Browsing by Subject "Model"
Now showing 1 - 20 of 27
- Results Per Page
- Sort Options
Item Advancing a nuclear magnetic resonance force microscopy (NMRFM) probe and simulating NMRFM in thin films(2023-12) Paster, Jeremy W.; Markert, John T.; Lai, Keji; Lewis-Peacock, Jarrod A; Tsoi, Maxim; Marder, MichaelThis research endeavor centers around the development of a nuclear magnetic resonance force microscopy (NMRFM) probe for investigating thin-film samples. Of particular interest is the conducting region that forms when lanthanum aluminate (LAO) is grown epitaxially on strontium titanate (STO). These materials are insulating in their bulk form. We propose NMRFM as a tool to detect whether there is diffusion of atoms across the interface, which could explain the emergence of the conducting region. While conventional scanning probe techniques are constrained to the surface of a sample, NMRFM features the non-invasive and subsurface detection capabilities of conventional nuclear magnetic resonance (NMR) spectroscopy. Unlike conventional NMR, for which a cubic millimeter-sized sample is required to produce a measurable signal, we can readily scale down NMRFM detection sensitivity, extending its application to smaller samples. In combination, these features suggest that NMRFM is well-suited to study the LAO/STO interface. Force detection of nuclear spins is made possible by coupling NMR spin flip sequences to a mechanical oscillator (cantilever). A small magnetic tip deposited on the cantilever establishes a large field gradient and an interaction force with the magnetic moments of the sample nuclei. This tip traces out constant-field slices perpendicular to the magnetic field. Within a particular slice, nuclear spins resonate with the perturbing oscillating field conventionally employed in NMR spectroscopy. We anticipate that evidence of atomic diffusion across the LAO/STO interface is limited to a 10-nanometer region. Before the reconfiguration outlined herein, this NMRFM probe previously resolved a sample whose smallest dimension was 60 microns. To develop this probe for thin films, we adopted the Interrupted OScillating Cyclic Adiabatic Reversal (iOSCAR) protocol. This method distinguishes the minuscule force interaction between the cantilever and the sample by implementing an NMR-induced modulation of the cantilever frequency. iOSCAR generates a distinguishable signal at an established frequency that is far from spurious artifact signals that limited the signal-to-noise ratio of the previous NMRFM protocol. This dissertation involves contextualizing NMR and NMRFM and assesses the need for further experimental investigation of LAO/STO. Furthermore, it details the evolution of an NMRFM probe to enable the exploration of thin-film samples using iOSCAR. While this research project largely involved the creation of experimental components, it concluded by modeling the expected experimental results. We created simulations of thin-film NMRFM, calculating the z component of the sample magnetization near an oscillating cantilever with a magnetic tip. These simulations explore the dynamic interactions between a thin-film sample and a cantilever as sample nuclei undergo magnetic resonance.Item Application of dynamic optimization methods for foam floods in stratified reservoirs(2018-08-17) Tang, Brandon Chok-Yie; Nguyen, Quoc P.Efficient recovery of oil from heavily stratified carbonate reservoirs can be very technically challenging, even when applying waterflood, gasflood, or WAG (water-alternating gas) processes. To date, relatively few field or pilot applications of foam flooding have been conducted due to an incomplete understanding of how foam will behave in the field. The reservoir of interest studied in this work is oil-wet and consists of a stratified upper high-permeability zone overlaying a lower low-permeability zone. This study seeks to assess the performance of the foam flooding process in oil recovery and develop an optimum field injection strategy based upon various objective functions. In the process, the impact of initial waterflooding and varying foam strength on the optimum project termination time, as well as the sensitivity of foam parameters on the optimum field injection strategy is investigated. Two main optimization techniques are tested: static optimization, where the injection parameters are set once at the beginning of the simulation, and dynamic optimization, where injection parameters are optimized in five-year intervals over the life of the well. The dynamic optimization was performed in two ways: a local dynamic optimization and an early-time weighted optimization. In general, the dynamic optimization outperformed the static optimization with respect to all objective functions. Over the course of the study, a variety of objective functions were utilized. The objective functions began with maximizing cumulative oil recovery and evolved to maximizing