No description available.

Master of Science / Department of Animal Sciences and Industry / Jeffrey S. Stevenson / We hypothesized that pregnancy outcomes may be improved by inducing luteal regression, ovulation, or both before a control CO-Synch + CIDR program (100 mcg GnRH i.m. [GnRH-1] and insertion of a progesterone-impregnated intravaginal controlled internal drug release [CIDR] insert on d -10, 25 mg PGF2alpha (PG) i.m. and CIDR insert removal on d -3, and 100 mcg GnRH i.m. [GnRH-2] and timed AI [TAI] on d 0) in suckled beef cows. This hypothesis was tested in 2 experiments, in which cows were treated with either PG or PG + GnRH before initiating a control CO-Synch + CIDR program to increase the proportion of cows starting the program in a low (< 1 ng/mL; Exp. 1) or high (≥ 1 ng/mL; Exp. 2) progesterone status, respectively. Blood was collected before each injection for later progesterone analyses. In Exp. 1, cows at 9 locations (n = 1,537) were assigned to either: (1) control or (2) PrePG (same as control with a PG injection on d -13). The PrePG cows had larger (P < 0.05) follicles on d -10 and more (P < 0.05) ovulated after GnRH-1 than controls (60.6 vs. 36.5%). Incidence of estrus between d -3 and 0 was greater (P < 0.05) for treated multiparous cows than multiparous controls and treated and control primiparous cows (74.1 vs. 64.3, 58.6, and 59.1%, respectively). In Exp. 2, cows at 4 locations (n = 803) were assigned to: (1) control (same as Exp. 1) or (2) PrePGG (same as control with PG injection on d -20 and GnRH injection on d -17. Cows with BCS > 5.0 or ≥ 70 d postpartum at TAI were more (P < 0.05) likely to become pregnant than thinner cows or those with fewer days postpartum. Treated cows in both experiments were more (P < 0.05) likely than controls to have luteolysis after initial PG injections and reduced (P < 0.05) serum progesterone. In both experiments, pregnancy rates at d 35 did not differ between treatment and control; however, cows classified as anestrous before d -10, but with elevated progesterone on d -10, had increased (P < 0.05) pregnancy outcomes than remaining anestrous cows with low progesterone concentrations. In summary, luteal regression and ovulation were enhanced by treatments before the 7 d CO-Synch + CIDR program; however, pregnancy per TAI was not improved.

Includes bibliographical references / In this thesis, the development of a strongly coupled, partitioned fluid-structure interactions (FSI) solver is outlined. Well established methods are analysed and new methods are proposed to provide robust, accurate and efficient FSI solutions. All the methods introduced and analysed are primarily geared towards the solution of incompressible, transient FSI problems, which facilitate the use of black-box sub-domain field solvers. In the first part of the thesis, radial basis function (RBF) interpolation is introduced for interface information transfer. RBF interpolation requires no grid connectivity information, and therefore presents an elegant means by which to transfer information across a non-matching and non-conforming interface to couple finite element to finite volume based discretisation schemes. The transfer scheme is analysed, with particular emphasis on a comparison between consistent and conservative formulations. The primary aim is to demonstrate that the widely used conservative formulation is a zero order method. Furthermore, while the consistent formulation is not provably conservative, it yields errors well within acceptable levels and converges within the limit of mesh refinement. A newly developed multi-vector update quasi-Newton (MVQN) method for implicit coupling of black-box partitioned solvers is proposed. The new coupling scheme, under certain conditions, can be demonstrated to provide near Newton-like convergence behaviour. The superior convergence properties and robust nature of the MVQN method are shown in comparison to other well-known quasi-Newton coupling schemes, including the least squares reduced order modelling (IBQN-LS) scheme, the classical rank-1 update Broyden's method, and fixed point iterations with dynamic relaxation. Partitioned, incompressible FSI, based on Dirichlet-Neumann domain decomposition solution schemes, cannot be applied to problems where the fluid domain is fully enclosed. A simple example often provided in the literature is that of balloon inflation with a prescribed inflow velocity. In this context, artificial compressibility (AC) will be shown to be a useful method to relax the incompressibility constraint, by including a source term within the fluid continuity equation. The attractiveness of AC stems from the fact that this source term can readily be added to almost any fluid field solver, including most commercial solvers. AC/FSI is however limited in the range of problems it can effectively be applied to. To this end, the combination of the newly developed MVQN method with AC/FSI is proposed. In so doing, the AC modified fluid field solver can continue to be treated as a black-box solver, while the overall robustness and performance are significantly improved. The study concludes with a demonstration of the modularity offered by partitioned FSI solvers. The analysis of the coupled environment is extended to include steady state FSI, FSI with free surfaces and an FSI problem with solid-body contact.

