The Graduate Research Showcase is comprised of an elite group of graduate students who will present their research orally or with a poster.
Save this date for our 2023 competition and check back soon for more information.
Scroll down to see 2022 winners anounced!
To participate in the 2022 Graduate Poster Session, Submit this form by April 6th.
Students will display their posters and present for judges on Tuesday, April 12 from 10am-1pm in the 2nd Floor Havener Atrium. Students will be judged on their poster display, how well they present to the judges, and how well they interact and answer judges questions.
Dress for the event is business attire.
The top 3 contestants will be announced the week of April 18th.
Participants (who are not CDF students) will be responsible for printing their own poster. We suggest using S&T Printing Services.
Questions? Email Erica Reven at email@example.com.
Daniel Bowerman, Veronica Lee, Matthew Luebbe, and Hailey Swain
Mine Haul Roads: Data Visualization of Maintenance Condition
Abstract: In the last few decades, larger and larger tucks have been introduced straining the capabilities of current mine haul road construction. Large weights, constant traffic and harsh weather conditions lead to constant maintenance. Currently, visual inspection is the primary method to identify areas to perform road maintenance. This is subjective, often inaccurate, and delays corrective action. Predictive modeling of road condition will allow optimization of road maintenance resources, resulting in lower total mining cost. Vibration level, an indicator of road condition, along with GPS and other telemetry data is analyzed to find patterns in haul trucks’ vibration response. Vibration is found to be higher and with a greater variance at higher operating speeds. Different sections of road exhibit significantly different vibrations levels and while haul trucks overall have similar levels of vibration, individual road defects can influence trucks differently. These findings will be used to develop a predictive road maintenance model.
Performance Assessment of GFRP Rebars Used in an Existing Bridge
Abstract: Since the use of steel reinforcement is related to corrosion issues, other alternatives needed to take place to mitigate the corrosion problems. One of these alternatives is glass fiber (GFRP). It presents itself as a strong candidate to replace steel reinforcement due to not only its high resistivity to corrosion, but also its economic efficiency. The goal of this study is to assess the performance of GFRP bars implanted in two bridges and exposed to a real-time weather environment for over ten years. The bridges are Southview Bridge in Missouri and Sierrita de la Cruz Creek in Texas. Cores were extracted from each bridge containing some parts of GFRP reinforcement. Several tests were conducted on the GFRP bars and surrounding concrete. These tests include: scanning electron microscopy, energy dispersive spectroscopy, short bar shear, fiber content, pH, chloride content, and carbonation depth. Regarding the results, scanning electron microscopy showed no existence of microstructural degradation in the glass fiber and interfacial transition zone. Energy dispersive spectroscopy did not show any changes in the GFRP’s chemical composition. However, short bar shear showed some changes from control bars. The outcome of this research provides new evidences that can be used to enhance GFRP durability data in civil engineering codes and standards.
Recyclability of 304L Stainless Steel in the Selective Laser Melting Process
Selective laser melting (SLM) is an additive manufacturing (AM) process that uses a laser to bond layers of powder for creation of three-dimensional components. During part fabrication, a large amount of energy is induced into the melt pool causing laser spatter and condensate to be generated, both of which have the potential to settle in the surrounding powder-bed compromising its reusability. In this study, 304L stainless steel powder is subjected to five reuses in the SLM process to assess its recyclability through both powder and mechanical property characterization. All powder was characterized by particle size distribution and shape measurements, oxygen content with combustion analysis, and phase identification by x-ray diffraction. The evolution of powder properties with reuse was also correlated to tensile and Charpy impact behavior. The results show that reused powder coarsens and accrues more oxygen with each reuse causing no change in the tensile properties but a decrease in the toughness.
Microscale Residual Stress Measurement in Rolled Materials
Residual stress exists within a material without an applied external force. Depending on its direction, it can improve or deteriorate a material's mechanical performance. A microscale residual stress measurement technique, micro-slotting, was developed at Missouri S&T and has been applied to rolled steel bars. Micro-slotting makes use of a focused ion beam (FIB) system, digital image correlation (DIC), and finite element analysis (FEA) to evaluate residual stresses with measured relief strains. Limitations to the technique that were not previously considered have been studied as well.
Tangential and Angular Joint Estimation Using Polynomial Chaos
Abstract: Polynomial chaos expresses a probability density function (pdf) as a linear combination of basis polynomials. If the density and basis polynomials are over the same field, any set of basis polynomials can describe the pdf; however, the most logical choice of polynomials is the family that is orthogonal with respect to the pdf. This problem has been studied for the joint estimation of states that are purely translational or angular, and for the independent estimation of mixed states of translational and angular random variables. It is proposed that real valued polynomials that are orthogonal with respect to measures on the unit circle can be used as basis polynomials in a chaos expansion, which would reduce the additional numerical burden imposed by complex valued polynomials.