oil recovery while minimizing gas utilization ratio, and finally maximizing net present value (NPV). From the results, it was ultimately shown that the global dynamic optimization of NPV was the most useful way of obtaining a field injection strategy. The optimal process design parameters indicated that high volumes of surfactant as well as gas in the lower zone needed to be injected early in the life of the project to best maximize NPV. From the optimal termination time study, it was found that the optimal termination time for the project was around ten years. Varying extents of initial waterflooding and alteration of foam strength did not have an impact on the suggested termination time. From the foam strength sensitivity, it was found that among the factors (water saturation, oil saturation, surfactant concentration) considered, the maximum dry-out water saturation had the most profound impact on the NPV. Ultimately, this work develops the framework necessary to create a field injection strategy for foam flooding in the stratified oil-wet reservoir used in this study, but can be extended to other types of reservoirs.Item Comparison of models for numerical simulation of low salinity waterflood(2021-08-12) Santra, Ritabrata; Sepehrnoori, Kamy, 1951-; Delshad, MojdehAccurately modeling Low Salinity Water Injection (LSWI) is essential for reliable predictions of oil recovery which affects exploration project planning and investment decisions. During LSWI, we modify the ions present in water before injection into an oil reservoir which helps maintain reservoir pressure and recover more oil from the reservoir, as compared to untreated regular water injection. Thus, understanding the primary mechanism and their effect of improved oil recovery due to wettability alteration during LSWI, and accurately modeling it, is essential to reliably predict and maximize oil recovery. However, there are several proposed models for numerical simulation of this novel method of LSWI and there exists no comparison for choosing the best model for an accurate simulation study. This study uses two simulators: (1) coupled reservoir simulator with geochemistry capabilities, UTCOMP-IPhreeqc and (2) commercial simulator, CMG’s GEM. We compare three models for numerical simulation of LSWI: (1) calcite dissolution, (2) total ionic strength, and (3) Extended Derjaguin, Landau, Verwey, and Overbeek (EDLVO). Most importantly, we also perform comparisons at both field and core scale. We describe the modeling capabilities of the two simulators and perform literature review to summarize the proposed mechanisms and the theory behind existing models. Finally, we simulate on (1) a synthetic carbonate field case, (2) a sandstone coreflood from a published literature, and (3) another sandstone coreflood, each with distinct mineralogy and petrophysical properties, to compare the three models. Results show that only the EDLVO model implemented in UTCOMP-Iphreeqc was able to accurately model the wettability alteration by estimating the change in contact angle during LSWI for all cases. While predicted recoveries from some of the models were similar, further investigation into the results uncovered the shortcomings of the other two models which resulted in incorrect calculation of the interpolating parameter. We concluded that the EDLVO model in UTCOMP-IPhreeqc works for all minerology while the other two models are scale, mineralogy, and case dependent. In future, we aim to develop a screening guide to choose model depending on the case, for simulating LSWI in commercial simulators which lack some of the mechanistic modeling capabilities of UTCOMP-IPhreeqc.Item Consumer-Data Approach to Assess the Effect of Residential Grid-Tied Photovoltaic Systems and Electric Vehicles on Distribution Transformers(IEEE, 2014-06) Uriarte, F. M.; Toliyat, A.; Kwasinski, A.; Hebner, R. E.The authors examine the impact of residential photovoltaic arrays and electric vehicles on distribution transformers by using 3-D surface and 2-D filled contour plots. These visualizations, somewhat unorthodox to power distribution analysis, elucidate the impact of hundreds of assets on distribution transformers on a single view. The visualizations are created with a smart grid computer model that accepts residential electrical recordings in one minute intervals. An analysis of simulation results shows that the electrical footprint experienced by a residential community and its distribution transformers stems from photovoltaic arrays rather than from electric vehicles. Additionally, the results indicate the existing distribution assets may be ready to support the proliferation of photovoltaic arrays and electric vehicles, a common concern across utilities in the United States.Item Drilling performance improvement : Brett and Millheim model adaptations for interaction effects and multiple learners(2012-08) Coddou, Ginny Anne; Groat, Charles