Environmental and Occupational Health Sciences Institute
Rutgers University Community RepositoryNot a member yet
58345 research outputs found
Sort by
This Woman’s Work: Arbeitswelten in der ,Frauenliteratur‘ der Zwischenkriegszeit
No full textDuring the interwar period, both the production and reception of ‘female writing’ underwentsignificant changes. So-called ‘women's literature’—texts written by and for women—received
unprecedented publicity, with publishing houses, particularly Ullstein, launching large, multimedia
campaigns to promote their female authors. This dissertation explores how changes in work practices and the publishing industry afterthe First World War encouraged female authors to reflect on the connections between care, career,
and happiness. Unlike the relatively well-researched literature that focuses on female employees,
this thesis examines four texts featuring protagonists with different forms of employment: a nurse,
a teacher, female doctors, and a scientist. The works analyzed include Vicki Baum's successful
novel stud. chem. Helene Willfüer (1928), Hilde Maria Kraus' Ärztinnen (1929)—a novel highly
regarded at the time of its publication—and Neun Monate (1931), as well as Rahel Sanzara's Die
glückliche Hand (first published in the Vossische Zeitung in 1933 and posthumously in 1936).
These texts explore social expectations and personal fulfillment through the depiction of work
while simultaneously reflecting on the conditions under which women could write and express
themselves. The incompatibility of professional and private life is portrayed and examined differentlywithin each narrative. While Vicki Baum suggests a possible happy ending for her protagonist,
Hilde Maria Kraus presents the compatibility of work and private life as questionable. Rahel
Sanzara, however, ultimately rejects the notion of female happiness, thereby challenging the
compatibility of motherhood, family, love, and career. I will argue that the challenges of ‘female writing’ become evident through the depictionof writing scenes and letters within these narratives. Unlike avant-garde approaches to exploring
the form, the selected texts exhibit formal breaks as they attempt to portray the relationship
between professional life and care. The emphasis on the active hands of the protagonists and
their respective spatial environments demonstrates the authors’ exploration of a new form of
writing and literary creation.Ph.D.Includes bibliographical reference
Genetic analysis and molecular breeding tools for improving summer patch tolerance in hard fescue
No full textHard fescue (Festuca brevipila Tracey) is a cool-season turfgrass species that is commonly categorized within the fine fescue group. This species is well-known for its excellent performance under low-maintenance conditions, such as landscapes with limited water, pesticides, and fertilizer inputs. However, the popularity of hard fescue is limited by its susceptibility to summer patch disease. Summer patch is a root disease caused by Magnaporthiopsis poae and Magnaporthiopsis meyeri-festucae. This highlights the necessity of breeding hard fescue cultivars with improved tolerance to this disease.The purpose of this thesis was to gain a better understanding of the inheritance of summer patch tolerance in hard fescue and to explore the application of QTL mapping for summer patch tolerance in hard fescue. The ultimate goal was to improve the efficiency and efficacy of selection for summer patch tolerance in hard fescue. The specific research objectives of this thesis were divided into three chapters, as follows: (1) to determine the heritabilities of summer patch disease tolerance in populations derived from a diallel crossing design between tolerant and susceptible hard fescue clones, (2) to construct a high-density, single nucleotide polymorphism-based (SNP-based) genetic linkage map for hard fescue, and (3) to identify quantitative trait loci (QTL) associated with summer patch tolerance in hard fescue.
For the first objective, full-sib progeny populations were created by crossing three tolerant and three susceptible parents in a diallel cross. One hundred progeny from each of the 15 crosses and reciprocals were established in mowed spaced-plant trials in 2017 and 2019. The disease severity of hard fescue clones was assessed by visual rating during the summers of 2018, 2019, 2020, and 2021. The observed variation in disease responses among the progeny suggests that inheritance is controlled by a few major genes. The estimate of narrow-sense heritability was 0.20 (± 0.01), while the estimate of broad-sense heritability was 0.67 (± 0.08). These heritability estimates are modest but still suggest that selection and breeding schemes could be useful for improving summer patch tolerance in hard fescue.
