region Proposal Network (E-RPN) to estimate vehicle orientation and draw a bounding box. The performance of the algorithm was evaluated by comparing the estimated vehicle class, position, and orientation to the ground truth. An Ouster OS1, Ouster OS0 ultra-wide, and a theoretic high density LiDAR sensor were chosen due to their different capabilities and point cloud densities. The vehicle classes ranged from a small hobbyist drone represented by a 3DR Iris+, a high payload drone represented by the Aurelia X8 Max, and an advanced air mobility 2-person air taxi. The presented work provides an understanding of the effects of LiDAR sensor parameters on algorithm performance, as well as displays a RADAR alternative to detecting small non-metallic frame aerial vehicles.
Abstract ID: 49DCASS-124
System Identification and Next-Generation Reservoir Computing for a Micro Turbojet Engine
Colton Wright
Ohio University
Jay Wilhelm
Ohio University
Nicholas Biederman, Brian Gyovai, Daniel J. Gauthier
None
Gas turbines continue to play a large role in the aeronautics industry, and the dynamics of gas turbines are an important area of research. Accurate simulations and modeling of gas turbines are needed for performance optimization, the design of suitable controllers, as well as the integration of the turbine into larger systems. In this work, two different models of a JetCat P100-RX turbojet were created. A physics model of the engine was created using System Identification to estimate parameters required for accurate engine thrust prediction. The physics model is static and does not model engine transients but can be used for steady-state engine thrust predictions and for the validation of other models. Another model was created using Next-Generation Reservoir Computing (NG-RC), a Machine Learning algorithm that is well-suited for learning dynamical systems from observed time-series data. The NG-RC uses no physical model of the engine and is completely data-driven. The experimental data used to train the models was collected by a custom sensor system which measured thrust, exhaust gas temperature, engine shaft speed, requested shaft speed, and pump voltage. Several engine tests were run to collect data for the training and validation of the models. The steady-state performance of both models was evaluated and compared.
Experimental Methods
Abstract ID: 49DCASS-015
Shock-Induced Trailing Edge Separation Effect on F-16 Limit-Cycle Oscillation
Donald Kunz
Air Force Institute of Technology
Eric D. Stubblefield
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-016
Temporally Separated Same-color Dual-plane Stereo-PIV technique Using a Two-legged Burst-Mode Laser
Zifeng Yang
Wright State University
Sirui Wang, Jianyi Zheng, Lei Li, Xunchen Liu, Yi Gao, Fei Qi
None
Dual-plane stereoscopic particle image velocimetry (PIV) is capable of providing three-component (3C) velocity measurements in two near planes simultaneously, thus 3C vorticity vectors can also be evaluated. Traditional dual-plane PIV measurements utilized either two-color lasers or polarized direction of light. In this work, a temporally separated same-color dual-plane stereo-PIV system is proposed and used to measure a swirling flame. A two legged high-repetition burst-mode single-color laser is adopted to avoid the complicated polarization setup or two-color laser requirement. Specifically, a short delay of 2 us is added to separate the two laser sheets and thus the enable separate imaging processes for the two pairs of high-speed cameras using imaging straddling method. The two laser sheets with a gap of 2 mm are generated with the same wavelength of 532 nm and a pulse cyclic frequency of 20 kHz. 3C velocity vectors of the swirling flame are evaluated based on the sequential particle images from each laser sheet. In spite of non-perfect simultaneous PIV measurements on two planes, the error caused by the 2 us delay on top of a 50 us measurement duration is insignificant. The overall uncertainties for the in-plane and out-of-plane velocity vector estimation are 2% and 8.5%, respectively. The uncertainties in the calculated X-, Y-, and Z-component vorticity vectors are 3%, 6% and 6%, which is similar to traditional dual-plane stereo-PIV measurements. This temporally separated dual-plane PIV technique intensively simplifies the experimental setup for dual-plane stereo-PIV measurements.
Flight Dynamics & Controls
Abstract ID: 49DCASS-057
Reconstructing and Reassessing Neil Armstrong's "First Man" Flight in the North American X-15
Will Lorenzo
AFIT Contractor
Timothy Takahashi
Air Force Institute of Technology
This paper uses ASU’s newly developed hypersonic vehicle Stability & Control screening
methodologies to reverse-engineer a Neil Armstrong X-15 flight involving an inadvertent
atmospheric skip, as showcased in the 2018 First Man movie. We provide insight on hypersonic
stability and controllability issues developed by high-speed bank-to-turn vehicles. It appears
that the atmospheric skip was the result of Armstrong losing lateral-directional control of the
vehicle due to a confluence of then overlooked, yet fundamental, aerodynamic instabilities.
Abstract ID: 49DCASS-058
A Reassessment of the Controllable Flight Envelope of the Bell X-1A
Will Lorenzo
AFIT Contractor
Timothy Takahashi
Air Force Institute of Technology
In this paper, we make a total-flight-envelope assessment of the stability and controllability of
the Bell X-1A. This follow-on to the famous Bell X-1 Rocket Plane had an uprated engine and
an increased fuel load that would allow it to accelerate to speeds in excess of Mach 2.5. These
propulsion upgrades were made without changing either the wing, horizontal or vertical tail.
Our flight-envelope approach to understanding performance, stability and control reveals the
“predestined” doom for pilots who attempt to fly this aircraft in excess of Mach 2. We clearly
show that Chuck Yeager’s near-tragic supersonic spin in 1953, was not due to faulty piloting
but due to an airframe lacking fundamental static directional stability.
Abstract ID: 49DCASS-078
Max Range Optimization in Pseudo 5DOF for Lifting Bodies with Heating and Survivability Constraints
Emma Webb
Air Force Institute of Technology
Robert A. Bettinger
Air Force Institute of Technology
This report will detail the methods and results of an optimal control problem (OCP) used to understand the maximum range of a reentering vehicle under heating and deceleration design constraints. A quick assumption of max range can be made on a reentry body by assuming a control effort to obtain the maximum lifting coefficient throughout the reentry. These trajectories often assume planar flight for quick calculations and are useful for determining the absolute practical distance of a reentry. This is often an oversimplification of the problem. Most reentry vehicles have certain design constraints that would limit the available reentry range such as deceleration limits for manned flight, heating constraints, and final velocity constraints for reentering missile platforms. These constraints turn the simple planar reentry problem into an OCP requiring numerical methods to solve. This research found that with an optimal control solution, a 60\% reduction in total stagnation heat load allowable results in an 11\% reduction in the performance metric.
Abstract ID: 49DCASS-079
Towards a Process for Characterizing Control and Stability Properties of Hypersonic Vehicles for Conceptual Design
Jose Camberos
Air Force Institute of Technology
Trevor D. Smiley
Air Force Institute of Technology
To ensure a viable design, the conceptual design phase needs to address vehicle stability and control. This capstone project describes the methods, processes, and tools selected to perform a basic aerodynamic assessment of a hypersonic vehicle and characterize its stability and control properties. The analysis begins with geometry generation using Engineering Sketch Pad (ESP) which then provides a surface ready for meshing. Using CBAero (Configuration Based Aerodynamics software tool), the aerodynamic response provides a means for generating stability derivatives ready for trajectory optimization. The results of this approach provide insight into a hypersonic glide vehicle's stability and control characteristics along with a conceptual-level process for incorporating parametric geometry in the analysis and design of hypersonic systems.
Abstract ID: 49DCASS-090
Enhancing Legacy Interceptors: An Advanced Technique to Emulate Modern Guidance Laws on Classical Guidance Systems
Melody N. Mayle
University of Cincinnati
Rajnikant Sharma
University of Cincinnati
A novel technique is introduced herein intended to augment the capabilities of antiquated defense systems. The solution proposed enables a classical guidance system to follow the trajectory generated by a modern guidance law, extending the operational lifespan of a conventional missile. By inserting a virtual target into an engagement and strategically computing its velocity and turn rate commands, an attacker, despite being guided solely by proportional navigation (pro-nav), can follow the trajectory generated by a guidance law other than pro-nav and intercept the real target, as the attacker pursues the virtual target. The simplicity of this method is captured through a brief analysis and its feasibility is demonstrated through the simulation of multiple modern guidance laws. Enhancing legacy interceptors, this approach contributes modestly, yet significantly, to the Department of Defense's ongoing initiatives towards modernizing the nation's weapons systems while managing resources.
Fluid Dynamics
Abstract ID: 49DCASS-002
Wall-Resolved Large-Eddy Simulation of Flow over a Parametric Set of Gaussian Bumps
Donald Rizzetta
Air Force Research Laboratory
Daniel Garmann
Air Force Research Laboratory
Wall-resolved large-eddy simulations were carried out for the flow over a parametric set of Gaussian bumps, that are representative of surfaces generating smooth-body separation. The geometry and flow conditions were motivated by an experimental investigation, which was conducted in order to provide data for validating approximate numerical approaches. Because the high Reynolds number and three-dimensional shape of the experimental model is challenging, even for approximate numerical techniques, a prior investigation was initiated in order to provide benchmark results that were accessible via wall-resolved large-eddy simulation. It was found that by increasing the bump height, the Reynolds number could be reduced and flow separation would occur. The modified bump then served as a surrogate for the original Gaussian bump producing smooth-body separation. In the present study, solutions were obtained to the unsteady three-dimensional compressible Navier-Stokes equations, utilizing a high-fidelity computational scheme and an implicit time-marching approach. A series of simulations were carried out for bumps of varying heights, for both the three-dimensional configuration and a spanwise-periodic subset, corresponding to flow at the bump midspan. The series includes both fully attached and separated flow situations. A number of metrics are provided to attest to the accuracy of simulations. Comparisons are made between the spanwise-periodic subset, and the three-dimensional configuration, and features of the flowfields are described. For three-dimensional configurations where separation occurred, a highly unsteady arch vortex structure evolved about the speed bump geometry. Its generation and methods to visualized it are outlined.
Abstract ID: 49DCASS-004
Positively Staggered Vertically Offset Propellers in Tandem
Jielong Cai
University of Dayton
Sidaard Gunasekaran
University of Dayton
Michael OL
Folderol, LLC
We continue our study on the aerodynamic investigation of two propellers mounted partially overlapping and in tandem configurations. In the previous study, the rear propeller was placed above the plane of the front propeller (negative stagger). The current study placed the rear propeller beneath the plane of the front propeller (positive stagger). Different vertical and horizontal spacings, incidence angles with respect to the freestream, and advance ratios were tested experimentally at the University of Dayton Low Speed Wind Tunnel. Thrust and torque were measured separately on each propeller. A significant increment in the power required of the tandem propeller system was observed when compared to the results in the previous study under negative staggered conditions. A higher loss in system efficiency at small forward flight speeds is also observed due to propeller wake interaction. The advance ratio at which this performance loss occurs is affected by the propeller spacing and incidence angle.
Abstract ID: 49DCASS-005
Wake Structure of the Prandtl-D
Patrick Hammer
Air Force Research Laboratory
Daniel J. Garmann
Air Force Research Laboratory
A numerical study is undertaken to investigate the loads and 3-D flow structure on the Prandtl-D, a flying-wing demonstrator designed to have a bell-shaped lift distribution. The viscous flow is simulated using Reynolds-Averaged Navier-Stokes (RANS) computations with the NASA Navier-Stokes solver, OVERFLOW. Two Reynolds numbers are considered: 3.43 × 10^5 and 3.00 × 10^6. High-order spatial accuracy is achieved using a fifth-order WENO scheme. The wing was twisted such that it produced a bell-shaped lift distribution at an incidence of 8?. At this balanced condition, no tip vortex is present. Above8?, however, a tip vortex forms on the wing’s upper side, while an underside tip vortex of opposite sign results at incidence below the design condition. The flow also separates in the vicinity of the wingtip at lower angles of attack and over the central portion of the semispan at higher angles of attack. Inspection of the wake at the bell-spanload condition revealed a thin vorticity sheet, consistent with previous results. The lack of a clear vortical structure when the spanload is bell-shaped suggest this condition represents a crossover between tip vortices of opposite sense. The flow-field at the higher Reynolds number had similar character as the lower Reynolds number, with the exception of delayed flow separation.