G.; Jablonowski, Christopher J.This work reviews concepts in drilling-based learning curves and proposes modifications to the Brett and Millheim learning curve model to enable its use for multiple learners and to characterize interaction effects between learners. Enabling the model’s use for multiple learning scenarios at once improves modeling efficiency. Interaction effects are present when learners improve from their own experience and the experience of those in close proximity to them. Quantifying interaction effects leads to a more complete understanding of performance improvement and enables more effective forecasting of drilling resources and expenditure requirements.Item Dynamic subdivided relative humidity model of a polymer electrolyte membrane fuel cell(2013-05) Headley, Alexander John; Chen, Dongmei, Ph. D.The development of a control-oriented dynamic relative humidity model for a polymer electrolyte membrane (PEM) fuel cell stack is presented. This model is integrated with a first law based thermal model, which tracks energy flow within four defined control volumes in the fuel cell; the cathode channel, anode channel, coolant channel, and fuel cell stack body. Energy and mass conservation equations are developed for each control volume. On top of mass conservation, electro-drag and osmosis models were also implemented within the model to account for the major modes of vapor transfer through the membrane between the anode and cathode. Requisite alterations to the thermal model as well as mass flow rate calculations are also discussed. Initially, the model utilized a single lumped control volume for the calculation of all values each channel (anode and cathode). This lumped value method is computationally inexpensive, and makes the model optimal for control design. However, investigation of the mass-based Biot number showed the need for greater granularity along the length of the channels to properly capture the relative humidity dynamics. In order to improve the resolution of the model, while still minimizing the computation expense, the model was subdivided into a series of lumped value models. The cathode channel was the point of focus as it is the major concern from a controls perspective. This method captures the proper trends found in far more complex CFD models, while still maintaining a quick calculation time. Different levels are subdivision (3 and 6 submodels) are investigated, and the differences discussed. Particularly, temperature range, relative humidity range, the effect on the modeled voltage, and calculation time are compared. This control-oriented model is low order and based on lumped parameters, which makes the computational expense low. Formulation of this model enables the development of control algorithms to achieve optimal thermal and water management.Item The effect of restrictive diffusion on hydrate growth(2016-05) Andris, Ryan Gerald; Daigle, Hugh; Mohanty, KishoreMethane hydrate is formed naturally in a number of geologic settings around the world. The most predominant methane hydrate reservoirs are found in shallow oceanic basins at low temperatures and high pressures. A widely observed phenomenon in these oceanic sequences is extensive fine-grained sediments containing little to no hydrate interbedded with highly saturated sand bodies (20-60%). At Walker Ridge Block 313 in the Gulf of Mexico, one particular coarse-grained bed (approximately 3m-thick) is estimated to have methane hydrate occupying as much as 60% of the available pore space surrounded by hydrate-free clay. Here, I develop a numerical model that simulates methane hydrate growth in shallow oceanic basins in order to test whether diffusive transport of methane is a viable transport mechanism for forming highly saturated sand layers. I conclude that methane diffusion is likely responsible for the key identifying features of hydrate formation in interbedded sands and shales (i.e. greater hydrate saturations at the sand boundaries surrounded by hydrate-free zones in the fine-grained matrix). In addition, I show that the key parameters affecting the hydrate saturation profile include the amount of available methane for hydrate growth, thickness of the sand layer, and the radius of the fine grained pore space. I also discuss the shortcomings of the developed model and what complexities need to be added to more accurately reproduce hydrate growth throughout intricate hydrogeologic systems.Item Effects of solar heating and insulation on model biodigester temperature(2022-07-29) Leigh, Rush; Ellzey, Janet L.; Mullett, JohnBiodigesters are important tools for the faecal sludge management in developing countries and in emergency situations such as refugee camps. The effectiveness of biodigesters, however, is limited by local circumstances such as infrastructure issues, skilled labor shortages, and most importantly for this project: lack of temperature control of the