El objetivo es la implementación de sistemas inteligentes para ajustar modelos hidrológicos, comparando las series temporales con redes neuronales, que permitan el aprendizaje y ajuste de parámetros para la obtención de modelos que realicen predicciones óptimas de alturas del río Paraná, en períodos de inundaciones. El interés radica en su aplicación en la provincia de Corrientes, Argentina, afectada por inundaciones que ocasionan pérdidas en la economía regional. Se realiza un análisis previo con series temporales que permite establecer las variables y factores que determinan las alturas hidrométricas, en períodos de inundación en la localidad de Corrientes. Posteriormente se presenta un pronóstico a corto plazo en períodos de crecidas, que predice las alturas hidrométricas a tres días implementando redes neuronales con función de penalización modificada. Se finaliza con un pronóstico a mediano plazo, para períodos de inundación, de alturas hidrométricas a siete días que se realiza con redes neuronales con diferentes arquitecturas. / The aim is the implementation of intelligent systems to adjust hydrological models comparing time series and neural networks which allow learning and setting parameters for models that make optimal predictions of the Paraná river heights in flood periods. The interest lies in its implementation in the province of Corrientes, Argentina, hit by floods causing losses in regional economy. We performed a time-series analysis to discover the variables and factors that influence the hydrometric height in flood periods in the town of Corrientes. Subsequently we present a short-term prediction for flood periods, which predicts the hydrometric heights three days in advance, using neural networks with a modified penalty function. Then we obtain a medium-term forecast for flood periods, seven days in advance, using neural networks with different architectures.

The aim of this master's thesis is to build artificial neural network that is able to calculate varying single tooth-pair mesh stiffness of spur gear for given input parameters. The training set for this network was determined by computational modelling by finite element method. Therefore, creating of computational model and mesh stiffness calculating were a partial aim of this thesis. Input parameters for stiffness calculation were number of driving and driven gear teeth and gear loading. Creating of computational model and performing series of simulations was followed by creating artificial neural network. Multilayer neural network with backpropagation training was chosen as a type of the network. Created neural network is sufficiently efficient and can determine varying mesh stiffness in input set range for learned input parameters and for values of parameters that are not included in training set as well. This neural network can be used for varying single tooth-pair mesh stiffness estimation in input set range.

Voice assistants are a remarkable example of the potential for AI to become further entwined with social life. However, they are produced by some of the world’s largest tech corporations and are rooted in capitalistic processes that depend on user data. This thesis presents a qualitative exploratory study of voice assistants. Through a combination of interviews and theoretical analysis, it focuses on participants’ perceptions and experiences with these AI agents and how they are embedded in the bigger picture of surveillance capitalism. The findings reveal the physical characteristics and personality traits that participants in this study ascribe to voice assistants, highlighting the implications of treating voice assistants as personified agents and the factors contributing to these perceptions. Further, this thesis examines how surveillance capitalism is present in participant interactions with these technologies and identifies how its reach into people’s lives is provoked by their design and background contexts. Lastly, it provides an overview of corporate power in the tech industry and how the structural, cultural, and political circumstances enable and legitimize big tech’s authority in digital environments and how this situates the individual and their capacity to contend with technological issues. / Graduate / 2023-07-12