Abstract ID: 49DCASS-040
A Fluid-Structure Interaction Method for Rarefied Flows
Ethan Huff
University of Kentucky
Hailong Chen
University of Kentucky
Savio Poovathingal
University of Kentucky
A gap currently exists in numerical simulation capabilities for fracturing solids in rarefied flows. Current techniques are unable to capture the effects of solid fracture and rarefied gases simultaneously. One of the primary limitations imposed by this gap is in the study of atmospheric ice particles interacting with the shock layers created by hypersonic vehicles. Potential impact energies are high for vehicles traversing hypersonically through atmospheric ice clouds, creating the danger of surface damage to these vehicles. The small length scales involved in the ice-shock processes necessitate accurate treatment of noncontinuum effects, and the brittle fracture of the ice expected in these conditions requires time-accurate treatment of crack propagation. This presentation reviews a novel computational framework for fluid-structure interactions which can both simulate rarefied gases and general brittle solids. The system is a two-way, loosely-coupled framework. The framework partitions the fluid and solid solution methods and couples them using the marching squares algorithm. Fluids are simulated using the direct simulation Monte Carlo method, a particle-based, stochastic technique for simulation of dilute gases. Solids are simulated with the lattice particle method, a technique well-suited for fracture studies which models solids as collections of material points bonded to one another by spring-like potential energies. Using the positions of material points in the solid, the marching squares algorithm can repeatedly generate updated outer boundaries to place in the fluid domain as the simulation progresses. Aerodynamic surface forces can then be passed from these outer surfaces back to the solid material points. Verification simulations are presented to showcase the accuracy of the framework.
Abstract ID: 49DCASS-054
Verification of Effectiveness of Laminar Separation Control Actuation Frequencies for Eppler 387 Airfoil
Vincent Sheeler
Wright State University
Mitch Wolff
Wright State University
Christopher R. Marks
Air Force Research Laboratory
A variety of aerodynamic devices operate at low Reynolds number conditions, such as unmanned aerial vehicles and low-pressure turbine blades in aircraft gas turbine engines. At low Reynolds numbers, laminar boundary layer separation can reduce the aerodynamic performance of airfoils, impacting the loading and drag. The effectiveness of active actuation methods can be improved when forcing is at locations and frequencies which excite instabilities in the flow. Gross et al. conducted a resolvent analysis for separation control of a Eppler 387 airfoil. Based on the findings, for wall-normal blowing and suction, maximum forcing response occurred at a frequency of 434Hz located at 46.2% of the chord with a forcing amplitude equal to 5% of the freestream. Verification of this study will occur at the U.S Air Force Research Laboratory’s (AFRL) Low Speed Wind Tunnel Facility (LSWTF). The work will utilize acoustic speakers as zero-net mass-flux actuators to generate wall normal blowing and suction on the suction surface. Measurements will include flow velocity, lift, and drag over a range of actuation frequencies to determine the flow response and most effective frequency compared to the analysis. Perfectly matching the actuator used in the resolvent analysis is not possible, as a result, additional study will be required regarding the ideal amplitude and frequency for effective flow control. This presentation addresses actuator characterization thus far, the design requirements of the physical model, and the planned experiment setup.
Abstract ID: 49DCASS-055
Boundary Layer Control with Spanwise-Swept Shallow-Inclined Fluidic Wall Injector
Chase Vansickle
University of Cincinnati
Daniel Cuppoletti
University of Cincinnati
A spanwise-swept, shallow-inclined fluidic injector within a standard flat plate is introduced to the crossflow of a subsonic wind tunnel to study boundary layer parameters. Jet in crossflow using a fluidic injector perpendicular to a surface is a repeated study in which the near field remains the area of interest for determining characteristics of flow interaction. This study concentrates on the far field effects of a shallow-inclined fluidic injector at various skew angles, with the goal of modifying the boundary layer parameters. Fluidic injectors are supplied with several momentum fluxes in addition to the varying geometries to assess how spanwise and streamwise boundary layer parameters including thickness, displacement thickness, and momentum thickness are affected by individual injector variables.
Abstract ID: 49DCASS-062
Dependency of PIV results on acquisition frequency in turbulent flows
Keerthan Ganeshan
University of Cincinnati
Daniel R. Cuppoletti
University of Cincinnati
Particle image velocimetry has manifested to be a reliable method in measuring flow, particularly in wall-bounded flows with unconventional paths, facilitating the investigation of the intricate interactions within the flow field. However, numerous fluid flow explorations have employed high-speed particle image velocimetry (HS-PIV) without a thorough understanding of the impact of acquisition frequency. This presentation is dedicated to initiating the comprehension of this discrepancy in the convergence of the results in PIV. The assessment involves velocity and turbulent kinetic energy (TKE) data derived from a case study of flow structure transport behind a cylindrical bluff body through a shape-transitioning nozzle, also known as a favorable pressure gradient (FPG) nozzle, with a global favorable pressure gradient, which serves as context for the current study. Ensemble results from two acquisition frequencies will be compared based on the analytical calculations in the time domain. Additionally, convergence assessments of the flow field results will be conducted to determine the minimum number of images required to precisely reconstruct the solution. This study will signify the preliminary efforts taken toward nuanced understanding in achieving accurate and converged results. The findings from the comparative study shall contribute to valuable insights for researchers engaging PIV in fluid flow studies, particularly complex flow such as internal flows through shape-transitioning ducts.
Abstract ID: 49DCASS-071
Application of Linear Stability Analysis to Compute Global Stability Modes of Canonical Fluid Flows
Mayur Mahale
University of Cincinnati
Ephraim Gutmark
University of Cincinnati
Linear stability analysis is a well-developed branch of mechanics used to analyze partial differential equations. This method yields the dominant frequencies and their apparent stability. Applying this to a resolved flow field gives us the possible instability modes of the flow, potentially helping predict the behaviour of said flow. The flow field is approximated as a steady state field and a time-varying component. The time-varying component is approximated by a complex exponential function. This reduces the initial equation to the time-varying component form, conserving the initial transformation matrix. Resolving the eigenvalues to this sparse matrix yields the growth pre-factor and frequency. Simple canonical flows are analyzed using this technique to validate and reveal unstable modes in the flow. This study aims to analyze the results obtained from canonical flows and compare them to observed unstable behaviour.
Abstract ID: 49DCASS-083
Thermochemical non-equilibrium hypersonic flow over a rectangular cavity embedded on a compression ramp
Jeremy Redding
University of Cincinnati
Jacob Gamertsfelder
University of Cincinnati
Luis Bravo
Army Research Office
Prashant Khare
University of Cincinnati
This presentation discusses the basic thermochemical non-equilibrium physics that arises when air at Mach number of 11 comes into contact with a rectangular cavity of aspect ratio 2 embedded on a 25-degree compression ramp. To accomplish this research, we use a finite-volume computational framework that solves the compressible form of the Navier-Stokes equations in two dimensions. To model thermal and chemical non-equilibrium processes, a two-temperature model coupled with a five-species 12-reaction chemical kinetics is used. Two simulations are conducted, one with an isothermal boundary and the other with conjugate heat transfer (CHT), to determine the impact of energy transmission to the material on surface heat flux. Primary oscillatory modes and shear layer dynamics are found using fast Fourier transforms and analyzing the near-wall velocity profiles both inside and outside the cavity. We will discuss the dynamical behaviors in the cavity, including the separation region before the cavity, trailing edge effects, frequency analysis of probe data collected at several key locations, and the effect of CHT on surface heat flux, using two states that are representative of the minimum and maximum deflections of the shear layer. It is found that the transrotational temperature close to the cavity's trailing edge is positively correlated with the shear layer oscillations and that the flow characteristics at the cavity's core have a significant impact on the separation upstream of the cavity.
Abstract ID: 49DCASS-105
Path Definitions for Nozzle Ducts with Curvature and Transition
Vincent Onoja
University of Cincinnati
Keerthan Ganeshan
University of Cincinnati
Daniel Cuppoletti
University of Cincinnati
The focus of this research endeavor centers on the physics of internal flows through exhaust nozzles with curvature and shape transitions. These nozzles are designed in a way that the passing flow encounters regions with a global favorable pressure gradient and local adverse pressure gradients. The growing interest in these uniquely shaped nozzles stems from a renewed focus on embedded engine designs, aiming for increased integration between the engine propulsion system and the aircraft body frame.
Previous works have explored the internal flow development problem by designing nozzles based on global geometrical parameters like Length-to-Diameter ratios, area ratio, aspect ratio and centre-line distributions. A common theme in these research efforts monitor the impact these parameters have on flow structure transport and nozzle performance; particularly in nozzle ducts with a centre-line distribution defined by curves as in Lee and Boedicker [1] and cross sections defined by equation of super-ellipses. We wish to extend the discussions by numerically and experimentally exploring the role local internal radius of curvature has in the flow development. With this approach, it is possible to establish fundamental relationships between local internal pressure gradients and nozzle performance.
We present a methodology employing cubic Hermite polynomials that describe the locus of transition from a square inlet to a rectangular cross section at the exit. Consequently, the nozzle comprises four walls: two planar walls (with optical access), a converging wall, and a diverging wall. The converging and diverging walls are characterized by cubic Hermite polynomial functions, allowing precise control of the local curvature at any point along the curves. This way, at any given streamwise location, the influence of local rate of convergence can be compared to rate of divergence and insights can be drawn on how these parameters affect overall flow development.
References
[1] Lee, C., and Boedecker, C., “Subsonic diffuser design and performance for advanced fighter aircraft,” Aircraft design systems and operations meeting, 1985, p. 3073
Abstract ID: 49DCASS-113
Characterization of Nozzle Spray Geometry in Propeller Wake
Sidaard Gunasekaran
University of Dayton
Brock Greenwood
University of Dayton
Abdul Khan
University of Dayton
The interest in agricultural drone spraying is increasing exponentially with more applicators turning to autonomous drones for spraying chemicals such as pesticides, insecticides, and fertilizers. Our prior investigations indicate that there is a drastic change in droplet size distributions from the nozzle when sprayed under the influence of propeller wake. In the current study, a 3D spray patternator was developed which was integrated into University of Dayton Spray Test Section. The patternator was used to quantify the spray geometry from different standard ASABE agricultural nozzles, with and without the influence of a propeller. The nozzles were placed directly underneath the hub of the propeller. The patternator was used to capture the spray pattern from the nozzle tip to the bottom of the spray at every 2 inches. The resultant images were then used to reconstruct the spray geometry for a given nozzle. The differences in the spray geometry with and without the propeller were quantified.
Abstract ID: 49DCASS-120
Expanding the Lower Operating Range of Hypersonic Inlets With the Use of Fluidic Injection
Ryan O'rorke
University of Cincinnati
Daniel Cuppoletti
University of Cincinnati
One current method for expanding the lower operating range of a hypersonic inlet is to use variable geometry. Although variable geometry is an effective control for achieving optimal off design performance for the inlet it comes as a cost of high weight requirements for the mechanical components to move the inlet. Our goal is to achieve similar results in operating range expansion with the use of perpendicular and boundary layer fluidic injection. This will result in a similar result at a much lower weight cost allowing the hypersonic vehicle to perform at a wider range of flight conditions.