biodigester. This thesis delves into the state of the art of biodigesters deployed in emergency situations, identifies many of the specific problems with biodigester effectiveness, and focuses on the importance of regulating the temperature of the biodigester. The primary question addressed by the research is: Can insulation and heating provided by photovoltaic panels (PV) effectively maintain a biodigester (of differing sizes and environmental conditions) within the temperature range for mesophilic digestion (35℃ to 37℃)? Starting with physical experimentation, data was collected from small scale mock biodigesters, under different conditions (e.g., covered by a tarp vs. uncovered, insulation vs. no insulation). Using heat transfer equations, a computational model was developed and compared to experimental data, in order to validate the model. The model results showed good agreement with the experimental measurements in all cases studied. For all experiments, the average difference between the model and actual data was less than 1°C, with the exception of one experiment (Heated vs. Heated Model), which had an average difference of less than 2°C. Using the validated model, four predictive cases were created: (1) Flexigester in coastal climate; (2) Flexigester in seasonal climate; (3) Anaerobic Lagoon in coastal climate; (4) Anaerobic Lagoon in seasonal climate. The scale of the flexigester is approximately 30 m³, and the Anaerobic Lagoon is 1,400 m³. These cases revealed the necessary amount of insulation and number of solar panels necessary to maintain the temperature within the desired range. Although exacts costs are not possible to obtain from these simulations, the results can still be used to determine the cost of the specified amount on insulation and number of solar panels. The estimated costs for retrofitting the four biodigester cases with insulation and solar panels to achieve optimal temperature ranges for mesophilic digestion are as follows: (1) $825, (2) $1,875, (3) $33,750, (4) $78,750. In summary, insulation in combination with heating provided by PV is a promising approach for maintaining biodigesters at optimal temperatures.Item An efficient hybrid model reduction for use with the AMLS method for frequency response problems(2010-05) Li, Qinqin, 1980-; Bennighof, Jeffrey Kent, 1960-; Sirohi, JayantA hybrid model reduction for use with the automated multilevel substructuring (AMLS) method is presented for frequency response analysis of complex structures. Structure responses to harmonic excitations and quasi-static responses to dominant damping forces are included in a reduced approximation subspace. Both types of responses greatly increase the efficiency of the subspace for solving the frequency response problem (FRP) for systems with high modal density and structural damping, and provide a good preparation for future frequency-dependent problems. A distilled subspace assumed to provide accurate frequency responses is generated from the finite element (FE) models by using the AMLS method. Then, the hybrid model reduction method is used to reduce the distilled subspace into a small new subspace. Three types of vectors are used to construct this subspace. The first type is distilled subspace dynamic response vectors (DRVs), which are exact solutions in the distilled subspace at certain chosen frequencies, called the DRV frequencies. The second type is modal DRVs, which are inexpensive approximate solutions calculated in an eigenspace. The third type is damping deformation vectors (DDVs), which provide information about response of the structure to damping effects. As exact responses, the distilled subspace DRVs eliminate frequency response errors at the DRV frequencies, and improve the accuracy at nearby frequencies as well. A small number of DRV frequencies are chosen carefully to offer maximum benefit with minimal computational cost. The modal DRVs are approximated very inexpensively from a suitable eigenspace. Only the diagonal entries in the modal coefficient matrices are used, along with low-rank updates that improve the accuracy of the modal DRVs and are applied using the Sherman-Morrison-Woodbury formula. Because of their low cost, a large number of modal DRVs constitute the major part of the reduced subspace. A small number of DDVs represent response to provide damping with minimal computational cost. The dimension of the final subspace is minimized by removing any redundancy through a special implementation of the QR factorization. This method results in a much smaller new subspace than the one from traditional modal truncation while achieving the same FRP accuracy. Such an efficiency also establishes a good foundation for future application in frequency-dependent problems.Item Electrical effects and computer simulations of PhotoBioModulation of the brain(2023-08-21) Huang, Li-Da; Abraham, Jacob A.; González-Lima, Francisco, 1955-; Pan, David