SHAREOK Repository
Not a member yet
49261 research outputs found
Sort by
Machine Learning Models to Predict Total Skin Factor in Perforated Wells
Get PDFAn accurate total skin factor prediction for an oil well is critical for the evaluation of the inflow performance relationship, and the optimization of the appropriate stimulation treatment such as acidizing and hydraulic fracturing. Performing well testing regularly is not economically feasible, and the equations used for total skin damage may not be accurate. In this work, the goal is to build machine learning (ML) models that can predict the total skin factor in perforated wells using accessible field data. Nine distinct ML algorithms such as Gradient Boosting (GB), Adaptive Boosting (AdaBoost), Random Forest (RF), Support Vector Machines (SVMs), Decision Trees (DT), K-Nearest Neighbor (KNN), Linear Regression (LR), Stochastic Gradient Descent (SGD), and Artificial Neural Network (ANN) are meticulously developed and fine-tuned using a substantial dataset derived from 1,088 wells. The dataset encompasses 19,040 data points, thoughtfully split into two subsets: 70% (13,328 data points) for training the algorithms, and 30% (5,712 data points) for testing their predictions. The parameters used are mostly gathered during well completion and conventional well testing operations, including liquid flow rate, water cut, gas oil ratio, bottomhole flowing pressure, reservoir pressure, reservoir temperature, reservoir permeability, reservoir thickness, perforations diameter, perforations density, perforations penetration depth, well deviation, and penetrated portion of the net pay thickness. In this study, the total skin factor acquired from conventional well test analysis serves as the model's output. K-fold cross-validation and repeated random sampling validation techniques are used to assess the performance of the models against the total skin obtained from the conventional well test analysis. The K-fold cross-validation outcomes of the top-performing ML models, specifically GB, AdaBoost, RF, DT, and KNN, reveal remarkably low mean absolute percentage error values reported as 3.2%, 3.2%, 2.9%, 3.3%, and 3.8%, respectively. Additionally, the correlation coefficients (R2) for these models are notably high, with values of 0.972, 0.968, 0.975, 0.964, and 0.956, respectively. In conclusion, ML models demonstrated their ability to predict total skin factor for different reservoir fluid properties, well geometries, and completion configurations. ML models offer a more efficient, quick, and cost-effective alternative to the conventional well-testing analysis.N
PLANNING FOR CLIMATE CHANGE-INDUCED DISPLACEMENT: SOCIAL INTEGRATION, UNCERTAINTY, DECENTRALIZATION, ADAPTABILITY, AND FAIRNESS
No full textThe looming climate crisis is a significant driver of the displacement of communities. The adverse effects of slow-onset climate change are anticipated to strike people worldwide, causing displacements in significant quantities. These displacements will lead to large-scale movements from high-risk and less resilient areas to safer or more resilient ones, creating a relocation problem: where people should go and when. This is a problem with distinctive characteristics that has received limited attention in the field of operations research. The complexities of the problem, including the need for long-term planning, uncertainties about the future, the involvement of multiple stakeholders, diverse populations, and different locations experiencing varying levels of climate change impacts, call for unique ways. It requires taking up approaches that can assist in developing urgently needed high-level relocation plans to manage climate change-induced movements in a timely manner and with a long-term outlook while using the resources effectively and protecting the peace, well-being, and dignity of displaced people and receiving communities. This dissertation presents a comprehensive proposition for high-level and long-term relocation planning amidst the escalating climate crisis, contributing to the field of humanitarian operations research for societal good. It comprises three studies designed to assist decision-makers in preparing for future actions at the strategic level, even those that may unfold years from now. Each study addresses a challenge associated with optimizing high-level relocation planning in response to climate change-induced forced displacement: future uncertainties, decentralized systems, and ensuring fairness. The first study proposes a two-stage stochastic programming model that optimizes relocation decisions under demand uncertainty with a focus on societal integration outcomes based on diversity indicators. The second study introduces a consensus-driven decentralized optimization framework that balances global utilitarian goals with local interests inherent to relocation planning utilizing a combination of altruistic and self-centered models, negotiations, and a bi-level optimization model that considers culture-based social integration, irregular movements, the associated costs of social conflicts, and fairness among different origins. The third and final study investigates the fairness of destination selection and flow assignment decisions within the context of the high-level relocation problem, providing a comparative analysis of multiple fairness metrics from the perspectives of various stakeholders and considering different principles