Abstract ID: 49DCASS-121
Discrete Vortical Gust Encounter and Mitigation using Closed Loop Control
Andrew Porterfield
University of Dayton
Andrew Killian
University of Dayton
Sidaard Gunasekaran
University of Dayton
Michael Mongin
Air Force Research Laboratory
The open and closed-loop response of a wing to discrete vortical gusts of differing strengths was characterized using Time Resolved Particle Image Velocimetry in the University of Dayton Water Tunnel. Discrete vortical gusts were generated by rapidly pitching a flat plate gust generator upstream of the wing. Simultaneous Time Resolved Particle Image Velocimetry and force measurements on the flat plate wing downstream of the gust generator were collected. A closed loop Proportional Integral Derivative control system was implemented in order to mitigate the influence of the discrete gust on the wing. TR-PIV was used to correlate the flow physics responsible for the force histories experienced by the wing. Peak-to-peak and average gust mitigation using closed loop control were quantified based on the deviations from open-loop results. The controller reduced the force transients during the gust entry by an average of 67% whereas the gust exit transients were only reduced by an average of 6%.
Abstract ID: 49DCASS-123
Using Unsupervised Machine Learning to Reduce the Energy Requirements of Active Flow Control
Jared Kerestes
Air Force Research Laboratory
Christopher Marks
Air Force Research Laboratory
John Clark
Air Force Research Laboratory
Mitch Wolff
Wright State University
In this work, steady vortex generator jets (VGJs) are used to control a laminar separation bubble (LSB) that develops along the suction surface (SS) of a new high-lift high-work low-pressure turbine (LPT) airfoil at low-Re conditions. The goal of this work is not to demonstrate the efficacy of steady VGJs. Rather, the goal of this work is to explore a novel strategy for reducing their energy requirements. Relative to pulsed VGJs, steady VGJs require significantly more energy to be effective but are more realistic to implement in actual application. The underlying idea of this work is simple: activate the VGJs less. Obviously, less activation means less energy consumption. The question becomes, how can the VGJs be activated less while maintaining approximately the same effectiveness? This work proposes that activation be dependent upon whether the LSB is short or long. If the LSB is short, the VGJs are not activated. If the LSB is long, the VGJs are activated only to the extent that the LSB is forced short but not suppressed entirely. As will be demonstrated, this control strategy reduces the energy requirements of steady VGJs considerably while maintaining their effectiveness. One immediate challenge of the proposed strategy is in determining whether the LSB is short or long. Short and long are very abstract concepts that do not lend themselves well to being measured. Machine learning, specifically fuzzy c-means clustering (a type of unsupervised machine learning), is used to determine the LSB type from only three pressure measurements taken in the neighborhood of peak suction. The machine-learning-based control strategy is discussed at length, with emphasis on its benefits relative to a more traditional strategy (i.e., a strategy in which actuation is independent of LSB type) and its overall feasibility.
Distribution Statement A: Approved for Public Release;
Distribution is Unlimited. PA# AFRL-2024-0568
Heat Transfer & Thermal Management
Abstract ID: 49DCASS-014
Influence of Environmental Barrier Coatings on the Overall Effectiveness of a Film Cooled Ceramic Matrix Composite Plate
Shane Lindsay
Air Force Institute of Technology
James Rutledge
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-017
Scaling the Overall Effectiveness of a Film-Cooled Flat Plate Based on a Matched Coolant Reynolds Number
John Di Lella
Air Force Institute of Technology
James L. Rutledge
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-024
Biot Number Error in Low Temperature Inconel Overall Effectiveness Experiments
Bailey Hopkins
Air Force Institute of Technology
James L. Rutledge
Air Force Institute of Technology
Jacob A. Hughes
Air Force Institute of Technology
Carol E. Bryant
Air Force Academy
Awaiting public release.
Abstract ID: 49DCASS-028
Development of a custom supervised learning network to determine the recession rate of ablative thermal protection systems materials.
Vijay Mohan Ramu
University of Kentucky
Savio J. Poovathingal
University of Kentucky
Carbon-carbon and Carbon phenolic ablators are used as heat shield materials to regulate the surface temperature of hypersonic vehicles. The thermochemical phenomenon experienced by the hypersonic vehicles as a consequence of aerodynamic heating has been characterized by physical experiments and detailed flow simulations. However, the state-of-the-art flow simulations are unable to resolve the effects of microscale changes in the carbon based ablators on the mesoscale parameters like recession rate which determine the performance of thermal protection systems (TPS). In this work, we propose a supervised learning model capable of capturing the recession rate of carbon based ablators influenced by oxidation of carbon fiber microstructure. The proposed supervised learning model is inspired by conventional neural networks used for temporal image-based prediction. The training images consists of snapshots depicting the physical changes of carbon fiber microstructure that are obtained through Direct Simulation Monte Carlo (DSMC) simulations in conjunction with a synthetic microstructure generation code called Fibergen, where snapshots of a carbon fiber microstructure subject to oxidation are recorded. A user defined surface function used to capture the overall profile of the microstructure acts as the encoder and decoder of the proposed supervised learning model. Support Vector Regression (SVR) forms the heart of the proposed predictive model by relating the user defined surface function parameters with input variables (porosity, time and microscopic probabilities). The developed supervised learning model is capable of resolving the effects of needling and thinning type of oxidation of carbon fibers on the recession rate. Once trained, the developed supervised learning model is able to predict the recession rate of a carbon fiber microstructure in a very short period (few seconds) as opposed to expensive DSMC simulations. The final model consists of analytical functions directly dependent on the input variables which facilitates the smooth integration of the supervised learning model with the macroscale solvers. In other words, the proposed supervised learning model provides an effective pathway to determine the effects of microscale changes on mesoscale parameters like recession rate which improve the accuracy of state-of-the-art macroscale material response solvers.
Abstract ID: 49DCASS-032
Stochastic uncertainty quantification of structured carbon ablator (FiberForm)
Luis Chacon
University of Kentucky
Ayan Banerjee
University of Kentucky
Donghyun Kim
University of Kentucky
Savio J. Poovathingal
University of Kentucky
FiberForm is a carbon ablator thermal protection system (TPS) used to mitigate the heat load and reduce the maximum surface temperature of spacecraft during atmospheric entry. FiberForm is a material with a structure, and previous research has been done on its general microstructure properties. However, our approach is to create a distribution function of its properties by analyzing the material using different cropping methods. Ten samples of FiberForm were extracted from a billet, and X-ray computed tomography (XRCT) was used to extract the microstructure of the samples. By cropping sub-volumes from different regions of each sample, distribution functions were created, and relationships between the properties were studied. The direct simulation Monte Carlo (DSMC) technique is used to solve the flow of each microstructure and extract the effective permeability using Darcy’s law. Our goal with this study is to better inform material response codes by providing a distribution of the material properties instead of single values as it is currently being done.
Abstract ID: 49DCASS-046
Effect of spectral absorptivity and emission from non-isothermal medium on the thermal response of ablative materials
Ahmed H. Yassin
University of Kentucky
Savio J. Poovathingal
University of Kentucky
Ablative materials are used as thermal protection systems to protect the entry vehicles from extreme heat fluxes attained during hypersonic speeds. The vehicle surface is subject to convective and radiative heat fluxes. Since modeling the radiative heat transfer in participating mediums is computationally expensive, current state-of-the-art material response solvers assume all absorbed radiation fluxes and emissions occur at the surface. When calculating the emission from the material, current state-of-the-art models use surface temperature as blackbody temperature in the Stefan-Boltzmann law ($\sigma T_w^4$) to calculate the blackbody emission. Stefan-Boltzmann law is only valid for isothermal mediums. However, the material is in a non-isothermal state and characterized by a high-gradient temperature and optically thin properties, where emissions occur from in-depth of the material. Also, the current state-of-the-art uses total and hemispherical radiative properties to characterize the emissivity and absorptivity of the medium. However, ablative materials have spectral properties, which is a strong function of wavelength. Thus, the current talk introduces two new models to the current state-of-the-art; the first is an exponential weighted effective temperature (EWET) emission model to correlate the emission from a non-isothermal medium, and the second model is a Bézier fitting function to calculate the spectral emissivity of ablative material as a function of their radiative properties. The present study compares the effect of using the EWET emission model and spectral emissivity with the current state-of-the-art. It was found that assuming blackbody temperature equals the surface temperature results in overestimating the cooling emission term in the surface energy balance equation. This, in turn, causes an underestimation of the temperature profile. The use of total emissivity and absorptivity underestimated the radiation problem, particularly for mediums characterized by high scattering albedos.
Abstract ID: 49DCASS-081
Void Fraction to Quality Correlation Study Applying the Separated Flow Model to Pulsed Power Loads
Zach Carner
Wright State University
Abdeel Roman
Air Force Research Laboratory
Mitch Wolff
Wright State University
Aircraft platforms are continually upgraded with increasingly high quantities of high-powered electronics. As such, efficient thermal management systems are crucially important to overcome system instabilities. To create effective thermal control systems, it is imperative to thoroughly examine and account for a variety of potential instabilities, including those likely to occur due to flow boiling. Thermal instabilities can severely damage electronics and hurt the overall reliability of the aircraft. These instabilities can be described by both void fraction and quality. Electrical Capacitance Tomography (ECT) allows for the collection of capacitance data from the fluid moving through the tube. The vapor, liquid, and the transition between the two, will have different permittivity values depending on the state of the fluid. These permittivity values can be found and recorded by the ECT sensor and easily translate to void fraction measurements. Finding the quality of the refrigerant for pulsed loads becomes more troublesome, especially when trying to apply traditional methods using the energy balance equation. The energy balance equation provides unrealistic results for quality during these profiles. Using correlations between void fraction and quality will provide a more accurate and realistic representation of the quality in the system at a given instant in time. This is because the void fraction is a measured quantity from the ECT sensor and, therefore, is not susceptible to the same flaws as the energy balance equation. Several correlations, each a variation of the Separated Flow Model, are evaluated and compared for accuracy. The results of this experiment will significantly enhance the ability of future thermal control systems to swiftly and efficiently adapt to abrupt fluctuations in the thermal loads they are required to manage.
Abstract ID: 49DCASS-082
Material Response Modelling with an Influence of an RTV Layer
Hilmi Berk Gur
University of Kentucky
Rui Fu
University of Kentucky
Alexandre Martin
University of Kentucky
The advanced volume-averaging thin-layer approach, considered state-of-the-art, proves to be a highly sophisticated technique for accurately predicting material responses, as evidenced in previous research across various materials. Extensive validation in these earlier studies consistently confirms its reliability and precision. In our present analysis, we specifically opted for intricate geometries like iso-q to closely examine the thin layer's behavior. For this investigation, the chosen thin layer material is identical to the bulk material, TACOT, which undergoes a phase change during simulation. The newly introduced method seamlessly aligns with outcomes from the established KATS-MR solver, demonstrating a perfect correlation. The primary goal of this study is to model an RTV binding agent on TACOT material. To accomplish this, we have meticulously computed Arrhenius rate coefficients. This data will be integrated into the existing code to facilitate a comprehensive simulation.
Abstract ID: 49DCASS-087
Thermal Exergy-Based Analysis of the Generic Hypersonic Vehicle
Neal Novotny
University of Dayton
Markus Rumpfkeil
University of Dayton
Jose Camberos
Air Force Institute of Technology
David John Neiferd and Joshua Deaton
Air Force Research Laboratory
The development of future novel aircraft concepts calls for a holistic approach to vehicle design, analysis, and optimization. Aircraft designers can no longer consider components as separate subsystems since these have become more interconnected and multidisciplinary. Hence, we continue to advocate for a new approach to aircraft design and a new way to quantify total aircraft performance. The novel approach to aircraft design we propose relies on the use of exergy methods that can evaluate disparate systems using a universal measure of performance to map global system performance. This paper introduces a new functional and its adjoint gradient in Multidisciplinary-design Adaptation and Sensitivity Toolkit (MAST) to determine thermal exergy destruction rates for aircraft structural analysis. The functional and adjoint gradient implementation demonstrate discrete agreement with analytical test cases. We consider various cases, including a simple fuel cooling subsystem for the Generic Hypersonic Vehicle (GHV) model assuming inviscid and viscous ?ow to obtain the temperature distribution. We compare results to aerodynamic losses in a true one-to-one fashion. Inviscid temperature distributions result in a drastically larger thermal exergy destruction rate than the viscous case due to the decrease in vehicle temperature. From a purely thermal perspective, our results strongly motivate the need to operate the vehicle at a high temperature with as little cooling as necessary.