Numerical Study of Paramagnetic Elliptical Microparticles in Curved Channels and Uniform Magnetic Fields
Abstract: By utilizing direct numerical simulations, we numerically investigated the dynamics of a paramagnetic elliptical particle immersed in a low Reynolds number Poiseuille flow in a curved channel and under a uniform magnetic field. Based on an arbitrary Lagrangian-Eulerian approach, a finite element method analyzed how a single curved channel and the strength and direction of a magnetic field can affect the rotation and the net migration of the particle. In the absence of a magnetic field, the rotation is symmetric, but the particle-wall distance remains the same after one periodic rotation. In the presence of a magnetic field, the particle’s symmetric rotation is broken, and the particle radially migrates further away from the channel wall as the magnetic field strength increases. The consequence of the radial migration is due to the particle angular velocity caused by the magnetic torque that constantly changes directions as the particle is transporting inside the curved channel. We compared the particle-wall distances for a particle in a curved channel and a straight channel. This research benefits non-spherical particle studies for industrial and biological applications.
In-Situ Radiometric Inspection of Laser Powder Bed Fusion
Abstract: Laser Powder Bed Fusion (LPBF) is an Additive Manufacturing (AM) process that enables the manufacturing of complex geometry parts with high resolution. A major challenge in LPBF is the qualification of parts that may have inadequate engineering properties due to intrinsic events. Part qualification is typically performed post process adding time and expense to the manufacturing process. The layer-to-layer nature of AM permits the opportunity for in-process monitoring of the thermal history experienced by parts, which is a critical factor that controls properties. This poster presents a radiometric inspection framework for in-situ qualification of LPBF 304L stainless steel microstructure on a local basis.
Special Case of the Thanksgiving Lemma Conjecture
Abstract: The Finite Congruence Lattice Representation Conjecture posits that every finite lattice is isomorphic to the congruence lattice of some finite algebra. The Thanksgiving Lemma Conjecture is an important step in our attempt to prove the Finite Congruence Lattice Representation Conjecture. The Thanksgiving Lemma Conjecture states that given a lattice of height 2 with exactly n atoms and a finite transitive G-set minimal-representer, then if x and y are in the principal congruence identifying u and v, where u is not v, then u and v are identified by the principal congruence identifying x and y. We prove the Thanksgiving Lemma Conjecture for the specific case in which the minimal-representer G-set consists of the two-dimensional vector space over the n^2-element field, whose fundamental operations are affine invertible functions. These affine invertible functions are polynomials of degree 1 with coefficients in the n^2-element field. In the case where the minimal “witnessing chain” is of length 3, we can assume that this chain consists of a single fundamental operation application and a single application of transitivity to the identification of u and v. Given these assumptions, applying a single fundamental operation to the identification of x and y provides a “witnessing chain.”
Numerical Determination of Spacecraft Formations in Deep Space
Abstract: The work this poster describes identifies continuous natural relative trajectories near the collinear libration points numerically from the nonlinear circular restricted three-body problem (CR3BP) differential equations of motion by utilizing a shooting method with a two-level differential corrector. The focus is on determining natural formations in the vicinity of the Earth-Moon L2 point when using the nonlinear equations of motion of the CR3BP. Upon identifying these relative trajectories analysis is conducted to determine drift rates of the uncontrolled trajectories compared to linearized formations near the collinear libration points. In addition, stationkeeping techniques are applied to determine the required maneuvers and frequencies and to assess propellant costs to maintain the nominal trajectories.
On Stackelberg Signaling and Its Impact on Receiver's Trust in Personalized Recommender Systems
Recommender systems have relied on many intelligent technologies (e.g. machine learning) which have procured credibility issues due to several concerns ranging from lack of privacy and accountability, biases and their inherent design complexity. Given this lack of understanding of how recommender systems work, users strategically interact with such systems via accepting any information with a grain of salt. Furthermore, the recommender system evaluates choices based on a different utilitarian framework, which can be fundamentally different from the user's rationality. Therefore, in this paper, we model such an interaction between the recommender system and a human user as a Stackelberg signaling game, where both the agents are modeled as expected-utility maximizers with non-identical prior beliefs about the choice rewards. We compute the equilibrium strategies at both the system and the user, and investigate conditions under which (i) the recommender system reveals manipulated information, and (ii) trust regarding the recommender system deteriorates when the true rewards are realized at the user.
Human-AI Cooperation in StarCraft II
Abstract: The world is becoming increasingly driven by technological advances. In the midst of this, the interactions between humans and various forms of Artificial Intelligence (AI) is growing drastically. Whether its your car, or phone, or even your coffee pot, its important to optimize these AI to the needs of the humans interacting with them. In this study, we looked at several different AI agents cooperating with humans, and asked humans how they perceived each of the AIs. We then took this feedback, and compared it with the performance in the assigned tasks, to see if performance was the key factor human perception of AI.