FATAL HUMILIATION: THE ROLE OF SHAME IN HOMICIDAL MOTIVATIONS
No full textFew criminological researchers have focused on the role of shame in homicide motives. However, those that have (Anderson 1994, Websdale 2010), have emphasized shame as an important contributing factor to homicides, particularly when there is an interpersonal conflict involved. This study ultimately focuses on the significance of shame as a predictor of expressive homicides and homicides that arise from an interpersonal conflict. It begins with an inter-rater agreement study that empirically assesses coding decisions for the key variables later used in this study: homicide circumstance, instrumental/expressive homicide type, shame, and victim-offender relationship. The results indicated a moderate to substantial level of agreement for the coded variables. In the second part of the study, instrumental and expressive homicides are analyzed with victim-offender relationship as the key independent variable. The results confirmed the centrality of that variable in predicting homicide type and identified victim’s sex, offender’s race/ethnicity, and weapon choice as significant predictors of expressive homicides. Finally, in the third part of the study, I examine the role of shame, as well as victim and offender demographics, victim-offender relationship, and weapon choice in homicide types (instrumental or expressive) and the circumstances surrounding the homicide. The results confirmed the salience of shame in understanding homicide motives. Like in the second part of the study, victim-offender relationship, offender’s race/ethnicity and sex, and weapon choice were also significant predictors of expressive homicides. Additionally, offender race/ethnicity and sex, offender age, and weapon choice were also significant in predicting the circumstances surrounding homicides
Perseveration in Misophonia: A Neuropsychological Approach
No full textMisophonia is categorized by strong, negative reactions to auditory stimuli, often orofacial (e.g., chewing, throat clearing), which may significantly impact daily functioning. Research on how misophonia may affect cognitive functions remains limited, though it shares characteristics with conditions known to influence cognition (e.g., Autism Spectrum Disorder, Obsessive-Compulsive Disorder). Of particular interest, those who have these disorders may display perseverance, the repetition of an inappropriate behavior rather than adapting to new stimuli. Given the similarities between these disorders and evidence of behavioral rigidity in previous studies, perseveration may also be present in misophonia. Perseveration has typically been assessed using the Wisconsin Card Sorting Test, a visual task. The current study tested perseveration and behavioral rigidity through an auditory learning task to target the sensory domain implicated in misophonia. Students from the University of Oklahoma participated by completing a detailed online survey that gauges experiences and severity of misophonia, followed by an audiology workup, then battery of behavioral tasks while Electroencephalography (EEG) data was collected to correlate behavioral outcomes with neural markers of auditory sensory fidelity. While this study did not replicate behavioral rigidity, and neural markers did not correlate with behavioral measures as anticipated, one neural marker showed a relationship with a perseverative measure. Although explicit perseveration was not observed, correlations between perseverative and behavioral measures warrant the need for further investigation. This work may provide insight into the cognitive factors that influence misophonia
MAGNETO DIELECTRIC MATERIALS AND THEIR USE IN THE MINIATURIZATION OF ANTENNAS USED IN AIRBORNE RADAR SYSTEMS
No full textHistorically, the reliance on materials with a relative permeability of one has limited the potential for antenna miniaturization, leaving magneto-dielectric materials underexplored due to their lack of commercial availability. This limitation becomes critical in applications like NASA’s EcoSAR airborne radar, where the large and heavy antenna array, weighing approximately 200 lbs, significantly increases operational costs and limits the advancement of smaller, more efficient aircraft. Addressing this, magneto-dielectric substrates such as MAGTREX 555, with a relative permeability of 6, offer a promising solution by enabling miniaturization and improving antenna performance. This thesis investigates the capabilities of MAGTREX 555 in designing and fabricating a stacked microstrip patch antenna. The study explores the trade-offs between miniaturization, bandwidth, and radiation efficiency. A 6′′ × 6′′ × 0.12′′ antenna was designed to operate at the EcoSAR radar’s target frequency of 435 MHz, achieving a 9.2% bandwidth (40.14 MHz) and cross-polarization isolation greater than 30 dB. While the realized gain was limited to 11.5 dBi due to substrate losses, the design demonstrated a weight reduction of 66.85%, bringing the array weight down to 66.3 lbs from the original 200 lbs. Additionally, the study examines the impact of substrate thickness, coupling coefficients, and ground plane dimensions on antenna performance, showing that increasing the ground plane size from 6′′ × 6′′ to 11′′ × 11′′ improved directivity and realized gain from 3.4 dBi and -6.8 dBi to 5.7 dBi and -1.6 dBi, respectively. The findings illustrate that while MAGTREX facilitates substantial size and weight reductions, its limitations in bandwidth and efficiency warrant further optimization. Nevertheless, this research underscores the potential of magneto-dielectric materials in revolutionizing radar systems for airborne and spaceborne applications, paving the way for lighter, more cost-effective, and versatile designs