For the second objective, full-sib progeny populations were created through crosses of T10 (♀) X S5 (♂) and S5 (♀) X T10 (♂). A total of 178 progeny, with 91 from the T10 (♀) X S5 (♂) population and 87 from the S5 (♀) X T10 (♂) population, along with their parental genotypes, were sequenced to call SNP markers. After filtering out SNPs shared by less than 80% of the progeny in each case, 3876 SNPs for the T10 background dataset and 4432 SNPs for the S5 background dataset were obtained. Two linkage maps with 21 linkage groups (LGs) each, one for T10 and another for S5, were independently constructed. The T10 linkage map comprised 1020 SNPs, spanning a total of 1981.1 cM, while the S5 linkage map contained 897 SNPs, covering a total of 1644.8 cM.
For the third objective, the T10 (♀) X S5 (♂) and S5 (♀) X T10 (♂) populations were visually evaluated for the percentage of green present in each individual in 2018, 2019, and 2020 at Rutgers University Horticultural Farm II located in New Brunswick, NJ. QTL analysis was then applied to identify the association between the markers in the linkage map and summer patch tolerance. Nine QTLs were identified with significant LOD scores exceeding the genome-wide LOD thresholds. These QTLs were distributed across three regions on three linkage groups and explained phenotypic variations ranging from 4.1% to 5.3% for summer patch stress.
These are the first studies to report heritability estimates for summer patch tolerance in any turfgrass species, including hard fescue, and to report the first QTL mapping of summer patch tolerance in hard fescue. The findings of these studies will aid in determining the most effective and efficient selection procedures for summer patch tolerance in hard fescue. The successful application of the techniques used in this study on hard fescue will also be valuable for future studies investigating heritabilities and marker-assisted selection in closely related plant species. Last but not least, the development of the T10 X S5 populations and the high-density linkage map will serve as valuable resources for upcoming studies on hard fescue for other traits of interest.M.S.Includes bibliographical reference
Trends in polychlorinated biphenyl concentrations in Upper Hudson River Water
No full textThis thesis explores the congener patterns of toxic polychlorinated biphenyls (PCBs) measured in the water of the Hudson River from 2017 to 2021. The Hudson River has been declared a Superfund site due to contamination with PCBs from two General Electric (GE) capacitor manufacturing plants located on the river near Fort Edward, New York. From 2009 to 2015, large areas of the Upper Hudson River (UHR) were dredged to remove PCBs residing in the sediment. Since that time, routine (monthly or weekly) monitoring of PCBs in the water column has been conducted at several sites downstream. These samples were analyzed using EPA method 1668C, which quantifies all 209 PCB congeners. In this work, this data set of hundreds of samples was analyzed using Positive Matrix Factorization (PMF), an unsupervised machine learning algorithm, to determine the congener patterns or fingerprints that contribute to the measured PCBs. The PMF analysis isolated eight factors, seven of which are related to GE, representing PCB formulations used by GE or their biodegradation products. One factor was higher in MW than the others and contained congeners that are characteristic of PCB formulations that were never purchased by GE for use in the UHR. This factor is more abundant downstream near the more urbanized portion of the UHR, and therefore represents PCBs from non-GE sources such as stormwater, treated wastewater discharges, and discharges from Combined Sewer Outfalls (CSOs). This non-GE fingerprint explains 4% of the PCB mass detected in water samples from the entire UHR, but more (7%) near Albany, just above the Federal Dam at Troy where water from the UHR flows into the Lower Hudson River and the New York/New Jersey Harbor. Although this non-GE source explains a relatively small proportion of the PCB mass at Albany, it contains higher MW PCBs and thus bioaccumulates to a greater extent than the lower MW GE-associated PCBs. It may therefore be problematic in both the Upper and Lower Hudson River.M.S.Includes bibliographical reference
Protective strategies within relationship social comparison contexts
No full textAlthough people often compare their romantic relationships to others’, little is known about the defensive strategies people employ to protect their relationships from adverse comparisons. Across two studies, I extended well-established protective processes in the self-threat literature to relationship threats, testing whether relationship partners would engage in devaluation, discounting, and/or bolstering after a relationship social comparison threat. In Study 1, I tested for the emergence of these protective strategies under relationship social comparison threat and explored whether such processes serve to buffer against reductions in relationship satisfaction. In Study 2, I tested whether devaluation, discounting, and/or bolstering buffer against downturns in affective, behavioral, and cognitive components of relationship satisfaction. To test