Abstract ID: 49DCASS-100
Characterizing Coupled Radiative and Conductive Heat Transfer in Atmospheric Re-Entry Vehicles
Colby Gore
University of Kentucky
Savio Poovathingal
University of Kentucky
John Maddox
University of Kentucky
The process of atmospheric re-entry subjects the spacecraft to intense thermal stresses, resulting in the formation of a high-enthalpy boundary layer that surrounds the payload. The resulting thermal stress caused by atmospheric re-entry mandates the use of Thermal Protection Systems, which are capable of limiting the heat transfer from the high-enthalpy boundary layer. To evaluate the induced radiative heat transfer during atmospheric re-entry, two initiatives are currently under development. The first set of experiments will be conducted at the University of Kentucky Paducah Campus, and the second set of experiments will occur in the Hypersonic Materials Environmental Test System (HyMETS) arc-jet facility located at the NASA Langley Research Center (LaRC) in Hampton, Virginia. The local experiments will consist of a modified cut-bar conductivity experimental setup, using near infared (NIR) quartz lamps to induce radiative through fiborous LI-2200. The LaRC campaign will seek to characterize the spectral response, by varying the exposure to the arc-jet and the thermal radiation produced from the NIR quartz lamps. The experimental findings will be employed to broaden computational models utilized to develop TPS materials.
Abstract ID: 49DCASS-101
An Investigation into the Effective Gaseous Thermal Conductivity of Fibrous Insulation Materials
James Senig
University of Kentucky
John F. Maddox
University of Kentucky
During atmospheric entry, vehicles are subjected to large amounts of aerodynamic drag. This aerodynamic drag is the primary mechanism for slowing down the vehicle as it descends through an atmosphere. It also creates a large heat flux across the surface of the vehicle. Thermal protection systems (TPS) are used to protect these vehicles and their payloads from this heat flux. Without a properly designed TPS, the vehicle and payload would get damaged or potentially burn up during entry. Phenolic-Impregnated Carbon Ablator (PICA) is an ablative heat shield material that has been used on several NASA missions such as the Stardust return capsule and the Mars Science Laboratory. PICA and other heat shield materials are chosen based on their low thermal conductivity and ablative properties. FiberForm, a common substrate used for PICA, is a porous, fibrous insulation material. Heat is transferred in fibrous insulation materials by solid conduction, gaseous conduction, and radiation. Each of these modes originates from different processes occurring within the TPS material. Solid conduction occurs between the individual fibers that compose the fiber matrix. Fibrous insulation materials are highly porous, allowing the gases in the surrounding environment to infiltrate the pores of the TPS. These gases facilitate gaseous conduction throughout the TPS. Finally, radiation is emitted from the fibers of the TPS. These three modes of heat transfer vary with pressure, temperature, TPS geometry, and the participating gaseous species. As a vehicle travels through an atmosphere at hypersonic velocities, chemical processes such as ionization and dissociation begin to occur, changing the gaseous composition found inside the TPS. The gaseous composition can also vary depending on the composition of the atmosphere the vehicle is flying through. This research will investigate the gaseous conduction mode of heat transfer and how it varies with pressure and gaseous composition. The results will then be compared to calculations performed using kinetic theory.
Abstract ID: 49DCASS-102
Air-carbon Ablation (ACA) model evaluation with X2 expansion tunnel carbon sample experiment
Ares Barrios-lobelle
University of Kentucky
Savio J. Poovathingal
University of Kentucky
Alexandre Martin
University of Kentucky
Carbon-carbon composites at high temperatures are susceptible to oxidation from atomic oxygen and nitrogen. Attempts have been made previously to capture the gas-surface chemistry via finite rate models. Among the newest generation of models is the Air-Carbon Ablation (ACA) model. This chemistry model includes reactions involving oxygen and nitrogen while hopefully being able to avoid the predictive short comings of previous models of the past. There has been a lack of published literature comparing this model to experimental results and other chemistry. In this study, we take results from a previous run experiments at the X2 expansion tunnel facility in the University of Queensland. The previous experiment and accompanying CFD analysis compared the Park and Zhuluktov-Abe models and found both models lacked the ability to capture the generation of CN in the flow field across a range of surface temperatures. Running these CFD cases with the ACA model can demonstrate the capability of the new oxidation model to it’s predecessors in a case where they were insufficient.
Abstract ID: 49DCASS-106
Implementing two-temperature model in a unified solver: accounting non-equilibirum effect
Seungyong Baeg
University of Kentucky
Raghava S. C. Davuluri
University of Kentucky
Alexandre Martin
University of Kentucky
Spacecraft undergoes high aerodynamic heating during re-entry mission. Majority of this high heat energy dissipates into the surrounding via advection, but the remain amount transfers to the vehicle through conduction and radiation which could damage the vehicle. The space vehicle needs thermal protection system(TPS) to protect itself from this high enthalpy flows. Ablation is one of the TPS method using mechanism that reducing heat transfer to the vehicle by sublimating material and dissipating the heat energy toward opposite side. The numerical research to predict this ablation process has focused on the interaction between fluid and solid. The methods to simulate this gas-surface interaction can be categorized into 4 methods, uncoupled method, weakly coupled method, strongly coupled method, and unified method, and the solver which employs unified approach is used in this research. Two-temperature model is implemented to account non-equilibrium effect in hypersonic flow. Verification work is conducted, and the result of this verification agrees well with reference data. The simulation of solid-fluid interaction case is conducted, and its results are presented. The results show the energy exchange between translational-rotational energy and vibrational-electronic energy modes.
High School
Abstract ID: 49DCASS-010
Creation of Non-Newtonian Blood Analogue with Similar Shear-thinning Properties
Chungyiu Ma
St. Andrews College
Jared Chong
Wright State University
Hang Yi
Wright State University
Zifeng Yang
Wright State University
An vitro experimental study on blood flows requires transparent blood equivalent to enable the inside flow field measurements. To meet this requirement, blood equivalents are constructed of DI water, Xanthan gum (XG) and fluorescent polymer particles (10-45 ?m). The viscosities varying with shear rates are measured using an IKA Rotavisc lo-vi viscometer under the room temperature. The viscosities for pig blood samples collected from a local slaughterhouse in Cincinnati is also measured for validation. The combination of gum and polymer particles with water successfully simulates the shear-shinning properties of blood with an overall relative difference less than 5%. Several artificial blood equivalents with different mixing ratios of DI water, XG, and polymer particles are created and tested. The artificial equivalent solution with 225 ppm XG alone is generated to mimic the non-Newtonian blood viscosity profiles under the body temperature in healthy humans. The other solution with 185 ppm XG and 446 ppm polymer particles with similar viscosities enables the particle image velocimetry (PIV) to measure flow field inside the vascular model. It is well-known that blood viscosity depends on body temperature. To simulate the temperature effect, different mixing ratios are tested to mimic the blood properties under different body temperatures. PIV measurements for flow inside an intracranial aneurysm model are conducted. Detailed flow field is successfully visualized and quantified with the created blood equivalent.
Abstract ID: 49DCASS-013
Leveraging Generative AI and Human Factors Techniques for Building Image Datasets
Isaiah Christopherson
Greeneview High School
Kara Combs
Air Force Research Laboratory
Jose Camberos
Air Force Institute of Technology
As uncrewed aerial vehicles (UAVs) continue to gain in popularity, there are instances when they are presented with scenarios outside of their training library. There is a need to enhance UAVs’ reasoning capabilities to handle these unexpected queries based on its current knowledge. However, high-quality real-world visual reasoning data is difficult to obtain. We propose leveraging text-to-image visual generative artificial intelligence (GAI) models to expand UAV visual datasets and evaluate the quality of results through human factors techniques. The models, Craiyon and Text2Image, were used create AI-generated images that were evaluated on three criteria: number of objects, image quality (evaluation of the image clarity, realisticness, etc.), and representativeness (how well the image “represents” the text used to generate it). An ideal image would have a low number of objects as a high number can be distracting, high image quality (Likert scale of 1-5), and high representativeness (Likert scale of 1-5). Text2Image generates more accurate images compared to Craiyon based on our criteria. Furthermore, we noticed a negative correlation between the number of objects and representativeness and a positive correlation between the image quality and representativeness. Implementing this methodology in a repeatable, automatic form will allow for GAI to continuously evaluate its own images and adjust its settings accordingly, ultimately saving time and allowing for further advancements to implement image recognition within UAVs.
Abstract ID: 49DCASS-035
Investigation of Off-Axis Thrust and Center of Gravity Misalignment on the Dynamic Stability of Mid and High-Powered Rockets
Richard Lian
duPont Manual High School
Helen Bai
duPont Manual High School
In the design of mid and high-powered model rockets, a commonly accepted criterion for stable flight is to have a stability factor of above 1.5 calibers. In our preparation for student rocketry competitions such as the American Rocketry Challenge (ARC), it was observed that more than ten percent of the flights were unstable even though the stability factor was above 1.5 calibers. Our research investigates the possible factors causing rocket instability under statically stable conditions. For that purpose, we analyzed rocket altitude, flight trajectory, and weather condition data from 85 rocket launches and simulated flights using the RocketPy library. The study establishes a custom simulation platform using RocketPy to simulate off-axis thrust, off-axis center of gravity, and wind conditions for mid and high-powered rockets used in student rocketry competitions. Launch footage analysis was also employed to demonstrate that thrust misalignment induces larger pitch rotations after launch, leading to unpredictable flight trajectories that often deviate horizontally from a nominal vertical trajectory. The analyses reveal that off-axis thrust can be a key factor that causes unpredictable flight trajectories. The second factor is the off-axis center of gravity. To mitigate these factors, we propose increasing the static stability margin and increasing the longitudinal moment of inertia of the rocket. The findings are validated through RocketPy simulations that demonstrate a strong correlation between real-world flight data and simulation predictions of rocket altitude and flight trajectories in similar wind conditions. By accounting for lateral thrust and center of gravity displacement, the study provides robust recommendations for designing mid and high-powered rockets with increased stability and more predictable flight trajectories.
Abstract ID: 49DCASS-041
Deep Learning based Optical Flow Analysis
Daniel Zhang
Farragut High School
Zifeng Yang
Wright State University
Quantifying the velocity field in high-speed flows is crucial for understanding flow dynamics, turbulence, and flow-structure interactions. While optical velocimetry techniques typically offer sparse information, achieving dense velocity vector fields with high spatial resolutions is essential for in-depth analysis of complex motion patterns and accurate motion tracking within the field of view. In this study, two-dimensional (2D) Rayleigh scattering imaging at a rate of 10- to 100-kHz was employed to measure high-speed flow velocity. This was achieved through the application of deep learning-based optical flow analysis, incorporating data from Rayleigh scattering intensity profiles related to density and temperature fields. The high spatial resolution of Rayleigh scattering images, featuring smooth gradients and precise tracking of flow features, facilitated the extraction of 2D instantaneous velocity fields. The deep learning-based optical flow method utilized a recurrent neural network architecture to extract per-pixel features from input images, calculate correlations between all pairs of features, and train by recurrently updating the optical flow. This approach extended Rayleigh scattering imaging as a non-intrusive, non-seeded, and multiscalar measurement technique for high-speed nonreacting and reacting flows.
Abstract ID: 49DCASS-047
From Curiosity to Career: The Impactful Role of TechFest Dayton in K-12 STEM Outreach for AIAA and the USAF
Kara Combs
Air Force Research Laboratory
Jose Camberos
Air Force Institute of Technology
TechFest Dayton represents the Miami Valley’s premier K-12 STEM (science, technology, engineering, and mathematics) outreach event, held in February each year. At its core mission, TechFest Dayton seeks to inspire generations of STEM professionals through interactive exhibits, hands-on activities, and captivating demonstrations. The event provides an accessible and broad platform for students (of all ages) to explore the wonders of STEM across multiple disciplines, ensuring that all attendees discover their passions.
