University of Maryland, Baltimore County
Not a member yet
17643 research outputs found
Sort by
LASER Quench Hardening of Steel: Effects of Superimposed Elastic Pre-Stress on the Hardness and Residual Stress Distribution
No full textCold drawn AISI 4140 beams were LASER surface hardened with a 2 kW CO2 LASER. Specimens were treated in the free state and while restrained in a bending fixture inducing surface tensile stresses of 94 and 230 MPa. Knoop hardness indentation was used to evaluate the through thickness hardness distribution, and a layer removal methodology was used to evaluate the residual stress distribution. Results showed the maximum surface hardness attained was not affected by pre-stress during hardening, and ranged from 513 to 676 kg/mm2. The depth of effective hardening varied at different magnitudes of pre-stress, but did not vary proportionately to the pre-stress. The surface residual stress, coinciding with the maximum compressive residual stress, increased as pre-stress was increased, from 1040 MPa for the nominally treated specimens to 1270 MPa for specimens pre-stressed to 230 MPa. The maximum tensile residual stress observed in the specimens decreased from 1060 MPa in the nominally treated specimens to 760 MPa for specimens pre-stressed to 230 MPa. Similarly, thickness of the compressive residual stress region increased and the depth at which maximum tensile residual stress occurred increased as the pre-stress during treatment was increased Overall, application of tensile elastic pre-stress during LASER hardening is beneficial to the development of compressive residual stress in AISI 4140, with minimal impact to the hardness attained from the treatment. The newly developed approach for LASER hardening may support efforts to increase both the wear and fatigue resistance of parts made from hardenable steels
Damage and Crack Detection Methods Based on the Vibrational Characteristics of Damaged and Cracked Cantilever Beams
No full textDamage detection methods utilizing the vibration characteristics of a component or structure have been emerged over the last few decades as viable methods in assessing and monitoring their structural health. Such methods often utilize physics based models that are capable of predicting the frequency and modal response of the structure in the presence of localized damage including cracks. They are often easy to use and do not necessarily require access to the full structure since they may rely on local component measurements and response. This work aims at advancing damage detection methods through the development of a new family of such physics based models. More specifically, beam model theories capable of predicting the modal characteristics of beam structures are revisited for the purpose of establishing their range of validity when compared to 2D Finite Elements. The governing equations of a vibrating beam are presented within the Euler-Bernoulli as well as Timoshenko assumptions. The requisite eigenvaTRAClue problem is then formulated and solved. For comparison purposes, a requisite 2D finite element model as well as one dimensional beam finite element models are developed through which the frequencies and related mode shapes, as well as their slope and curvature profiles are obtained. Broad parametric studies comparing the Euler-Bernoulli, Timoshenko and finite element predictions are presented and validity envelopes for the beam theories are established. The Euler-Bernoulli, Timoshenko and 2D Finite Element beam models are further developed to simulate the presence of a localized damage zone of reduced stiffness within a cantilever beam. The models allow for the damage zone to be placed at a prescribed location. A modulus of elasticity reduction profile is introduced simulating a gradient damage characteristic within the prescribed damage zone. The results suggest that evidence of the presence, extent and degree of damage can be seen in the predicted frequency changes as well as in the localized changes exhibited by all mode shapes. More pronounced evidence of the existence of damage is exhibited by the mode shape slope and particular by the mode shape curvature profiles. Thus, appreciable changes in mode shape curvatures relative to their healthy counterparts are predicted offering a viable alternative to detecting the presence, extent and severity of localized dispersed damage in beam and most likely other structures. In further expanding the available family of models for damage detection, a 2D finite element model was developed aimed at studying the modal response and related mechanical characteristics of a cantilever beam with a horizontal sharp crack. Broad parametric studies are reported exploring the effects of beam aspect ratio, crack center location and crack length on the frequency and modal characteristics of such cracked cantilever structures. Similarities in the predictions between models are discussed while also establishing correlations between the discrete crack geometry and the profile characteristics of the stiffness reduction zones employed by previous models. The spatial and temporal response similar cracked beams were also studied via the method of finite elements. The finite element methodology used is developed along with the numerical finite difference integration scheme employed in solving the time dependent boundary and initial value problem. An impulse force and time dependent sinusoidal forcing functions at the upper right corner of the beam are used to model the time varying applied loading. As in the previous models, a broad range of beam length to height ratios ranging from 5 to 50 has been considered. Time profiles of selected beam displacement are presented and compared to known beam solutions. The presence of the crack is shown to cause a noticeable time shift in the beam deflection periodical response. Discrete Fourier Transform profile of the beam's tip displacement and related frequencies also show evidence of the presence of the crack. The simulation results obtained by the above new generation of models suggest that the presence of local zone of reduced elastic stiffness or the presence of a crack induced a shift in the transverse beam frequencies most likely due