these questions, participants in romantic relationships received bogus feedback suggesting they measured better (or worse) than another couple on a specific relationship dimension. Then, they answered questions about how much they valued the comparison dimension, how much they believed the comparison feedback, how well their relationship was performing, and items assessing (components of) relationship satisfaction. Consistent with my predictions, romantic partners who experienced threats to their relationship (i.e., who received upward relationship social comparison information), were more likely to devalue the comparison dimension and discount the feedback. However, the protective strategies did not influence romantic partners’ overall relationship satisfaction (Study 1). Instead, in Study 2, I found that the protective strategies were linked to specific components of relationship satisfaction, including negative behaviors and negative cognitions. Additionally, I found that two strategies--discounting and bolstering--were especially likely to emerge among people more committed to their partners. This work offers important insights into how romantic partners navigate relationship social comparison threats and contributes to the broader literature on relationship maintenance strategies.Ph.D.Includes bibliographical reference
Practical methods for fair and explainable decision making
No full textAlgorithmic decision making is used to make a wide variety of important decisions in today’s society, from bail decisions to school matchings. There is an increasing recognition of the importance of understanding exactly how algorithms make their decisions. This dissertation aims at bridging the gap between algorithm designers and stakeholders by providing metrics and tools to better understand the behavior of decision processes. Bridging this gap has the potential to empower stakeholders to make more informed decisions about how they interact with the algorithms and increase their trust in these systems that are increasingly critical to the running of society. For regular people to be full participants in algorithmic decision making, the algorithms chosen must be explainable and transparent so that no expert knowledge is required to understand their outcomes. The users of decision systems must also feel the algorithms that drastically affect the course of their lives are fair and reasonable. To further these explainability and fairness goals, this dissertation proposes explainable algorithms and metrics to promote decisions that are clearly fair and reasonable to the people they most affect. The first stage of algorithmic decision making is when decision makers choose the decision model to be used as an input to the algorithmic decision process; in many scenarios, decisions are based on ranking functions that aim at ordering participants to the decision. This dissertation provides explainable metrics to help the designers understand the diversity, fairness, and imputed weights of the developing ranking function. These metrics do not require expert knowledge to understand what they are measuring and how they are measuring it. By providing explainable metrics, decision makers can more easily trust the claims the metrics are making about the data. Specifically, this dissertation proposes transparent participation metrics to clarify the ranking process, by assessing the contribution of each parameter used in the ranking function in the creation of the final ranked outcome, using information about the ranking functions themselves, as well as observations of the underlying distributions of the parameter values involved in the ranking. In order to evaluate the outcome of the ranking process, diversity and disparity metrics are proposed to measure how similar the selected objects are to each other, and to the underlying data distribution. Metrics are evaluated on synthetic data, as well as on two real-world scenarios: high school admissions and decathlon scoring. Once explainable metrics are used in the design process, decision makers may want to automatically optimize the metrics without compromising the explainability of the outcome of the decision process. One such explainable metric is the disparity, which is closely related to the popular statistical parity measure of fairness. Optimizing this metric usually means minimizing the disparity between the population and the selected set. Traditionally, this would be done by minimizing disparity through an opaque transformation of the ranking functions or a set-aside system, in academic proposals about how to rank with fairness constraints and real-world school matching systems respectively. The same minimization can be done with bonus points without substantially increasing the complexity of the ranking function. Fairness goals are often added on top of an existing ranking function. By mirroring the language in which these ranking of functions are usually described, the fairness goals can be easily combined with the existing ranking function without reducing explainability. The conversion of the fairness goals into a specific number of bonus points is provided by a novel algorithm named the Disparity Compensation Algorithm(DCA). DCA uses a sampling-based