Zhigang; Akinwande, Deji; Liu, HanliBrain functions have been shown to be affected by external stimuli. Low- Level-Light Therapy (LLLT) using nearinfrared photons is one of the effective ways to modulate the hemodynamic activities in the brain. However, the biphasic hormetic dose-response where bioenergetics are stimulated at a low dose and inhibited at a high dose is well observed in all photon stimulations. The amount of photon energy delivered to the brain are affected by the wavelength as well as the multilayered head structure with variations of optical parameters (OPs). A real 3D volume head model is built for each participant in this study, and the boundary conditions of each OP in each layer is considered. The Monte Carlo simulation with wavelengths ranging from 650 nm to 1064 nm is implemented to investigate the energy delivered to the brain under different radiation profiles. Results show that 1064-nm photons penetrate deeper than 810-nm photons except for scalp absorption at the lower bound due to low melanin content. Collimated-beam radiation is better than diverging-beam due to a more uniform intensity distribution at the scalp surface. Further research to optimize LLLT dosage for each individual is imperative due to the high inter-person variability in structure and OPs. In addition, EEG measures voltage fluctuations resulting from ionic current within the neurons of the brain. Because ATP is the major energy unit consumed by the ion pump and gate, PhotoBioModulation(PBM) will theoretically affect EEG. We establish the relationship between PBM and EEG demonstrated that PBM could increase resting-state alpha, beta, and gamma power by an increment of prefrontal blood oxygen level.Item Forecasting off-street bicycle facility demands(2020-12-04) Hall, Jennifer; Machemehl, Randy B.Some efforts to address the problem of environmental injustice toward minority and low-income populations include the incorporation of bicycle facilities into an infrastructure. Some well-known positive impacts that come with bicycle facilities are better health, increase in food availability, employment access and ultimately regional sustainability. In order to begin the process of identifying all the other positive impacts that could come with the implementation of bicycle facilities one must estimate how many users these bicycle facilities will attract, in other words, forecast the user demand of these bicycle facilities. This thesis focuses on off-street bicycle facilities and begins by evaluating current and past predictive models that are used for forecasting off-street bicycle facility demand. Noting these past models, we created multiple statistical models from locally sourced data that connect bicycle facility counts to time, demographics, and weather data. Due to the lack in the sheer number of off-street bicycle counters throughout the City of Austin, correlation between demographic and bicycle count data was problematic, yet all three models can in fact be applied to different off-street bicycle facilities only if locally sourced data is acquired.Item A fundamental approximation in MATLAB of the efficiency of an automotive differential in transmitting rotational kinetic energy(2012-05) Vaughn, James Roy; Matthews, Ronald D.; Bryant, Michael D.The VCOST budgeting tool uses a drive cycle simulator to improve fuel economy predictions for vehicle fleets. This drive cycle simulator needs to predict the efficiency of various components of the vehicle's powertrain including any differentials. Existing differential efficiency models either lack accuracy over the operating conditions considered or require too great an investment. A fundamental model for differential efficiency is a cost-effective solution for predicting the odd behaviors unique to a differential. The differential efficiency model itself combines the torque balance equation and the Navier-Stokes equations with models for gear pair, bearing, and seal efficiencies under a set of appropriate assumptions. Comparison of the model with existing data has shown that observable trends in differential efficiency are reproducible in some cases to within 10% of the accepted efficiency value over a range of torques and speeds that represents the operating conditions of the differential. Though the model is generally an improvement over existing curve fits, the potential exists for further improvement to the accuracy of the model. When the model performs correctly, it represents an immense savings over collecting data with comparable accuracy.Item G.R.A.C.E. satellite thermal model(2012-12) Jones, Fraser Black III; Howell, John R.I developed a thermal model of the Gravity Recovery and Climate Experiment satellite for the Center for Space Research to use in verifying their thermal models and for developing the next generation of satellites for their experiments. I chose COMSOL to model the satellite and used ProEngineer and 3Ds