The potential to extend its impact beyond the classroom and into the professional world sets TechFest Dayton apart. Held at Sinclair Community College, TechFest finds itself uniquely positioned at the heart of innovation in Dayton, OH, with Wright-Patterson AFB nearby. Wright-Patterson AFB’s support not only enhances the event's offerings but also provides students with a rare glimpse into the exciting world of aerospace and national defense, a popular industry among STEM students.
The presentation will delve into the collaborative initiatives between TechFest, the American Institute of Aeronautics & Astronautics (AIAA), and Department of the Air Force personnel, showcasing how such partnerships contribute to shaping future STEM professionals who may find their calling in fields crucial to national security and technological advancement. In our presentation, we will highlight people-driven improvements to TechFest and success stories on how TechFest has impacted attendees’ interests and careers. We will also discuss current opportunities to support TechFest and ways by which TechFest can support the Dayton community, the local STEM ecosystem, and defense fields. Our talk will explore how this event serves as a catalyst for educational and professional growth. Through a combination of engaging STEM experiences and strategic partnerships, TechFest creates an environment that empowers students to transition from the realm of curiosity to a fulfilling STEM career, shaping the future of both education and national defense.
Abstract ID: 49DCASS-131
Film Cooling Effectiveness on Ceramic Matrix Composite
Sophia Majors
Air Force Research Laboratory
Within the gas turbine engine, a high pressure turbine is located immediately downstream of the combustor, leading to some of the highest temperature readings within the engine. Film cooling is employed to create a blanket of cooler air over turbine components, providing thermal protection from the combustion's hot exhaust gasses. The main goal of turbine cooling research is to maximize cooling effectiveness while using the least amount of coolant flow. In the Heat Transfer and Aerothermal Laboratory, researchers have been exploring the capabilities of Ceramic Matrix Composite (CMC) as an alternative to nickel-based superalloys within testing facilities. This is done using the Elevated Temperature Advanced Cooling Rig (ETACR). The ETACR is a high speed, high temperature heat transfer and cooling effectiveness wind tunnel. The primary measurement technique used in this research is infrared (IR) thermography, which detects and converts radiant emission into surface temperature distributions. Surface mounted thermocouples are used for calibration of the IR thermal images. CMC is a material of interest for hot section component design due to its high temperature capability, which exceeds that of nickel-based superalloys. CMCs also represent significant weight saving at approximately 1/3 the density of metallic systems. These CMC characteristics are key enablers for higher performance aircraft engines.
Imaging & Diagnostics
Abstract ID: 49DCASS-023
A Generalized Approach to Synthetic PIV images
Dilip Kalagotla
University of Cincinnati
Paul Orkwis
University of Cincinnati
The use of Particle Image Velocimetry (PIV) to study complex flow fields has become common because of its ease of implementation. However, the assumption that the particles track the flow faithfully remains. To tackle this issue, the whole PIV process has to be simulated. In our previous work, the entire framework developed to simulate the PIV was explored. The final step in the framework is the imaging phase in PIV. A new code (syPIV) was developed to generate synthetic PIV images. Traditionally, these images were used to test and validate PIV algorithms. In the current case, these images were used as a final step in the simulation of the whole PIV process. The syPIV code and its use will be explored in this work. The testing and validation cases used for the current code are presented with insights into the effect of particle presence in PIV.
Abstract ID: 49DCASS-068
Exploring the Validity of Inverse Filtering for Flow Rate Estimation: A Comparative Analysis of Direct and Indirect Measurements in a Vibrating Nozzle
Jacob Michaud-dorko
University of Cincinnati
Charles Farbos de Luzan
University of Cincinnati
Ephraim Gutmark
University of Cincinnati
Liran Oren
University of Cincinnati
Inverse filtering, a signal processing technique used to recover an original signal, holds significant promise in estimating flow rates from recorded differential pressure signals across a known fixed airway resistance. While delineating flow rates from recorded pressure signals has been widely employed in research, the formal validation of inverse filtering has encountered challenges, primarily due to the difficulty in obtaining direct measurements of flow rate waveforms. This study represents a crucial initial step toward validating inverse filtering by comparing the flow rate waveform derived from inverse filtering with direct flow rate waveform measurements at a pulsating nozzle exit.
The experimental setup incorporates a sound-producing oscillating valve connected to a downstream resonator, with volume flow fields measured using phase-locked tomographic particle image velocimetry. In addition, a circumferentially vented pneumotachograph device connected to the resonator outlet measures the differential acoustic pressure across a known airway resistance for outlet flow waveform calculations. The outlet flow rate waveform collected by the pneumotachograph device is then subjected to inverse filtering to delineate the flow rate waveform upstream at the vibrating nozzle exit.
This investigation explores the impact of varying upstream reservoir pressures and near-field and far-field constrictions downstream of the flexible nozzle. Results demonstrate that inverse filtering effectively approximates the flow rate waveform at the nozzle exit, with accuracy influenced by upstream pressure conditions and resonator constrictions. The observed differences between direct and indirect methods are attributed to increased acoustic impedance in the downstream resonator. This study contributes valuable insights to the ongoing efforts of validating the applicability of inverse filtering for precise flow rate estimation in engineering research.
Abstract ID: 49DCASS-107
Quantitative Concentration and Density Measurements in Subsonic and Supersonic Helium Jets Using Rainbow Schlieren Deflectometry
Taber Wanstall
University of Dayton
Henry Jacques
University of Dayton
Carson Running
University of Dayton
Rainbow Schlieren Defectometry (RSD) has been applied to acquire quantitative concentration and density measurements in both subsonic and supersonic conditions. Experiments were conducted using a helium jet expelled into the air for two distinct regimes: initially laminar, momentum-driven at subsonic speeds, and subsequently at supersonic speeds. Comprehensive full-field measurements were captured, encompassing the laminar, transition, and fully turbulent regions of the jet in both subsonic and supersonic conditions. The subsonic (momentum-driven) regime is validated against Rayleigh scattering data in the literature. The supersonic regime utilizes a novel approach for acquiring density, where the mixing field is delineated into three portions: the potential core, the non-isobaric mixing shear layer, and the isobaric far field. Results from the RSD technique demonstrate robust mixing measurement capabilities for both subsonic and supersonic applications. The results of this work show promise for future applications involving the use of quantitative RSD for fuel injection studies.
Abstract ID: 49DCASS-116
Multispectral Infrared Imaging of a Scramjet
Nathan Childs
Wright State University
Mitch Wolff
Wright State University
Timothy Ombrello
Air Force Research Laboratory
Mid-Infrared imaging was used to visualize the flow coming from the exhaust of a supersonic reacting flow. Previous work from other researchers encountered complications while imaging a supersonic combustor where the windows reduced the transmitted wavelength range to 1.5-3.5µm, which blocks the wavelengths of CO and CO2 emission. Images were collected of the radiation intensity coming from H2O, Hydrocarbons, CO, and CO2 which were acquired using a mid-infrared camera imaging the exhaust flow with a continuously rotating filter wheel. This expands the capabilities of the technique allowing for the CO and CO2 bands to be imaged. Future work will quantify radiance from each discrete wavelength band to calculate the temperature and concentrations of the species in the flow, and ultimately towards a computed combustion efficiency.
Materials & Structures
Abstract ID: 49DCASS-003
Taylor Test
Anthony Palazotto
Air Force Institute of Technology
Katie Bruggeman, Showmik Ahsan, Henry Young
Wright State University
John Hansen
Air Force Institute of Technology
An experiment has been performed at the Air Force Institute of Technology using their Taylor test facility. The Taylor test is an experimental system allowing high-velocity projectile movement into a rigid plate. The idea was to evaluate a specimen’s design with varying materials and record their high strain rates through their collision. In this set of tests, IN 718 was being compared at a velocity of 151m/s against a 1 ½ inch thick steel plate. The velocity was recorded by a photographic camera system obtaining photos at a range of 28000 frames per second. Two types of specimens were considered, one made from ½ “
x 2” heat-treated extruded IN 718, and the other additively manufactured (AM). It was found that the strain recorded between the two was quite different based up their Modulus of Elasticity. The density of the two types of specimens was very close giving a mass of equal quantity, but due to the moduli the strain and strain rates differed. The stress also differed. Stress values were based on the momentum reached in the experiment and became a function not only of the velocity but also the product of the modulus of elasticity and density. A major issue in the experiment turned out to be the aerodynamics. In certain cases, it was found that the projectile was greatly affected by the initial system support. Specific frictional restraint could produce erratic behavior of the small specimen. In order to overcome this, attention was directed to reducing the surface friction between the gun barrel and the specimen.
Abstract ID: 49DCASS-008
Computational Modeling of Single Shear Bolted Joint Hybrid Composite Laminates with and without Film Adhesive
Cameron Mcmahan
Air Force Institute of Technology
John Brewer
Air Force Institute of Technology
Micheal Gran
Air Force Research Laboratory
While composites are prized for their strength and stiffness, they are prone to unpredictable failure at stress concentrations such as bolted joints. The development of a hybrid composite with localized stainless-steel foils in between and in place of composite plies has proven to be effective in increasing joint bearing strength. This was originally shown using adhesive layers to bond stainless steel foils within the laminate during the primary cure. Then, Bellanova proved that the hybrid behaves predictably without the film adhesive, which created a stronger joint with less complexity.
Since Brewer and Bellanova focused on experimental efforts, their modeling efforts covered only double-shear configurations. This effort focuses on the Finite Element Analysis (FEA) of the more complex single shear configurations. Specifically, protruding head and countersink head configurations and considered in control, adhesive hybrid, and non-adhesive hybrid layups. These computational models are compared to experimental results in overall performance at the load-displacement level and also at the ply level.
Abstract ID: 49DCASS-011
Structural Index Parameter for Capturing Aerothermal Effects in Conceptual Level Vehicle Design
Samuel Atchison
AFIT Contractor
Jose Camberos
Air Force Institute of Technology
The three phases of vehicle conceptual design include parametric sizing, configuration
layout, and configuration evaluation. During the parametric sizing phase, the ability to
define and quantify the technology level of an aerospace system allows the assessment of
candidate designs based on feasibility given current technology or indicates if one must advance
a particular technology. To meet this need, the structural Index (Istr) parameter merits
exploration to consider structural and aerothermal effects during the parametric sizing phase
of conceptual design given materials, structural concepts, and manufacturing capability. This
study showcases the utility of this structural/materials technology parameter for high-speed
vehicles by modernizing and expanding upon Paul Czysz’s original structural index (Istr) versus
temperature map. The construction of the modernized and expanded structural index (Istr)
map is accomplished by selecting a temperature-through-thickness method for a given thermal
protection system (TPS), which simplifies a given temperature and pressure profile into a
constant heat pulse. One can then size the TPS to keep the structural temperature within
material limits. The newly generated structural index (Istr) maps allow one to observe trends
with variations in temperature, cruise time, average atmospheric pressure (Pavg), and TPS
materials. Finally, we present an outlook on improving the structural index (Istr) maps further
through multi-physics modeling and usage in trade space exploration to assist in directing
hypersonic materials and structures research.
Distribution Statement A: Approved for Public Release; Distribution Unlimited. PA# AFRL-2023-6446
Abstract ID: 49DCASS-019
Output only modal analysis of a flexible manipulator
Viet-hung Vu
Royal Military College of Canada
This presentation is about the output-only modal analysis of a flexible manipulator. The technique is to use only the acceleration responses from dynamic testing or from normal operations to identify the real modal analysis of the manipulator structure. The multivariable autoregressive model is employed to fit the signal data and the least squares method is used in the form of the QR factorization to estimate the model parameters. Modal characteristics such as the natural frequencies, damping ratios, and mode shapes are identified from the estimated state matrix. A good result matching with the traditional impact modal testing shows that the method can be used for the modal analysis of structure when it is in operation or when it is impossible to measure the excitation such as in aerospace systems.