approach modeled after Stochastic Gradient Descent to satisfy fairness goals orders of magnitude faster than competing algorithms and allows algorithm designers to accomplish fairness goals without compromising the explainability of the overall ranking function. The need for fairness and explainability can continue past the main part of the algorithmic decision process to the follow-up decisions that need to be made to respond to potential errors. Once an error has been detected, the outcome may need to be adjusted to avoid unfairly disadvantaging any participants due to the error. Adjusting the outcome to repair errors after the algorithm has already been completed needs a stronger explanation than simply running the algorithm as originally planned with no errors involved. Without this explanation, it can be difficult to convince both the harmed and unharmed participants that they are being treated fairly. A real-world example of this challenge arises in applications of deferred-acceptance (DA) matching algorithms, where errors or changes to the matching inputs are sometimes discovered only after the algorithm has been run and the results are announced to participants. Mitigating the effects of errors is a different technical problem than the original match since the decision-makers are often constrained by the decisions already announced. This dissertation proposes models for this new problem, along with explainable mitigation strategies to go with these models. Three different error scenarios are explored: In the first scenario, one of the participants is removed after the match is complete, leading the group matched to that participant unmatched. In the second scenario, some participants are unfairly disregarded and not ranked by a participant on the other side. In the last scenario, some participants are unfairly promoted over others. For each error type, the expected number of participants directly harmed, or helped, by the error, the number indirectly harmed or helped, and the number of participants with justified envy due to the errors are calculated. Error mitigation strategies must then arbitrate between minimizing the extra resources needed to mitigate the error and ensuring that no participants are unnecessarily harmed by the error. This dissertation provides algorithms, metrics, and models for fairer and more explainable algorithmic decision making. In particular, by providing metrics to help design ranking functions, explainable optimization, and sound error mitigation, this dissertation will allow for greater stakeholder participation in algorithmic decisions. As algorithms become increasingly cryptic it is essential to remember that key decision algorithms are designed to help humans make decisions at scale. If the algorithms are not fair and explainable to the humans they represent, they will not be successful no matter how effective they are. When explainability and fairness are centered, the algorithms can become natural extensions of the humans making the decisions.Ph.D.Includes bibliographical reference
Modeling, control, and evaluation of aircraft tire-runway interactions
No full textAlthough landing takes only 1% of the total flight time in commercial air transportation, 31% of fatal accidents happened during the landing stage for worldwide commercial jet fleet during 2013-2022. Aircraft landing tire-runway interactions plays a critical role for safe operation of aircraft, particularly under severe weather conditions, such as rain, snow, slush, etc. Understanding and quantification of tire-runway friction under different environmental conditions are among the most important tasks for aircraft takeoff and landing performance assessment. Research studies underscored the critical role of maintaining adequate tire-runway interaction friction forces, as it directly influenced accidents such as overruns by impacting braking distances. This significance is particularly pronounced when runways are wet, with data from various regions indicating that approximately 20% of fatal accidents occur in such conditions. The goal of this dissertation is to provide a modeling, analysis and control of aircraft tire-runway interactions.After reviewing the modeling and analysis literature, the second chapter of this dissertation presents tire-runway friction estimation under wet pavement conditions and hydrodynamics analysis for aircraft tire-runway interactions. Understanding wet tire-runway interactions is critical for predicting unstable operating conditions such as hydroplaning and for safety operation. We present an analytical method to compute water film pressure and thickness distributions on the viscous hydrodynamic region between the tire and the i ground. The hydrodynamic models and analysis are obtained for the smooth tire surface and tires with transverse tread elements. We present analytical conditions to predict the hydroplaning occurrence. The impact of wet condition on tire-runway friction forces is also discussed through the effective length of the contact patch and a LuGre friction model. Simulation results are presented to demonstrate the modeling development. Comparisons with other existing models illustrate the advantage of the proposed models and analysis.