to the increase beam compliance induced by the damage zone or the crack. In fact, beams containing cracks of larger length were shown to vibrate at lower frequencies when compared to those with shorter cracks and of course when compared to a healthy structure. In addition, distinct deviations from an otherwise smooth profile exhibited by the healthy beam were predicted in the damage or crack regions along the beam's free surface for the mode shape slope and curvatures with the latter exhibiting much higher sensitivity to the presence of damage or crack when compared to the slopes. The frequency reduction induced by the damage or a crack has also been validated through Discrete Fourier Transform displacement profiles which showed that larger frequency reduction is predicted for beams containing larger cracks. The above findings establish the foundation for the development of new novel model based and non-model methods for damage detection
The Mechanics of an Elastically Deforming Cantilever Beam with an Embedded Sharp Crack and Subjected to an End Transverse Loading
No full textOver the last two decades, the frequency response of a component or a structure has been used to assess structural health as one of several methods used in damage detection and structural health monitoring. In studies involving simply supported or cantilever beam components, the natural frequencies and mode shapes for linearly vibrating beams have been established using approximate Euler-Bernoulli or the shear-enhanced Timoshenko beam theories. The latter theory accounts for shear effects that become important in low aspect ratio beams as well in calculating high frequencies in otherwise slender beams. In most studies, the presence of damage is modeled as a localized reduction in the structural stiffness EI of the beam. While the Timoshenko theory allows for the independent reduction of the elastic stiffness E and beam cross-sectional properties such as the cross sectional area A or the second moment of inertia I, both the Euler and Timoshenko theories assumed a symmetric reduction of the beam properties leading to a symmetric reduction of the effective beam stiffness EI. As such, current techniques do not have the refinement needed to detect along with the location of damage, its shape and its orientation. In this study, both analytical and numerical methods are employed in developing robust models for damage detection. The study utilizes a cantilever beam with a fully embedded sharp crack and subjected to an end applied transverse force as the binary geometry for model development. Initially, a broad parametric finite element meshing algorithm capable of generating model geometries with fully embedded cracks at prescribed location, length and orientation is developed. The above algorithm is employed in conducting broad parametric studies of beams with a fully embedded sharp crack. The simulations helped to establish the load transfer and deformation mechanics of the cracked beam. In addition, for each model considered, the near-tip fracture mechanics are established revealing critical issues related to fracture mode mixity, stress intensity and crack surface closure effects. In parallel studies, analytical methods are employed in establishing for the first time simple but robust Mechanics of Materials models capable of predicting the mechanics of a cantilever beam containing a sharp horizontal crack. Model predictions on the dominant force and moment resultants in the crack region as well as on the extent of the transition zones in the near-tip region have been obtained and compared to their counterparts obtained numerically via the method of finite elements. A remarkable agreement was found between the beam model and finite element predictions. The beam model outcomes also enabled the development of analytical solutions for the complex near-tip mixed mode fracture fields. The latter models utilize the force and moment resultants in the crack region to obtain analytical expressions for the linear elastic energy release rate at each of the crack tips using both the compliance and the J-integral methods. Arguments regarding the structure of the transition fields and induced force and moment discontinuities at the crack tip cross sections led to unique separation of the mode mixity thus revealing the dominant fracture mode II nature of the beams and loading considered. Initially revealed by the numerical finite element models and then validated by the analytical beam model, surface curvatures of the cracked beam emerged as a viable predictor of crack damage. More specifically, it has been shown both numerically and analytically, that a measurable deviation from a smooth profile occurs in the free surface curvatures in beams containing a fully embedded sharp crack. Intriguing features on the curvature profiles provide critical clues on the location and extent of the crack. Such findings provide compelling evidence that non-model efficient methods can be developed in detecting damage. Crack surface closure effects are examined to improve the linear elastics structural response. Their effects on the load transfer and deformation mechanics of damage detection are investigated
Do Medicaid family planning waivers reduce low birth weight prevalence? A nationwide, state-level analysis