Abstract ID: 49DCASS-021
Spectral measurements of wavelength dependent radiation properties of LI-2200.
Yejajul Hakim
University of Kentucky
Savio J. Poovathingal
University of Kentucky
Michael W. Renfro
University of Kentucky
The generation of shock layers during atmospheric re-entry at the stagnation point of hypersonic
vehicles leads to exceptionally high heat loads, which are subsequently transported
to the Thermal Protection System (TPS) surface through convective and radiative heat
transfer, eliciting a significant thermal response. The objective of this experiment is to measure
the wavelength and angle dependent transmission and backscattering of radiation in
LI-2200 to support improved modeling of radiation transport through TPS materials. These
properties are assessed by focusing on a broad range of wavelengths, spanning from ultraviolet
(UV) to infrared (IR) (300-1100nm), employing two distinct types of fiber-coupled
light sources. Two samples of LI-2200, with thicknesses of 0.6mm and 1.3mm, are used
to measure these properties. These samples are encased between two glass slabs for secure
mounting. Two different fiber-collimator configurations are employed as the receiver to collect
the transmission through the TPS. Both the sample and the receiver are mounted in
motorized stages which can be rotated to vary the angle of incidence and angle of collection.
Measurements for transmission and backscattering are collected at 5 degree intervals. For both
transmission and backscattering measurements, a total of 25 different combinations of angles
were recorded. The spectral response of LI-2200 is characterized through normalization with
the 0 degree case. In these measurements, light source is of low power thus the surface’s heat load
can be considered negligible. However, modeling the transmission and scattering response
will be employed to ascertain fundamental radiative transport parameters, which can then
be utilized to calculate the overall heat load for reentry scenarios of interest.
Abstract ID: 49DCASS-026
Characterization of Axial-Torsion Yield Surfaces of Additively Manufactured Ti-6-4 at Varied Layer Heights
Joseph Puglisi
Air Force Institute of Technology
John Brewer
Air Force Institute of Technology
Ryan Kemnitz
Air Force Institute of Technology
Elizabeth Bartlett
Air Force Research Laboratory
Awaiting public release.
Abstract ID: 49DCASS-033
Observing and modeling differential ablation in C/C composites
Cameron Brewer
University of Kentucky
Savio J. Poovathingal
University of Kentucky
Differential ablation is a product of the differing densities found in the matrix and fibers of a C/C composite. This work seeks to observe, quantify, and model differing rates of ablation. Due to the diversity in C/C composite orientations and arrangements, the structures which have been selected for this study include the following: a loose carbon fiber weave with relatively low amount of carbon matrix, a tight carbon fiber weave with relatively high amount of carbon matrix, and randomly oriented carbon fiber bundles encased in a carbon matrix. The C/C composites are oxidized at 1000oC in a tube oven within a quartz glass tube. The generation of carbon monoxide and carbon dioxide gas is monitored by connecting the exhaust of the oxidizing sample in series with a residual gas analyzer (RGA). The microstructure of the C/C sample is imaged before and after oxidation is performed using a scanning electron microscope (SEM). In addition to experimentation, a model is being created to emulate the phenomena of differential ablation within the C/C sample in the case where the incoming flow is parallel to the fiber orientation. Non-uniform recession of the matrix surface is being performed through the development of a parallelized marching cubes algorithm which will be coupled with an existing support vector regression model for ablating the individual fibers. Direct Simulation Monte Carlo is used to render the flow of the incoming gas. The coupling of these two codes should result in a cohesive computational model for differential ablation.
Abstract ID: 49DCASS-042
High-Fidelity Melt Pool Prediction with a Physics-Guided Heat Source for Accelerated Laser Powder Bed Additive Manufacturing Simulations
Abdullah Amin
University of Dayton
Robert Lowe
University of Dayton
Nishat Sultana
University of Cincinnati
The grain size in parts fabricated using laser powder bed fusion (LPBF) additive manufacturing (AM) is significantly influenced by the dimensions of the melt pool. These melt pool dimensions, in turn, are determined by manufacturing process parameters such as laser power, scan speed, spot diameter, and hatch spacing. As these process parameters are crucial in controlling grain size, they directly and meaningfully influence the final properties of as-manufactured parts. Thus, there is significant need for simulation-based tools that (a) provide a fundamental understanding of the dynamics of melt pool formation and (b) predict its evolution with reasonable computational resources and within an acceptable timeframe. Customary physics-based approaches – although effective – are computationally intensive and time-consuming, owing to the complex interactions across multiple physical fields in the LPBF AM process. This significant computational expense limits the feasibility and applicability of these customary approaches at scale. To address this, we introduce a novel physics-guided methodology for accurately modeling the laser heating aspect of the LPBF process. This new approach, informed by experimental data, links input process conditions to an optimized definition of the heat source. Three different methods for approximating the heat source are evaluated and validated using benchmark experimental measurements (NIST AM-BENCH 2022). Our study focuses on simulating five key LPBF parameters: solid cooling rate, liquid cooling rate, time above melting, melt pool width, and melt pool depth. These were then compared with experimental measurements for IN718. The simulation results show excellent agreement with the experimental measurements, underscoring the effectiveness of our physics-guided heat source model. Notably, this approach has the potential to expedite computational modeling by several orders of magnitude compared to the pure physics-based modeling approach, suggesting potential feasibility at part scale and potentially offering a significant advancement in the field of additive manufacturing and part property predictions.
Abstract ID: 49DCASS-048
Statistical variance in radiative properties of porous materials
Ayan Banerjee
University of Kentucky
Luis Chacon, Yejajul Hakim, Ahmed H. Yassin
University of Kentucky
Michael Renfro, Savio J. Poovathingal
University of Kentucky
In the quest for planetary exploration, space vehicles play a significant role in the transportation of payloads. A crucial component of the space vehicle is the thermal protection system, which protects the vehicle from overheating while entering a planetary atmosphere. The entry is a high-speed process which can result in significant radiative heat loads on space capsules. To understand the material response of heat shields, an integrated simulation framework is required which couples computational fluid dynamics with material response solvers. The mechanism behind transport of radiative signatures is fundamentally different from the conductive mode of energy transport, and penetration of radiative signatures depends on the radiative coefficients of the thermal protection system (TPS) material that protects the space capsule. Radiative coefficients depend on material architecture and vary based on microscale features and manufacturing defects that result in inconsistency across billets of the heat shield. Combining with other microscale properties like permeability, stress, and thermal conductivity, variability in radiative properties will lead to variations in the response of heat shields. In this study, a statistical approach is used to generate radiative coefficients of FiberForm microstructures, and these coefficients are correlated to microscale features such as porosity.
Abstract ID: 49DCASS-073
A Skin-Stabilizing Constraint for Feature-Based Topology Optimization of a Wingbox
Hollis Smith
National Research Council
Joshua Deaton
Air Force Research Laboratory
Structural stability is an essential requirement of aerospace structures.
However, the substantial computational cost of conventional eigenvalue buckling constraints has posed a significant barrier to their inclusion in the topology optimization of large-scale airframe structures. To address this inadequacy, we introduce a heuristic buckling constraint for the 3D topology optimization of a wingbox that targets localized skin-buckling modes. The proposed formulation is defined on a field variable mapped from the design, and thus is not restricted to a particular design representation. The feature-mapping approach enables the proposed method to produce layouts composed exclusively of vertical plate-like members that can be manufactured using industrial processes consistent with those of conventional spar/rib construction. The results demonstrate the effectiveness of the method at retaining stabilizing sub-structure that stiffness-based optimizations remove. Crucially, the heuristic stabilizing constraint is evaluated in a fraction of the time of the elastostatic analysis, promoting skin-stiffening structures at a significantly reduced computational cost in comparison to eigenvalue buckling constraints.
Abstract ID: 49DCASS-110
A Novel and Straightforward Approach for Determining the Post-Necking True Stress-Strain Response of Aerospace Metals
Yatik Shah
University of Dayton
Robert L. Lowe
University of Dayton
To numerically simulate and predict the plastic deformation of aerospace metals and alloys during extreme impact events (e.g., turbine engine blade-out and rotor-burst events, bird strikes, and foreign object damage), accurate knowledge of the metal’s hardening behavior at large strains is requisite. Tensile tests on round cylindrical specimens are frequently used for this purpose, with the metal’s large-strain plasticity ultimately captured by a true stress vs. true plastic strain curve. During tensile testing, the strain field in the specimen gage section evolves from a nearly homogeneous profile prior to necking to a heterogeneous profile after the onset of necking. Concomitantly, the customary analytical relationships used to convert between engineering stress-strain and true stress-strain break down after necking, since the state of stress is no longer homogeneous or uniaxial after necking. Thus, a number of approaches have been proposed and employed to correct the post-necking hardening response. Although effective, these approaches are generally complex and/or computationally expensive, which can be particularly problematic when deployed within a large experimental program. In this talk, a novel and efficient post-necking correction method is proposed and benchmarked. Using the equivalent true strain history obtained from a digital image correlation virtual strain gage placed at the fracture location, an approximate first-order analytical approach is used to calculate the corresponding equivalent true stress. This true stress calculation is used to generate a simple post-necking hardening law, using linear interpolation between known true stress-strain states at necking and fracture. This approach is successfully benchmarked using experimental data from a suite of metals with different crystal structures and hardening behavior: Inconel 625, Inconel 718, 17-4 precipitation hardening (PH) stainless steel, and Ti-6Al-4V titanium alloy.
Abstract ID: 49DCASS-112
Preliminary Structural Design and Analysis of a 27U CubeSat
Matthew Evans
Air Force Institute of Technology
Robert Bettinger
Air Force Institute of Technology
John Brewer
Air Force Institute of Technology
Carl Hartsfield
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-114
Mechanical Properties of Low Dimensional Materials
Arun Nair
Air Force Institute of Technology
Carbyne is a one-dimensional carbon chain that could be considered the strongest material. Recent experiments have shown that stable chains with lengths of thousands of atoms can be developed. As carbyne manufacture advances, it is important to determine possible applications for use in nanocomposites. However, the interface properties of these carbon-based materials with metal matrix could affect the nanocomposite mechanical properties. Thus, it is important to study the interface properties and perform mechanical testing of the nanocomposites embedded with carbon-based low dimensional materials to determine the most efficient material combinations to improve the mechanical properties. We use computational methods to predict the mechanical properties of metal-carbyne nanocomposites composed of multiple carbyne chains sandwiched within a metal matrix.
Using molecular dynamics and density functional theory methods, we develop and study the interface properties of carbon-based materials embedded in metal matrix. We then conduct mechanical testing of the nanocomposites to determine if the nanomaterials improve their mechanical properties when compared to the metal matrix. From the mechanical property prediction and interface analysis, it is possible to determine the nanocomposite that would be most compatible for future development and applications.