Implementing grooves on aircraft runways has been proven effective in maintaining skid resistance under wet conditions. However, due to the inherent complexities involving hydrodynamics and the interaction between the tire and grooved pavement, the modelling of the friction force in such conditions has been rarely studied. In Chapter 3, we present a method to evaluate tire/runway friction under wet conditions and on grooved pavement sur faces. This evaluation involves examining the interaction force generated between the tire and grooved pavement, and studying of the drainage capacity of the grooves. Comparisons between the model-predicted friction coefficients and results from indoor testbed experiments are provided to demonstrate the feasibility of the proposed model and its potential for incorporation into aircraft braking systems to improve their performance.
Embedded sensors inside the tire rubber layer potentially enables attractive features to estimate and obtain the real-time tire-runway interactions information. The fourth chapter of this dissertation presents a new tire-runway force sensing and estimation scheme with embedded flexible sensors. A physics-based tread beam model is developed and integrated with the LuGre friction model to evaluate the rubber deformation as well as the contact pressure distribution. A sensor model is then proposed to build the relationship between the measured longitudinal stress of the tire tread rubber and the external friction force at the tire-runway contact patch. Feature points on measurement curves are analyzed and used to estimate tire runway friction forces. A tire friction testbed is developed for comprehensive experimental validation under conditions such as various slip, surface roughness, etc. Experimental results demonstrate the feasibility of using the force-sensitive sensors ii for predicting tire/road friction characteristics.
Compared with passenger vehicles, aircraft tire-runway friction measurements face different characteristics and additional challenges. Because of heavy weight, large size and high speed, it is more challenging and expensive to conduct full-size tire-runway friction tests for commercial aircraft than passenger vehicles. In the fifth chapter, we present a novel scaled testbed for studying aircraft tire-runway interactions during landing and braking stage. The indoor testbed is built on a platform with a rotational arm to support an aircraft landing wheel on a circular, re-configurable runway track. By designing the braking torque, the rotating arm motion, and the pneumatic normal load, the testbed possesses the capability to emulate the dynamic characteristics of aircraft tire-runway interactions and braking maneuvers under various pavement conditions. Besides the mechatronic design of the testbed, we present the modeling and control of the platform subsystems to demonstrate the capability and performance for simulation of various braking maneuvers under different pavement and wet conditions. The dimensionless analysis and design are presented to represent the equivalent simulation of braking process of full-size aircraft. Experimental results demonstrate the efficacy of the testbed design in studying the interactions between the aircraft tire and the runway pavement.
In the last chapter of the dissertation, we summarize the concluding remarks of the research progress and also present the ongoing research directions. We mainly focus on the friction modeling and analysis of the aircraft landing tire on different pavement groove patterns under dry and wet conditions. Using the developed scaled testbed, we plan to validate the modeling and analysis experimentally and further develop a real-time control of the tire-runway interactions on different groove patterns and wet conditions.Ph.D.Includes bibliographical reference
Subtle shifts: how the news shapes perceptions of deadly police-civilian encounters
No full textMy research examines how issues of racialized policing are discussed in partisan news sources and how the information provided in news articles has the power to impact reader sentiment regarding deadly confrontations between police and civilians. The impetus for this research was an incident in May 2020, which was broadly covered in the media and sparked a national reckoning on racial justice. On May 25, 2020, George Floyd, a 46-year old black man was killed after a Minneapolis police officer kneeled on his neck for over nine minutes. Black men being killed by police is not a novel development, however, the resulting news coverage and public response made this incident unique. Partisan news outlets have subtle differences in how they cover fatal police-civilian interactions. Over time, repeated exposure to these differences can cause readers to develop increasingly polarized opinions about the same incident (Klein, 2020; Cummings, 2022). For my study, I used a mixed-methods approach, conducting a qualitative content analysis for the first part and conducting a quantitative survey experiment for the second part. I began with a content analysis of 42 news articles of the George Floyd incident from four different news sources: Vox, CNN, Fox, and Breitbart. The Pew Research Center (2020) indicates that the former two sources appeal to those who lean politically left and the latter two sources appeal to those who lean politically right. As compared to right-leaning sources, left-leaning sources tended to provide more extensive coverage about the Floyd-Chauvin incident. The key findings from the content analysis that informed the survey experiment were: 1) there was a lack of victim-humanizing details