Max to generate the mesh from a .STEP file provided by DaimlerChrysler. I adjusted the model based on previous computer models and actual telemetry data from the GRACE satellite provided from 2002 through 2008. Using the model, I developed a sensitivity analysis of the satellites key thermal environment components and used that to recommend design changed for the next generation of satellites. Special attention should be given to redesigning the Star Camera Arrays and the heat transfer between the Main Equipment Platform and the Radiator.Item A generalized flow rate model for primary production and an analysis of gravity drainage through numerical simulation(2014-08) Vitter, Cameron Artigues; Balhoff, Matthew T.; Lake, Larry W.The age of “easy” oil has steadily declined through the years as many conventional land-based fields have been depleted to residual levels. Novel technologies, however, have reawakened old fields, allowing incremental oil to be added to their recoverable oil in place (ROIP). Underground Gravity Drainage (UGD), an example of one of these technologies, combines improved horizontal and deviated drilling technologies with the longstanding concept of gravity drainage. In this work, a better understanding of gravity drainage has been gained through (1) development of a numerical, three-dimensional, three-phase reservoir simulator (UT-EMPRES), (2) development of a universal, semi-empirical model of production rates through primary depletion, and (3) analysis of the important aspects of gravity drainage through simulation. UT-EMPRES is a new three-phase, finite-difference reservoir simulator, which utilizes a simple, easy-to-use Microsoft Excel interface to access MATLAB-programmed simulation code. This simulator produces nearly identical results to other well-established simulators, including UTCHEM and CMG. UT-EMPRES has some unique features, allows for easy post-processing in MATLAB, and has been utilized extensively in the other two areas of this thesis. The generalized flow rate model (GFRM) is a semi-empirical equation that is used to forecast the dynamic primary production rate of a reservoir with an arbitrary number of wells all operating at the same constant pressure condition. The model is an extension of the classic tank model, which is inherently a single flowing phase development. With the ability to make a priori predictions of production figures, users can screen various prospect assets on the basis of economic potential through optimization routines on the GFRM. Gravity drainage and its approximation through numerical simulation are analyzed. A sensitivity study was conducted on three-phase gravity drainage, leading to the conclusion that small changes in vertical permeability and portions of the relative permeability-saturation relationships can greatly affect production rates. Finally, two-phase (oil and air) and regions of three-phase (water, oil, air) flow simulations were found to exhibit exponential decline in phase production rates, which may enable the GFRM to be applicable to UGD-type processes.Item Harmonics diversity simulation of inverter based generators in large-scale power systems(2020-05-07) Osorio Perez, Fernando Elias; Santoso, SuryaThis thesis explores the application of inverter-based generator (IBG) models in EMTP-RV. The main goal is to analyze how accurately different models can simulate the harmonic current spectra of an IBG, and study the interaction of the harmonic currents with the grid. The models to be studied are the detailed switch model (DM), harmonic averaged model (HAVM), and automated current source model (ACSM). In this work, the harmonic current diversity is also studied; the diversity arises from the different generating set-points of an IBG and the grid configuration at the point of interconnection (POI). Furthermore, the application of the ACSM in power systems with conventional generators is elaborated through a step-by-step procedure.Item How information asymmetry affects contract design : paying for private firms with IOU's(2016-05) Jansen, Mark; Parrino, Robert, 1957-; Fracassi, Cesare; Almazan, Andres; Hartzell, Jay; Starks, Laura; Abrevaya, JasonThis dissertation examines a financing mechanism that is common in the acquisition of privately-held firms. Using a novel database of transactions in which the target firm is private, this paper shows that sellers receive a debt claim as a contingent payment for the firm that is being sold. The debt claim, which takes the form of seller financing, is secured by the assets of the target firm. I show that proxies for information asymmetry are correlated with the presence of seller financing as payment in the transaction. I also find that when the firm is more likely to have received a financial audit, the transaction is less likely to include seller financing. Since financial audits improve firm transparency, I interpret this as evidence that a