Orbital Mechanics
Abstract ID: 49DCASS-063
The Circular Restricted N-Body Problem (CRNBP) in the Jupiter-Europa System
Annika Gilliam
Air Force Institute of Technology
Robert Bettinger
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-067
Heuristic Search Method to Optimize Development of a Cislunar Space Situational Awareness Architecture
Jacob Dahlke
Air Force Institute of Technology
Robert Bettinger
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-086
Historical Retrospective: 3- and 6-Degree-of-Freedom Analysis of the Magellan Aerobraking Experiment at Venus
Erica Higginbotham
Air Force Institute of Technology
The first non-Earth aerobraking experiment occurred in 1993 as part of the Magellan spacecraft’s mission to Venus. Magellan reduced its orbit from elliptical to nearly circular over a span of 70 Earth days by leveraging the non-conservative force of aerodynamic drag to reduce its orbital energy via transiting the upper region of Venus’ sensible atmosphere. First tested with the Hiten probe in the Earth-Moon system in 1991, the subsequent Magellan experiment helped prove the viability of aerobraking for planetary missions and paved the way for its implementation in Mars missions starting in the late 1990s and the 2014 Venus Express mission. This research, for the first time in literature, investigates the Magellan aerobraking experiment from the perspective of both 3- and 6-degree-of-freedom analysis. Multiple atmospheric density models are considered, as well as J4 gravitational perturbations to understand the complexity and sensitivity of numerous aerobraking maneuvers in the thick Venusian atmosphere. Alternative vehicles of varying size are studied to determine their capability of maintaining the Magellan aerobraking flight profile. To examine a wide range of vehicle mass, drag reference area, and vehicle configuration options, the Hiten, Magellan, and Hubble Space Telescope (HST) satellites are modeled and their aerobraking performance analyzed for the hypothetical case of Venus orbital operations. When reconstructing the historical Magellan aerobraking maneuvers, 3DOF and 6DOF analysis yields minimal deviation with the Magellan and Hiten models, while more significant deviation with the larger HST model. Both 3DOF and 6DOF analyses also reveal the sensitivity of the atmospheric density model, relative to the coinciding solar cycle, and indicate a higher fidelity gravity model is necessary for Venusian analysis. Overall, aerobraking is a viable maneuver option for adjusting semi-major axis if the vehicle performing the maneuver is capable of countering aerodynamic torques in the range of 1-6 Nm depending on vehicle size and initial attitude orientation and can save a significant amount of propellant and cost when imparting a large change in semi-major axis.
Space Systems
Abstract ID: 49DCASS-030
Instrumentation Design for KRUPS Atmospheric Entry Capsules
Bruno Domingues Tacchi
University of Kentucky
Matthew P. Ruffner, Kirsten Ford, Logan Craig,
University of Kentucky
William Smith, Alexandre Martin, Savio J. Poovathingal
University of Kentucky
The Kentucky Re-entry Universal Payload System (KRUPS) is a small spacecraft developed to deliver flight data during atmospheric re-entry and serve as a test bed for thermal protection system (TPS) materials. After the success of the Kentucky Re-Entry Probe Experiment (KREPE), a new iteration of the KRUPS capsule was developed for KREPE-2, the upcoming second orbital flight of KRUPS. The main modifications are a new sensor suite consisting of six thermocouples, five pressure transducers, a spectrometer, an inertial measurement unit (IMU), an accelerometer, and a global positioning system (GPS).
This new iteration of the instrumentation suite is an important improvement from the first KREPE mission - where four thermocouples were the only sensors. The mini spectrometer being added to the capsule will capture spectral emissions from the shock layer. Five pressure ports and transducers will be added to the capsule’s TPS to capture pressure data, allowing a more accurate trajectory reconstruction and estimation of freestream pressure. The IMU, accelerometer, and GPS will measure the capsule’s accelerations, angular velocity, and flight path, which further increases the accuracy of the trajectory reconstruction. These new sensors are in addition to the six thermocouples, which provide TPS temperature data - vital information to design next-generation TPS materials.
This work aims to provide an analysis and scientific justification of the instrumentation for the KREPE-2 mission: information on how each sensor was chosen, its function, placement in the capsule, and the design process required to include it in the capsule.
Abstract ID: 49DCASS-031
Simulating Martian Environments for Testing of Localization Algorithms
Steven Kraine
University of Cincinnati
Anirudh Chhabra
University of Cincinnati
Donghoon Kim
University of Cincinnati
Accurate estimation and localization on the Martian surface is a critical challenge for various Mars exploration missions, including rover operations, aerial drone navigation, and habitat placement. Existing estimation and localization techniques designed for Earth-based environments often prove inadequate for Mars due to the planet's unique conditions, including the lack of traditional GPS systems, the presence of uneven and rocky terrain, and the different gravitational forces. Therefore, there is an urgent need to establish a comprehensive testbed specifically tailored to assess and advance estimation and localization technologies for Mars applications. To tackle this problem, a testing framework is proposed using the Robot Operating System (ROS) and Gazebo tools. To create the testbed, Martian terrain was generated by modeling Perlin noise along with additional craters to mimic the landscape. Then, ROS nodes were developed to emulate the ultra-wideband sensors used for creating the positioning system. Then, additional ROS nodes were developed for simulating a simple Kalman filter-based localization algorithm. In addition to these, statistical analysis nodes were created to show the real-time performance of the algorithms to be tested. Additionally, the testbed also monitors the CPU utilization for measuring the computational complexity. This testing environment will prove pivotal to future localization algorithm development for remote rovers as it provides a dedicated platform for testing multiple localization algorithms at the same time. Using the proposed testbed, the development cycle for remote sensor networks can be greatly reduced.
Abstract ID: 49DCASS-044
AFIT Shadow Imaging Research Overview
Douglas Ruyle
Air Force Institute of Technology
David Curtis
Air Force Institute of Technology
Shadow imaging has been used for decades in astronomical observation to reconstruct the silhouette of distant space objects from the object’s diffraction pattern within the shadow cast on the observer from a star. Synthetic Aperture Silhouette Imaging (SASI) applies this to space domain awareness to enable fine resolution silhouette images of satellites in the geosynchronous (GEO) belt to be collected with a linear array of hobby telescopes. As a satellite passes between a star and the observer on the ground, a North-South telescope array can detect the reduced stellar intensity as the shadow of the satellite passes over from West to East due to the apparent motion of the stars with respect to the satellite. A laboratory model propagates light 35.8 cm to match satellites at GEO according to a scale factor of 10,000 squared in the Fresnel diffraction integral. The silhouette is recovered by processing the resulting diffraction pattern through a Gerchberg-Saxton phase retrieval algorithm.
SASI arrays need to be designed with individual telescopes that are large enough to collect sufficient light to detect a drop in stellar intensity, but also small enough to individually sample smaller ripples in the diffraction pattern to produce a finer aperture-limited resolution. Recent research indicates that a SASI array may be constructed using an 11-inch aperture telescope with up to three spaces between telescopes where each space is the same size as the telescopes. The missing parts of the diffraction pattern can be estimated and the reconstructed silhouette maintains sub-meter resolution at GEO.
The resolution of reconstructed images from SASI arrays also depends on how narrow the spectral band is. Diffraction is wavelength dependent, so each wavelength diffracts around the object’s silhouette differently. A narrower spectral band produces less wavelength-dependent variation in the resulting diffraction pattern. Bandpass filters between 400 nm and 700 nm are added to the laboratory setup to create sharper diffraction patterns. The reconstructed silhouettes from multiple spectral bands are stacked and registered to the laboratory target satellite image and evaluated for how closely the stacked images match the truth image to quantify a multispectral improvement factor.
Future work involves completing an initial field test using a single 11-inch aperture telescope with a photon detector to measure the reduced intensity from a star as a target satellite occults the star. The target satellite of the initial field test will be the International Space Station (ISS). The larger surface area and closer range may help overcome the inherent inaccuracies in the orbital ephemeris data such that even if the main body of the satellite does not occult the target star some of the solar panels may still produce an occultation event.
This research effort will validate software models and better characterize the trade space for a preliminary SASI array design which includes individual telescope size, overall array geometry, and spectral filters. The array design will affect the cost, useable star magnitude, and detection frequency or wait time between occultations) which impacts the overall operational utility of the system.
Abstract ID: 49DCASS-053
The Variation of Radiative Heat Loss as a Function of Position for an Isothermal Square Twist Origami Radiator
Rydge Mulford
University of Dayton
Mohammed Farhan Aziz Najeeb
University of Dayton
Jeremy Price
University of Dayton
David Warburton
University of Dayton
Origami-inspired dynamic spacecraft radiators provide the ability for spacecraft to reject adjustable quantities of heat to account for orbital variations in onboard heating. These radiators might also find utility for Lunar applications, where more extreme temperature variations complicate the rejection of heat from Lunar power plants. The square twist origami tessellation provides large increases in rejecting surface area as the tessellation is actuated from fully closed to fully deployed. However, the variation of radiative heat loss as a function of actuation position is not found in the literature. This study determines the radiative heat loss behavior of an isothermal square twist origami radiator as a function of actuation position. The geometric location of a square twist origami tessellation panels are modeled using vector algebra for an infinitely thin (2D) square twist tessellation as well as a “thick origami” (3D) rigidly-foldable tessellation, both characterized by an adjustable closure angle.
The heat loss characteristics of both the 2D and 3D square twist origami radiators over a 180-degree range of actuation angle have been calculated. Monte Carlo Ray Tracing was used to simulate the radiative losses from the 2D surface over a range of emissivity values (0.1 ,0.5, 1.0). The 3D surface was modeled with a radiosity balance implemented in the ANSYS thermal modeling environment.
Results show a divergence between the 2D and 3D square twist origami radiators. At an emissivity of 1.0, the ratio of escaped to emitted rays exhibits distinct behavior for each tessellation. The 3D square twist origami radiator demonstrates a slower decrease in this ratio as the closure angle increases, culminating at a ratio of 0.4 at an actuation angle of 180 degrees. Conversely, the 2D square twist origami radiator exhibits a linear decline, reaching a ratio of 0.32 at an angle of 180 degrees. Thermal simulations across the range of motion with uniform surface temperatures showcase a turndown ratio (largest to smallest heat transfer) of 4.42. These results substantiate the practicality and efficacy of the rigidly foldable square twist model for radiative heat loss control.
Further analysis extends to derivative geometries, including the "StarTwist," "Sloped-edge Twist," "OctaTwist," and "Mini-core Square Twist." Ray tracer simulations were conducted, adhering to the same range of emissivities. At an emissivity of 1.0, the turndown ratios are 3.32 for the Square Twist, 2.52 for the Mini-core Square Twist, 1.96 for the Sloped-edged twist, 2.78 for the Octa Twist, and 2.88 for the Star Twist. At an emissivity of 0.5, these ratios are 3.27 for the Square twist, 2.50 for the Mini-core Square Twist, 1.95 for the Sloped-edge, 2.76 for the Octa Twist, and 2.87 for the Star Twist. Finally, at an emissivity of 0.1, the turndown ratios are 3.08 for the Square twist, 2.35 for the Mini-core Square Twist, 1.91 for the Sloped-edge twist, 2.67 for the Octa Twist, and 2.78 for the Star twist, demonstrating the variable thermal management efficiencies of these designs.
Abstract ID: 49DCASS-080
Online 6-DoF Spacecraft Control for Multi-Agent Inspection Operations
Mark Mercier
Air Force Institute of Technology
David Curtis
Air Force Institute of Technology
Multi-agent inspection missions are a key enabler for recent advances in on-orbit operations such as servicing, manufacturing, and debris removal. The ability to leverage distributed assets to collaboratively inspect a single target increases the applicability of an inspection to targets that may be in an uncontrolled rotation or have a strict time window in which inspection must take place.
Often optimal trajectory design for complex scenarios, such as a multi-agent inspection, is prohibitively computationally expensive to take place on orbit and must be done offline. Errors in estimated states and thruster noise may introduce deviations from this offline-generated, open-loop control solution. For this reason, an online-capable translational controller is required to track the reference trajectory while still retaining the original objectives of the original optimal control problem.
Additionally, the attitude requirements of a multi-agent inspection mission may necessitate an attitude controller capable of enforcing a desired attitude while avoiding certain constraint zones. By making some assumptions on the inspecting system, the translational and attitude controller can be decoupled and separately developed. This assumption is leveraged in this work to develop an LQR translational controller and an APF-based attitude controller. The attitude controller will enable each inspector to point towards the desired target while avoiding a sun exclusion zone.
In this work, these controllers are used to control multiple spacecraft tasked with inspecting a single target. A reference translation-only trajectory is assumed to be provided from an offline guidance scheme and tracked via a simple LQR controller, while attitude control is handled by an online constrained attitude controller tasked with keeping the RSO in view, while avoiding sensor blinding due to the position of the sun.