about Floyd, though the literature indicates that humanizing details can impact readers’ perceptions of favorability and empathy (Li & Shi, 2022) and 2) there was a lack of articles that included information on the prior complaints against Chauvin, and the literature indicates that misconduct history can change perceptions of police legitimacy and trust (Cruz, 2015). Guided by my qualitative findings, I designed a survey experiment with a 2x2 fully-crossed design with four vignettes describing a fatal interaction with a police officer. The manipulations for the vignettes were: one control vignette that simply described the interaction, one vignette that added humanizing details about the victim, one vignette that included officer misconduct history, and one that included both the victim-humanizing details and officer misconduct history. After a randomly assigned vignette was displayed, respondents (n=1,337) received survey questions assessing their opinions about the interaction (i.e. who was to blame, whether either party should have reacted differently, how appropriate the police response was, and recommended disciplinary action). Key findings from the survey experiment were that the inclusion of humanizing details about the victim and the officer’s misconduct history increased support for the victim and increased blame for the officer. When combined, the effect was most powerful - having access to both the victim-humanizing details and officer misconduct history in the same vignette resulted in the least blame attributed to the victim and the most blame attributed to the officer. Demographic details including gender, race, political ideology, and prior experience with unfair treatment by police significantly changed how respondents answered the survey questions, including how they apportioned blame to the victim and the officer. Key implications from my findings highlight the importance of providing relevant, comprehensive information in news articles, establishing a practice of transparency around police misconduct, and engaging the public in open discourse about media literacy and implicit biases.Ph.D.Includes bibliographical reference
Below the surface of Blackness: Negrura and communality in the Spanish-, French-, And English-Speaking Caribbean
No full textThis dissertation begins with the premise that relationality is key to defining blackness. In various forms of literature from across the Caribbean, characters define their blackness based on their relationships with themselves and others within their texts' world. In addition, their definitions of blackness are directly influenced by their imagined genealogical relationship(s) with characters from other texts that existed before and/or after them. This project is about how black and mixed-race female characters are able to see the other, “touch the other, feel each other,” (Fanon), and how they are transformed by their interpersonal (and often peripheral) relationships. My dissertation consists of four chapters in which I conduct discursive analysis to explore definitions of blackness as Negrura (a concept originally used by Lydia Cabrera which I redefine). The journey begins in Cuba and Puerto Rico (Chapter 1) through the analysis of Alejandro Tapia y Rivera's play "La cuarterona" (1867) and Cuban novelist Cirilo Villaverde's Cecilia Valdés (1839-1882). Chapter 2 travels to the Dominican Republic and Haiti (chapter 2) in the exploration of the characters in “La danza de Mingo” (The Dance of Mingo), a Dominican play written by Haffe Serulle (1947) in 1977, and Edwidge Danticat's novel The Farming of Bones (1998) focused on the 1937 massacre of Haitians and Dominicans of Haitian descent in the Dominican Republic. The Anglophone Caribbean connects to these narratives in my examination of Jamaica (chapter 3) through Michelle Cliff's Abeng (1995), and the U.S. Virgin Islands (chapter 4) in the writing of Tiphanie Yanique in Land of Love and Drowning (2014). In my comparativist study of the African Diaspora in the Americas, I use Diaspora Studies, Gender Studies, Latin American Studies, critical race theory, and Decolonial thought to engage concepts of relationality (Glissant), Blackness, touch, and spirituality in Caribbean literature.Ph.D.Includes bibliographical reference
Activities, affect, labor: a study of fans’ participation in döjin and stage exploration in Japan
No full textBased on interviews and participant observations, this study examines how fan activities can be understood through the concepts of labor and affect, by focusing on two activities in Japan: döjin activity (the production of books and other goods) and stage exploration (visits to places used as settings in media content). The concept of labor situates fan activities within the context of contemporary capitalism, where they can be harnessed for the production of surplus value. At the same time, the concept also brings attention to what fans exactly produce, which cannot be entirely subsumed under capital. Adding the concept of affect illuminates how bodies act upon one another, and can account for what specifically moves fans to labor for what they love and are passionate about, as well as how fans can be mobilized to labor in capitalism.My findings demonstrate that fans are moved by: the immediate atmospheres of the spaces in which they conduct their activities; the transmedia entities of narrative-worlds and characters, which are the “objects” of their fandom and consumption; and the notion of hobby – more specifically, the values it encapsulates – within capitalism, where waged work and other forms of “productive” activities are given centrality.