reduction in information asymmetries between the parties of a acquisition affect the deal structure. A complementary explanation for the use of seller financing is related to capital constraints faced by buyers in the financing of the transaction. I present evidence that contract structures are affected by cross-sectional and time-series changes in the supply of local investment capital for buyouts. I find that seller financing is less common in areas in which locally informed capital is more abundant. I also find that transactions contain a lower percentage of seller financing in city-years in which Small Business Administration provides loan guarantees for the acquisition and expansion of firm’s loan guarantees are higher. The evidence suggests that seller financing is solving a contracting problem because it is unaffected by controls for local banking activity.Item Investigation of analytical models incorporating geomechanical effects on production performance of hydraulically and naturally fractured unconventional reservoirs(2014-08) Aybar, Umut; Sepehrnoori, Kamy, 1951-; Patzek, Tadeusz W.Petroleum and Geosystems EngineeringItem Isogeometric analysis : applications for torque and drag models, drillstring and bottom-hole assembly design(2018-05) Hanson, Katy Lynn; Oort, Eric van; Foster, John T., Ph. D.The drilling industry today relies on torque and drag models to analyze and ensure success during all phases of well construction and operations, including planning, drilling, and completion. Analytical models are based on equations that are undergoing constant development and improvement. The finite element method is an alternative to complex analytical calculations that is used often to determine torque and drag forces that are present when a drillstring is lowered, raised, and rotated in a wellbore. Traditional finite element analysis (FEA), however, is not time efficient or computationally able to simulate the complexities of a real wellbore. Thus, we introduce an alternative to the traditional finite element approach: isogeometric analysis. Isogeometric analysis is similar to finite element analysis except that it uses NURBS (Non-Uniform Rational B-Splines), as opposed to interpolatory polynomials used in traditional FEA, as the basis functions. NURBS functions are the same as those used in CAD programs, and they are able to construct exact conic shapes, such as circles and ellipses. Adopting NURBS basis functions allows finite element analysis to be performed directly on the exact geometrical surface - not on an approximate geometric surface mesh, as in traditional FEA. IGA yields a significantly faster and more accurate simulation. This thesis presents a real-world application of IGA to a drag force model to determine the resultant surface hook load during run-in-hole (RIH) operations. Real well data is used, and IGA results are compared to a similar FEA analysis. The outcome shows that IGA is indeed a superior finite element method that has immense potential for further application in the industryItem Modeling of nanoparticle transport in porous media(2012-08) Zhang, Tiantian; Bryant, Steven L.; Huh, Chun; Delshad, Mojdeh; Prodanovic, Masa; Johnston, Keith P.The unique properties of engineered nanoparticles have many potential applications in oil reservoirs, e.g., as emulsion stabilizers for enhanced oil recovery, or as nano-sensors for reservoir characterization. Long-distance propagation (>100 m) is a prerequisite for many of these applications. With diameters between 10 to 100 nanometers, nanoparticles can easily pass through typical pore throats in reservoirs, but physicochemical interaction between nanoparticles and pore walls may still lead to significant retention. A model that accounts for the key mechanisms of nanoparticle transport and retention is essential for design purposes. In this dissertation, interactions are analyzed between nanoparticles and solid surface for their effects on nanoparticle deposition during transport with single-phase flow. The analysis suggests that the DLVO theory cannot explain the low retention concentration of nanoparticles during transport in saturated porous media. Moreover, the hydrodynamic forces are not strong enough for nanoparticle removal from rough surface. Based on different filtration mechanisms, various continuum transport models are formulated and used to simulate our nanoparticle transport experiments through water-saturated sandpacks and consolidated cores. Every model is tested on an extensive set of experimental data collected by Yu (2012) and Murphy (2012). The data enable a rigorous validation of a model. For a set of experiments injecting the same kind of nanoparticle, the deposition rate coefficients in the model are obtained by history matching of one effluent concentration history. With simple assumptions, the same coefficients are used by the model to predict the effluent histories of other