Initial results, informed by recent literature, show that a properly tuned LQR controller is not only adequate at tracking the reference trajectory, but also the best choice for maintaining the objective of fuel minimization during online trajectory tracking. The constrained attitude controller is based on a quaternion-feedback control law developed in literature but differs in the occasional “eclipsing” of the RSO in front of the sun so that the desired quaternion falls within the constraint zone. The use of an APF-based attitude controller appears to be a good solution to ensure controller stability despite this occurrence.
Abstract ID: 49DCASS-091
Origami Investigation of Space-Based Mirror System
David Garcia
Air Force Institute of Technology
Robert A. Bettinger
Air Force Institute of Technology
The imaging and inspection of Resident Space Objects (RSOs) is an increasingly important Space Situational Awareness (SSA) mission as space-faring nations and commercial enterprises alike seek to develop means to repair and refuel satellites, as well as de-orbit RSOs in order to reduce orbital debris. The lighting conditions for imaging and inspection are not always advantageous for a repair/refuel satellite; therefore, the use of mirrors deployed from servicer satellites is
proposed to reflect solar energy in order to illuminate dimly lit RSOs. In terms of a general concept of mission operations, the servicer satellites would control the reflected light beam and be positioned to illuminate RSOs for imaging, inspection, repair, and/or refuel. The servicer satellite is assumed to be either a 12U or 27U CubeSat, therefore an investigation of folding the mirror into a compact state utilizing origami will be critical. Specifically, the use of cubic
or rectangular origami flashers will be necessary for CubeSat applications.
Abstract ID: 49DCASS-126
Intelligent Control for Robotic Spacecraft Simulator with Kinematic Redundancy
Anirudh Chhabra
University of Cincinnati
Donghoon Kim
University of Cincinnati
We have developed a twelve degrees-of-freedom (DOF) robotic testbed for space mission emulation at the Intelligent Autonomous Systems Research Laboratory. This testbed is constructed using a six-DOF robotic arm mounted on top of a six-DOF Stewart platform. As the testbed aims to simulate six-DOF spacecraft motion, the redundant DOF can be used to achieve multiple secondary objectives such as avoiding joint limit violations, collisions, and kinematic singularities. These secondary objectives enhance the quality of manipulation. However, to control such a redundant robotic system, proper inverse kinematics (IK) must be defined that can translate the desired manipulator pose into a proper set of joint states that satisfy constraints and achieve the secondary objectives. Traditional closed-loop IK facilitates the design of a closed-loop control for achieving one secondary task but cannot guarantee the achievement of multiple tasks simultaneously. This work focuses on obtaining an IK solution that ensures the achievement of multiple tasks. A fuzzy logic-aided inverse kinematics (FLIK) approach is designed that aims to adaptively manage multiple secondary tasks by actively prioritizing between these tasks in real-time. The performance of the proposed methodology is evaluated using numerical simulations in MATLAB as well as high-fidelity simulations in Gazebo using ROS.
Turbomachinery & Propulsion
Abstract ID: 49DCASS-018
Testing High Work, High Lift Low Pressure Turbine Airfoils in a Transonic Cascade
Ryan Sauder
Wright State University
John Clark
Air Force Research Laboratory
Andy Lethander
Air Force Research Laboratory
Mitch Wolff
Wright State University
A family of low pressure turbine stages was recently developed to investigate the limits of achievable work and lift for future engines applicable to unmanned air systems. Subsequently, mid-span sections of a pair of turbine blades from those stages were manufactured for testing in a transonic cascade facility to verify predicted performance at both on- and off-design conditions. Both turbine blades were designed at the meanline level to a work coefficient of 2.80, and the velocity triangles and pitch-to-chord ratios were consistent with incompressible Zweifel coefficients of 1.60 and 1.78. At design conditions, the airfoils provided 123° of flow turning at an exit isentropic Mach number of 0.78. Both inlet- and exit-total pressure traverses were conducted, and the inlet turbulence intensity and length scale were measured at a pair of locations upstream of the instrumented airfoils in the mid-passage. The experimental datasets presented herein include loading variations as well as loss variations versus Reynolds number. It was found that neither airfoil experienced separation at the lowest Reynolds number achievable at design exit Mach number. Accordingly, it was necessary to operate the cascade at reduced total-to-static pressure ratios to observe significant effects of separation on the loading and loss results. The experimental results presented here are compared against design-level predictions using Reynolds Averaged Navier-Stokes (RANS) codes with transition modeling as well as Large Eddy Simulation (LES). The results are encouraging and bode well for the development of future engines that have reduced part-count, weight, and cost while providing acceptable performance lapse at altitude.
Abstract ID: 49DCASS-027
Design and Evaluation of a Centrifugal Compressor for High Stage-Inlet Temperatures
Caleb Degler
Air Force Institute of Technology
James Rutledge and Frederick Schauer
Air Force Institute of Technology
Awaiting public release.
Abstract ID: 49DCASS-050
A Novel, Microwave-Sustained Air Plasma Propulsion (MSAPP) Concept for Supersonic/Hypersonic Flights
Jincheng Wang
Iowa State University
Hui Hu
Iowa State University
Naibo Jiang
Spectral Energies LLC
Paul Hsu
Spectral Energies LLC
Sukesh Roy
Spectral Energies LLC
Aero-engines that rely on fossil fuels generate significant emissions, widely attributed as a major factor in the emissions observed at high altitudes. Their global impacts and local effects on ground-level air quality have made it a growing environmental concern. In the present study, a novel, microwave-sustained air plasma propulsion (MSAPP) concept is explored to power the atmospheric flights of future supersonic/hypersonic vehicles. The MSAPP system is based on a novel concept of efficiently converting electric energy to airflow kinetic energy via the generation of a high-temperature and high-speed microwave-sustained air plasma jet for thrust generation. While the MSAPP system is a purely electric-based propulsion system without consuming any fossil fuels or propellants, it can generate sufficient thrusts comparable to conventional aero-engines to power supersonic/hypersonic vehicles under various atmospheric conditions with zero carbon emissions. A comprehensive theoretical study is conducted to characterize the energy efficiency and thrust efficiency of the MSAPP system using a Ma = 2.74 supersonic converging-diverging nozzle. A feasibility analysis is also performed to assess the suitability of MSAPP to power an Airbus 320. The analysis also considers important factors such as the weight and dimensions of the MSAPP system, the electric battery requirement, and the duration of in-flight operation using the latest Tesla Powerwall batteries currently available on the market.
Abstract ID: 49DCASS-059
Rotating Detonation Engine Research at AFRL
Robert Fievisohn
Air Force Research Laboratory
Rotating Detonation Engines (RDEs) are being developed and studied at the Air Force Research Laboratory (AFRL) due to their potential benefits over current combustor technology. The detonation wave in an RDE propagates orders of magnitude faster than the deflagration flames in traditional combustors. These greater flame speeds enable reduced size combustors for a variety of applications such as ramjets and afterburners. Additionally, the heat release process in a detonation wave is thermodynamically more efficient than in a deflagration flame. This increase in efficiency manifests itself through an increase in the stagnation pressure across the combustor. Therefore, RDEs are also known as a form of Pressure Gain Combustion (PGC). This presentation gives a high-level overview of the benefits and potential applications that AFRL is pursuing with respect to RDEs. A discussion of the underlying physics and challenges associated with these devices is also included.
Abstract ID: 49DCASS-070
Preparation of Experimental Inlet Swirl Distortion on a High-Speed Compressor
Marcus Acton
Wright State University
Mitch Wolff
Wright State University
Michael List
Air Force Research Laboratory
For the design of more efficient aircraft, engine designers are researching the possibility of ingesting the aircraft’s boundary layer. This turbulent layer introduces distortions into the flow such as swirl distortion which can negatively affect performance. The SAE S-16 committee has proposed a methodology to assess the engine performance loss due to swirl distortion. This work follows the ongoing research into developing this methodology. Bulk and twin swirl patterns have been tested in a vertical wind tunnel to test the functions of a universal swirl generator. The swirl patterns were characterized using the same descriptors as the proposed SAE swirl methodology.
Abstract ID: 49DCASS-075
Fundamental Research in Low Pressure Turbine Aerodynamics at AFRL
Christopher Marks
Air Force Research Laboratory
Molly Donovan
Air Force Research Laboratory
Jared Kerestes
Air Force Research Laboratory
Nathan Fletcher
Air Force Research Laboratory
John Clark
Air Force Research Laboratory
The low-pressure turbine (LPT) is a key component in gas turbine engines as it extracts work from the flow to drive the fan which generates thrust. Modern axial flow LPTs are efficient devices; however, there are opportunities to decrease the component weight, overall engine footprint, and production costs by increasing the coefficients of lift and work. The potential gains are especially important in the age of unmanned aerial systems where this is a need for layered families of air vehicles that require a range of small and medium size propulsion systems. Realizing the benefits of high-lift and high-work technology is not trivial, as complex unsteady flows and low Reynolds number phenomena pose challenges in designing and developing efficient LPTs with higher lift. Furthermore, in smaller engines, the reduced blade aspect ratios mean that the influence of the endwall flows on the full passage performance is greater. This presentation will give an overview of recent fundamental aerodynamic research in high-lift high-work aerodynamics at the Air Force Research Laboratory (AFRL) with an emphasis on detailed experimental studies of the unsteady vortical endwall flows, the onset of separation across the span, and the impact of elevated incoming turbulence. In addition, recent numerical modeling using large eddy simulations (LES) and machine learning (ML) post-processing methods with the potential to accelerate design will be highlighted.
Abstract ID: 49DCASS-085
Adjoint Optimization of Non-axisymmetric endwall Contours to Reduce Losses in a High-Pressure Turbine
Austin Hendrickson
The Ohio State University
Spencer Sperling, Bao Nguyen, Richard Celestina, Jong Liu
Honeywell Aerospace
Hakan Aksoy, Jeremy Nickol
Honeywell Aerospace
Randall Mathison
The Ohio State University
As modern gas turbines continue to evolve, further efficiency gains become more difficult to realize. The parasitic loss from secondary flows become a larger component of the overall loss, making aerodynamic optimization of secondary effects ever more critical. Specifically, the secondary flow losses created by the horseshoe vortex can become a dominant loss factor. There has been much research and development over the past 40 years into the mitigation of these secondary flow structures, but manufacturing advances are opening up the realm of what designs can be practically implemented.
This paper discusses the development of an adjoint optimization method for three-dimensional endwall contouring using an adjoint solver in commercially available software. The primary goals of the study are to establish best practices for a holistic process of endwall optimization and compare the optimized design to the baseline geometry. An array of CFD analyses has been conducted for a high-pressure turbine stage to investigate the impact of the sensitivity radius parameter with regard to isentropic efficiency convergence. The results of these simulations are discussed and the effectiveness of the use of the adjoint model with respect to endwall contouring is discussed. Comparisons between the baseline geometry and optimized endwall shape are used to highlight how small changes in the endwall geometry impact the development of secondary flow structures such as the passage vortex.
Abstract ID: 49DCASS-129
The PATMI Technology: The Path to Rapid Decarbonization of the Energy Sector
Kamal Fernando
Kalindha Rashmi, LLC
Modern thermal power plants based on classical thermodynamic power cycles suffer from an
upper-bound efficiency restriction imposed by the Carnot principle. This work describes how to
break away from the classical thermodynamics paradigm in configuring a thermal power plant
without violating the Second Law. The result, the next generation of low-to-moderate
temperature (300 - 1100 degree C) thermal power plants to operate at significantly higher efficiencies
(50-70%), powered by a variety of fossil and non-fossil fuels and by renewable energy, leading
to low-emissions power generation for the foreseeable future. The proposed new paradigm
eliminates many issues related to conventional inertia-less electronic inverters such as the loss
of spinning reserves in the power grid, thus making the voltage and frequency control of the
main power grid highly economical.