An atmosphere of a space acts upon fans who reportedly absorb or breathe it in as “air”; in the process, fans become inspired and elated, or even cleansed of negative affects. In turn, fans’ presence and activities (e.g., walking, waiting), in conjunction with the things around them, are elements that shape the atmosphere. Narrative-worlds and characters are “incomplete” entities filled with “blanks” and “gaps” that induce fans to exercise imaginative and creative labors to materialize them in various forms. Döjin activity and stage exploration help build narrative-worlds and characters, which are owned as intellectual property, demonstrating a specific way through which media corporations can move fans to do the labor of consumption that can be captured as sources of value.
By exercising concrete labors in döjin activity and stage exploration, fans also develop ideas on what matters and what warrants labor, brought together under their conception of “hobby.” This notion enables fans to value various concrete labors in terms of their specific affective experiences (both feelings and capacity to act). It thus undercuts the pursuit of capital predicated on a measurement of various concrete labors vis-à-vis one another in terms of the capital they produce through indifference to their specificities. Furthermore, the notion of “hobby” can be deployed by fans to decenter work from their lives, which is part of a larger response to the capitalist imperative for activities to be productive. These fans affirm the value of a concrete labor even if it does not produce capital. Fan activities, informed by the notion of hobby, thus point towards the possibilities for contesting capitalist logics.Ph.D.Includes bibliographical reference
Nondestructive evaluation of damage in polyethylene pipes using ultrasonic testing
No full textIn pipelines for natural gas distribution, more than 50% of damage is caused by third-party interference, such as foundation loads, traffic loads, or rock impingement. The remainder of damage occurs due to the limitations of the pipe’s material in its service environment, such as resistance to slow crack growth (SCG), rapid crack propagation (RCP), thermal-oxidative aging, and stress-corrosion cracking (SCC). Due to the cost of excavating, user inconvenience, and required expertise, it is difficult to detect the onset of crack nucleation and propagation using conventional methods such as routine visual inspections or the hydrostatic pressure test. Therefore, effective nondestructive evaluation (NDE) techniques are sought so that cracks may be efficiently detected long before they compromise the integrity and functionality of a pipe system. However, due to high levels of material damping in plastic media, ultrasonic guided wave testing (UGWT) on plastic pipes is significantly unexplored compared to steel pipes. This study aims to develop a systemic guideline of using piezoelectric (PZT) transducers for actuating and sensing of ultrasonic guided waves (USGW) on polyethylene (PE) pipes to locate external or internal cracks and assess their severity. A series of tests are performed on PE pipe specimens, in which PZT arrays are configured in varying pitch-catch arrangements on the outer surface. UGWT is conducted to determine material properties in the pipe’s pristine state. Damage is subsequently introduced using high RPM cutting tools, and stress-waves are transmitted. Responses at multiple sensors are examined for each actuated pulse, and degree of signal strength decay is recorded. Subsequently, the pipe is numerically simulated, and input properties are fine-tuned until response signals match the results of the experimental tests. The conditions of the waveguide are controllably varied to incorporate crack damage of different lengths, depths, thicknesses, axial and circumferential locations, and orientations. As such, a synthetic database of signal response data is compiled and expanded. This data is analyzed for damage-sensitive features, which are subsequently used to develop damage detection algorithms, such as training and testing a machine learning (ML) model. To validate the detection analysis, the developed algorithms are run over sensed signal responses from the experiments. This database is used to develop damage classification algorithms, such as support vector machine (SVM) and convolutional neural network (CNN). These models are further updated with independent numerical and experimental cases. The results proved the ability of UGWT for NDE of PE pipe by sequentially locating cracks and quantifying the crack geometry.Ph.D.Includes bibliographical reference