experiments when experimental conditions are varied. Compared to experimental results, colloid filtration model fails to predict normalized effluent concentrations that approach unity, and the kinetic Langmuir model is inconsistent with non-zero nanoparticle retention after postflush. The two-step model, two-rate model and two-site model all have both reversible and irreversible adsorptions and can generate effluent histories similar to experimental data. However, the two-step model built based on interaction energy curve fails to fit the experimental effluent histories with delay in the leading edge but no delay in the trailing edge. The two-rate model with constant retardation factor shows a big failure in capturing the dependence of nanoparticle breakthrough delay on flow velocity and injection concentration. With independent reversible and irreversible adsorption sites the two-site model has capability to capture most features of nanoparticle transport in water-saturated porous media. For a given kind of nanoparticles, it can fit one experimental effluent history and predict others successfully with varied experimental conditions. Some deviations exist between model prediction and experimental data with pump stop and very low injection concentration (0.1 wt%). More detailed analysis of nanoparticle adsorption capacity in water-saturated sandpacks reveals that the measured irreversible adsorption capacity is always less than 35% of monolayer packing density. Generally, its value increases with higher injection concentration and lower flow velocities. Reinjection experiments suggest that the irreversible adsorption capacity has fixed value with constant injection rate and dispersion concentration, but it becomes larger if reinjection occurs with larger concentration or smaller flow rate.Item Modeling teacher effectiveness as a function of student ability(2013-05) Jackson, Christian Dennis; Lin, Tse-minIn 2010, the L.A. Times newspaper used the test results of Los Angeles County elementary students to assess and rank the elementary teachers. They then published the results on their website. Publicly ranking teachers in this manner has important implications on the careers of the teachers being ranked. It is, therefore, important that any model claiming to rank teachers be as accurate as possible. It seems plausible that a teacher's ability to help a student depends upon that student's prior academic ability. Some teachers might be better at teaching gifted students while others might be better at teaching remedial students. The L.A. Times did not account for this in their model. This paper looks at the results of allowing teacher effect to vary with prior student ability and how that interaction affects the relative rankings of the individual teachers. To assess this, the same Value-Added model the L.A. Times used is employed, with the exception that teacher effect is allowed to vary with the prior abilities of the students. New teacher ranks are then calculated and compared with the ranks calculated by the L.A. Times. The results of this analysis show a relatively small number of rank changes between the two models. In general, allowing teacher effect to vary results in a 5% to 12% change in the rankings of both the Math and Reading teachers relative to the L.A Times model. Other research on the same data has resulted in a 20% to 55% change in the rankings of the Math teachers and a 40% to 65% change in the rankings of the Reading teachers relative to the L.A. Times model. Although ranking teachers is a popular idea for determining the distribution of funding, the model shown in this paper as well as the other models reviewed, illustrate that a change in the model results in a change in the rankings of the teachers. A model that allows teacher effect to vary with prior student ability results in a better model fit than a model that does not. Whether or not this is a good thing is hard to say. Two examples are provided in this paper. One shows a teacher whose rank appears to be artificially inflated by this model and the other shows a teacher whose rank appears to be artificially lowered by this method. Although the fit of the model proposed by this paper is better than the model used by the L.A. Times, it does not result in radical changes in the rankings of the teachers. Rather, it seems that teacher rankings are sensitive to the particular model used and there are countless numbers of valid models. For this reason it is not wise to release such sensitive information to the public. It is probably true that the weak teachers are ranked relatively low in this analysis and that the truly good teachers are ranked relatively high. However, these rankings should only be used as one part of a larger metric to rank teachers and too much weight should not be placed on them for the purposes of rewarding or penalizing teachers due to the sensitivity of the model specification.