No full textToo many infants in the United States are delivered with unhealthy weights. Low birth weight is a noteworthy public health problem because it is a major risk factor for infant mortality. It consumes a disproportionately high level of health care resources and its prevalence continues to rise despite investment of public resources aimed toward improving birth outcomes. One solution has been to ensure satisfactory maternal health by preventing unintended pregnancy and promoting optimal birth spacing. Federal and state governments have responded to the need for managing the timing of conception using public policy to assure access to family planning services, particularly for low-income residents who might not otherwise be able to access or afford them. Policy implementation at the state level has relied on the Medicaid program to finance and deliver contraceptive medical care to populations who would not ordinarily be eligible for basic program enrollment, using Section 1115 family planning waivers. A policy question relevant to the problem of low birth weight is whether states with Section 1115 family planning waivers in effect exhibit significantly better infant birth weight rates as a result. This study evaluated the extent to which these rates varied in response to implementation of waiver policies, while controlling for inter-state differences in population demographic composition, labor force factors, and other non-medical determinants of health. Public use data from the U.S. Natality files and the Current Population Survey were used to construct a panel data set including observations for 50 states and the District of Columbia for each year from 1990 through 2010. Multivariate linear regression models including state and year fixed-effects consistently produced ordinary least squares (OLS) coefficients on the policy intervention variable that were statistically insignificant and substantively small in all but non-Hispanic Black populations. Estimated effects observed for this sub-group suggest that implementation of Section 1115 family planning waivers is associated with significantly higher low birth weight rates, an unanticipated outcome. Results produced by this study failed to support the research hypothesis that Medicaid family planning waivers are effective in lowering the prevalence of infants that weigh 2,500 g or less at birth, when measured as a percent of total live births. Waivers may nevertheless be functioning to reduce the number of infants at healthy and unhealthy weights alike, for all populations studied except non-Hispanic Black. It is critical to note that although low birth weight rates were significantly higher for non-Hispanic Black populations following policy implementation, it seems unlikely that absolute prevalence increased. These results are more likely a measurement artifact related to an overall decline in birth rates combined with possible changes in composition of the full birth weight distribution. Nevertheless, ensuring access to contraceptive medical care appears to be an effective approach to mitigate low birth weight prevalence by reducing the number of these births, but policy alternatives other than subsidized contraceptive medical care ought to be considered for reducing the proportion of total births that are under 2,500 g
EXAMINING THE ROLE OF THE COMPTON-THICK, X-RAY REPROCESSOR IN TYPE 1 ACTIVE GALACTIC NUCLEI
No full textStudies have indicated that black holes and their host galaxies must co-evolve, although the mechanism linking the two is not yet clear. X-ray observations of the actively accreting systems (active galactic nuclei, hereafter AGNs) over a valuable probe of conditions in the inner environs of the supermassive black hole, as X-ray production within these systems comprises a significant fraction (5%-40%) of the bolometric luminosity and originates close to the nucleus. Detailed spectroscopy in this bandpass has established that the inner environs comprise multiple X-ray absorbing zones with column densities extending into the Compton-thick regime (NH > 1024 atoms cm-2). Compton-thick absorbers are known to have outflowing velocities up to 0.3c. The kinetic energy of oufltowing material with velocities > 0.1c may possibly be comparable to the gravitational binding energy of the stellar bulge and serve as a link between the black hole and the host galaxy. In addition, weak outflows may have a signicant effect on star formation in the host galaxy. In this study, we use X-ray observations of radio-quiet AGN to examine some aspects of the flow of material between the black hole and the host galaxy. First, we model a small sample of unabsorbed Seyfert galaxies, finding their X-ray spectra to be consistent with arising as reflection from tens to hundreds of rg in a Compton-thick, accretion disk wind of solar abundances seen face-on. Then, we explore properties of the local AGN population in the very hard band, above 10 keV, where there has been scant data available to date. We finnd a high flux for local AGN above 10 keV, and thus a very hard spectral shape over the Suzaku bandpass. Taken together, the spectral hardness and equivalent width of Fe K emission are consistent with reprocessing by an ensemble of Compton-thick clouds that partially cover the continuum source. Simple considerations place the distribution of Compton-thick clouds at or within 1017 cm. This work demonstrates not only that a Compton-thick wind can have a profound effect on the observed X-ray spectrum of an AGN, even when the system is not viewed through the flow, but that at least 50% of the continuum in type 1 AGNs is partially-covered by Compton-thick gas, suggesting that type 1 AGNs may not offer a completely direct view of the primary continuum, as once thought
Exploring variations of pairwise coverage for prioritizing web application test cases
No full textRegression testing is a critical phase in the software development lifecycle. One of the challenges during regression testing, called regression test selection, is to identify and organize the test cases to execute, in order to evaluate the correctness of the modified software system. Test case prioritization is one of the test selection methods that aim at ordering test cases in order to find faults quickly. The ordering is obtained by subjecting test cases to criteria that can identify test cases that are likely to locate faults early in the test execution cycle. We focus on event-driven systems in this work, and in particular web applications. In prior work, several criteria are developed for event-driven systems including, pairwise coverage of parameters, frequency of windows coverage, and count-based criteria. These criteria are black-box criteria, in that they do not use the source code when prioritizing test cases. Instead, they use information about the order and frequency in which pages are accessed to prioritize test cases, where a test case is itself a sequence of pages that can be visited on the web application along with any associated data, also known as parameter-;values. The previously proposed pairwise coverage of parameters criterion, called 2-way, selects