Temperature Based Identification of Thermal Properties of a One Dimensional Transient Convection Model of a Slender Cylindrical Fin
Abstract: This study investigates an approach to determine the thermal conductivity of a material from temperature data. Due to the simple geometry and small amount of material needed, this method can be applied to additively manufactured materials on a batch by batch basis. For this study, a one-dimensional transient heat diffusion PDE with a closed-form solution is used to model the slender test coupons. The inhomogeneous boundary condition is handled using eigenfunction expansion and Green's formula. A modified Levenberg-Marquardt nonlinear least squares algorithm is used to determine the thermal conductivity from the experimental data using a flux boundary condition. The results from the experimental data compared well with published values for the materials tested.
Effect of Turbulence Model Uncertainty on Scramjet Strut Injector Flow Field Analysis
Abstract: The effect of turbulence model closure coefficient uncertainty on the Reynolds-averaged Navier-Stokes solution of a scramjet strut fuel injector flow field is investigated with an uncertainty quantification and sensitivity analysis study for the Menter-BSL and Wilcox-2006 k-ω turbulence models. Simulations were performed using the VULCAN flow solver. Nonintrusive polynomial chaos theory was used for efficient propagation of uncertainty, and Sobol indices were utilized to quantify the sensitivity of various solution metrics to the variation of each closure coefficient. The output metrics of interest were evaluated at three crossflow planes, and included the integrated quantities of mixing efficiency, circulation, total pressure recovery, and one-dimensional Mach number, as well as the pointwise vorticity and eddy viscosity distributions. Influential sets of closure coefficients were identified for each turbulence model, with the Karman and diffusion constants being the most prominent. These results will assist future efforts aimed at reducing the uncertainty in the numerical design of scramjet fuel injectors through the identification of closure coefficients and physical aspects of the flow that warrant further investigation
A Co-free Precipitation-Hardened High-Entropy Alloy
Abstract: High-entropy alloys (HEAs) have been shown to have high strength, thermal stability, irradiation resistance, and other desirable properties. However, most HEAs studied are single-phase and contain Co, limiting their strength and making them unsafe for nuclear applications. Our group has developed a Co-free high-entropy alloy with L12 (or y') precipitation-hardening potential. Characterization with scanning electron microscopy (SEM), transmission electron microscopy (TEM), and atom probe tomography (APT) shows the formation of a B2+Chi phase network within the FCC+L12 matrix. This structure shows impressive strength, mostly from the B2+Chi precipitates, but may show limited ductility due to those precipitates. This alloy will provide a good base for composition variation to improve other properties such as corrosion resistance.
Securing Autonomous Vehicle Interactions with Blockchains
Abstract: This poster presents the key ideas behind the author's proposed solution to securing vehicular ad-hoc networks (VANETs) of autonomous vehicles using blockchains. Blockchains are a tool used to reach consensus in an untrusted distributed system. This solution produces significant monetary savings over prior solutions, including the Security Credential Management System which was created by the United States Department of Transportation (USDOT). The key difference is that the author's proposed solution leverages a peer-to-peer system, producing a cost saving of approximately $2.5 trillion initially and $121.5 billion per year in maintenance compare to the system proposed by USDOT.
A Chirp Pulse Fourier Transform Microwave Spectrometer with Multi-Antenna Detection (MAD-CP-FTMW)
Abstract: In 2019, our group showed that it is possible to detect microwave radiation in a chirp pulsed Fourier transform microwave (CP-FTMW) experiment that does not copropagate with the incidence of radiation. It was determined that this could be achieved with the broadcast antenna by the implementation of a circulator, switch, and low noise amplifier in order to also detect off of the same antenna. This discovery provided both the preliminary data and the impetus for a new hypothesis to be tested: Could free induction decays (FIDs) be detected at other points not in the direct linear path of the microwaves? To address this, we utilized our new microwave three-wave mixing (M3WM) instrument based upon the four-antennae design of Pate in order to utilize the orthogonality of the setup for the purposes of CP-FTMW FID collection. The design of the new four-antennae MAD-CP-FTMW at Missouri S&T will be presented along with the exciting results of detection in traditional, 180 degree, and quadrature angles.