THERMOCHEMICAL MODELING, TECHNOECONOMIC OPTIMIZATION, AND CONTROL OF FREEZE DESALINATION PROCESS
No full textFreeze desalination (FD) is a method in which saline water is cooled below its freezing point and freshwater is separated from the brine in the form of ice crystals. FD is relatively insensitive to the salinity of the feed solution, making it suitable for desalination of high concentration brines such as the brine rejected from the seawater desalination plants. The design of the FD system and the thermochemical behavior of the brine upon freezing are critical factors in the energy performance of this method. To date, thermochemical properties of the concentrated seawater during cooling, such as the threshold of formation of ice and salt-hydrates and their corresponding cooling load of formation, are not well known. Likewise, the optimal configuration of the FD system to achieve the maximum energy efficiency has not been investigated. This work provides comprehensive data about the cooling load of freezing of concentrated brine rejected from seawater desalination plants along with the threshold of formation of ice and salt-hydrates backed-up by validation. Furthermore, the optimal configuration of the FD system is identified and the effects of the compressor isentropic efficiency and effectiveness of the system’s heat exchangers on the work consumption of the FD system were investigated. In addition, the lack of knowledge of FD’s economic performances represents a critical research gap that impedes the technology’s progression toward commercialization. To address the issue, this work presents a cost optimization framework for tuning the equipment size and identifying the optimal control parameters at the design stage. Utilizing the developed framework, the outcomes provide benchmark economic metrics for the FD technology considering practical ranges of electricity prices, brine disposal cost, and feed total dissolved solids (TDS). For a feed brine TDS of 75,000 ppm and at a freshwater selling price of 1.5 /bbl, the FD costs /bbl and 0.62 $/bbl respectively. The generated benchmark cost metrics can inform the future market analysis and commercialization of the FD technologies. In addition, this work investigates the potential for enhancing the power flexibility of the FD technology being a thermal desalination method known for its high energy consumption. The study introduces a design modification to the FD configuration, aiming to improve its adaptability to fluctuating power availability. By employing a causal modeling approach, the modified design is analyzed with a focus on minimizing brine treatment costs over a 24-hour period. The effectiveness of this design is evaluated through scenario-based testing, with a particular emphasis on its potential for brine treatment cost savings by participating in day-ahead electricity markets
Old Lady Yells at Cloud: a consideration of AI in libraries
No full textWhat does AI have to do with creativity in libraries? A lot, but maybe not in the ways that you expect. It has become a truism that AI promoters push the ability of AI to do what have been traditionally human creative activity for centuries: writing and visual art. What does this mean for librarians? Wouldn't we rather have AI that is going to shelve all the books or fund a new library classroom? Is AI going to have all the fun while we librarians end up with the drudge work? Perhaps we should look more closely at the promises that AI is making and see what, if anything, is behind them. This presentation will examine some of the excitement around AI and consider if what AI can do for libraries is worth the cost
Maintaining Momentum: How One Regional Public University Library is Staying Afloat
No full textIn the midst of various efforts to become more involved and visible in the University of Arkansas Fort Smith community, the Boreham Library on campus is undergoing three major faculty changes, which account for 50% of our total faculty positions. Despite being short-staffed during the processes of filling these roles, the remaining faculty and staff have maintained the momentum we built during previous semesters through outreach and shared governance efforts. The librarians have kept control of a Student Academic Success course which the previous director advocated for us to teach while also participating in several campus committees, including Faculty Senate. Our exceptional support staff have ensured that we have been able to continue offering the same events and hours of operation for our community during these extra challenges