a next test case that covers the most uncovered 2-way parameter-value interactions. However, the 2-way criterion considers interactions between parameter-values across pages in a test case. The 2-way criterion also does not consider the sequence in which pages are accessed. This thesis proposes and evaluates two main hypotheses: (1) Test cases that are ordered in descending order based on the number of parameter-value interactions between consecutive pages they contain are likely to have high rate of fault detection. (2) Test cases that are ordered in descending order based on the number parameter-value interactions within a web page they contain are likely to have high rate of fault detection. Our work thus aims to explore two variations of the 2-way criteria. With the first criterion, 2-way intra, our aim is to expose faults that may occur due to parameter-value interactions within the same page in a web application. On the other hand, the 2-way-inter-consecutive criterion exposes faults based on pairwise interactions between parameter values on consecutive pages. We first implemented the criteria in CPUT, a Java-based automated tool that converts usage logs to XML based test suites, prioritizes test cases and creates test orders that was developed in prior work. To evaluate the effectiveness of these criteria, first we gathered user sessions of two open source web applications, Schoolmate and OSCommerce. Using CPUT, the test case orderings for the two criteria were obtained. We then manually seeded faults in the applications. The ordered test cases were then run against the fault seeded versions of the applications. The struct, struct++, diff, path and content oracles were used for comparing clean & faulty outputs. Based on the comparison, the rate of fault detection and Average Percentage of Faults Detected (APFD) were calculated. We report and discuss the results of comparison of these metrics with that of the pre-existing criteria
Genocide: What Genocide?
No full textSince the end of the Cold War, millions have been killed in a series of genocides. The dismantlement of the Soviet Union left the United States with new decisions to make regarding human rights violations. The United States government no longer had hard interests to intervene in cases of human rights violations. I show through a comparative study of the genocides in Bosnia, Rwanda, and Darfur that it takes political pressure for the United States to intervene: it takes multiple calls for action from members of Congress and the American people for the United States to step in and end the violence. I used congressional meeting minutes and hearings to show the political pressure and calls for U.S. intervention. In addition to the media that congressional representatives use in their calls for action, I looked at New York Times and Washington Post articles to show the media attention given to the acts of genocide. To study the response to the genocides, I examined the patterns that emerged in the US response to them. To emphasize the comparisons between the three cases I show how they all follow the same pattern of U.S. intervention, consisting of four phases. The pattern is distinct in all three acts of genocid
Iounn
No full textThe aim of this thesis is to explore the relationship that the Internet, social media, and consumer loyalty have on teenagers' self-perception today in contrast to earlier generations. For two years I mined Apple Store display computers for photographs that visitors left behind. These photographs tell a story about teenagers being forced to look at themselves through the lens of technology, the Internet, and online peers. I also mined YouTube for what is known as Apple Store Dances, and explored the physical and psychological relationship teenagers have with themselves in the mirroring process of dancing before computer screens in public. I discuss these two image sets in relation to mythology, sociology, postmodernism, the Apple Store's architecture and design, and the grid. Finding that I often reverted back to and indulged in being a teenager myself in this process, the project is both personalized and reflects an artist-researcher's analysis
Examining development induced geomorphic change through time using multi-temporal LiDAR-derived Digital Elevation Models
No full textDuring development, topography is cut, filled, and graded, altering natural surface drainage patterns. The spatiotemporal characteristics of drainage network and hillslope changes throughout the development process are to date unknown. For this study we used multiple temporal LiDAR-derived DEMs spanning the development of two small agricultural watersheds to examine geomorphic changes caused by urban development. A third forested watershed was also examined to track geomorphic changes not attributable to development. We first present a technique to extract drainage channel heads and drainage networks through time utilizing surface tangential curvature and compare its ability to predict field verified channel heads against two other delineation techniques. Second, subwatersheds of the delineated networks were used to summarize topographic changes in the forested and developing watersheds through time. Finally, simplified hydrologic response functions are generated for each time step utilizing width functions augmented by the curvature delineated networks. Landscape dissection and hydrologicconnectivity increased throughout development, indicating an increased efficiency of water removal from the landscape. Substantial topographic changes manifest as variations in curvature standard deviations through time in first order catchments, highlighting redistribution of high and low gradient regions. Idealized hydrologic responses reflect increased landscape connectivity with a distinct shift toward more rapid storm response shapes throughout the development process. Unexpectedly high temporal variance observed in the forested watershed across analyses indicates inherent differences in temporal digital topographic datasets. Results of this thesis highlight fundamental geomorphic changes caused by development practices while developing methods and rationale for temporal DEM-based geomorphic studies