Particle Comminution Using Waterjet Based Mill: Coupled CFD-DEM Study
Abstract: The high energy inefficiency of Waterjet based mill has been the focus of much of modern-day comminution research, and this current research attempts to understand the dynamic energy events taking place inside the mill to improve the efficiency of the mill. The current study aims to develop a mechanistic model that accurately captures the process of fluid accelerating particles and predict the crushing behavior of particles inside the mill by means of numerical simulation. The model is developed by coupling computational fluid dynamics (CFD), which resolve the fluid flow field (liquid and air) with discrete element model (DEM) which solves for particle motion (solids).The most distinguished things in the current work are as follows: (a) the particle/fluid, particle/particle and particle/wall interactions which are characterized by the CFD-DEM coupling method. In this poster, only the results of Validated CFD model will be presented. A detailed discussion on the selection of suitable CFD turbulence model for the accurate prediction of flow field inside waterjets cutting head is made by comparing the predictions among and against experimental data. The findings of the CFD model highlight the importance of the choice of turbulence models in CFD solution and its effects on the final coupled CFD-DEM simulations.
Cyclic dynamics of misfires and partial burns in a dilute spark-ignition engine
Abstract: Cyclic dynamics of cumulative heat release data past the edge of stability in a dilute spark-ignition engine are investigated. Increased dilution either from excess oxidizer or combustion products causes combustion to become more strained, and incomplete combustion events (i.e. misfires and partial burns) start to occur. These incomplete combustion events cause combustion instabilities and create a dilute limit in spark-ignition (SI) engines. These events are often followed by a higher-energy cycle due to the feed-forward mechanism present in the residual gases. These patterns are deterministic and increase the coefficient of variation to undesirable levels. Symbol sequence analysis was used to investigate the cyclic dynamics of these low-high patterns. The heat release was partitioned on an energy basis to give physical meaning to each partition and each sequence created when analyzing the symbol sequence results. This partitioning method provided insight into the differences in the dynamics when operating in the misfire or partial burn regime. These differences could impact the control method used.
Improving the Efficiency of Nondominated Sorting Genetic Algorithms
Abstract: Genetic algorithms using nondominated sorting as a selection mechanism have emerged as some of the premier methods to handle multiobjective optimization problems. While these algorithms have shown the ability to converge to a well-distributed set of Pareto efficient solutions, they are constrained by the computational complexity of the nondominated sorting algorithms they employ. To combat this limitation, considerable effort has been expended looking to improve the efficiency of nondominated sorting procedures. This work proposes and demonstrates the efficacy of an ideal-point-based inferred dominance algorithm, shown to improve upon the state-of-the-art in a limited investigation.
POD Approximations and Errors
Abstract: When using proper orthogonal decomposition (POD) as a model order reduction technique it is of particular interest to know how these approximations behave. We present refinements and extensions of our earlier results concerning POD projections and approximation errors. We focus on new, exact error formulas for POD approximations with a generalized framework. An example is given to illustrate the new error bounds.
Mixed-Modality Learning for Lifelong Machine Learning (M2L for L2M)
Abstract: All modern AI algorithms and advancements fall under the umbrella of Machine Learning. A new framework called Lifelong Machine Learning (L2M) is the next paradigm of research in pursuit of general artificial intelligence. This poster provides a definition of the paradigm and preliminary theoretical research.
Investigating Asphaltene Stability in Crude Oil During Carbon Dioxide Injection
Abstract: Hydrocarbons are almost always though of as being fluids, either liquid or gas. This is the main reason why they can be mobilized from the reservoirs and formations that may exceed 10,000 ft underground in several cases. One component of hydrocarbons that is not a liquid however is asphaltene. Asphaltene is a solid hydrocarbon component that is stabilized in the crude oil via resins or as a nano-colloid. At some conditions, this component may begin to separate from solution and deposit in the formation pores, perforations, wellbore, surface facilities, and pipelines. This can result in severe safety and production hazards include plugging of all the aforementioned, and pressure buildups which may result in blowouts. This research studies the impact of different factors on asphaltene stability in the crude oil and determines the conditions at which asphaltene is most likely to deposit. Also, the research attempts to mitigate asphaltene damage my using multiple organic solvents to dissolve the asphaltene. By undergoing this research, a better understanding of the stability of asphaltene and how to mitigate its deposition can be achieved. This in turn can help resolve or alleviate the damage caused by this solid hydrocarbon.
Necessary Conditions for Hypergraph Transitivity
Abstract: A hypergraph is a generalization of a graph whose edges can contain an arbitrary number of vertices. A hypergraph is said to be vertex-transitive if its vertices are indistinguishable, in the sense that any vertex can be sent to any other vertex by a permutation that preserves the structure of the hypergraph. A hypergraph is said to be edge-transitive or flag-transitive if the hypergraph's edges or flags are indistinguishable in the same way, where a flag is a pair of a vertex with an edge that contains it. The goal of this research is to find necessary conditions that a hypergraph (V,E) must satisfy to be vertex-, edge-, or flag-transitive. This is done by generating additional hypergraphs which are also preserved by the structure-preserving permutations of (V,E), and describing how the vertices, edges, or flags of the hypergraph must also be indistinguishable with respect to these additional hypergraphs.