Thesis (Ph. D.)--West Virginia University, 2006. / Title from document title page. Document formatted into pages; contains xx, 320 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 156).

La production de canneberges au Québec est l’une des plus importantes au monde. L’avancée de la production de canneberges au Québec est due, en partie, à une gestion optimisée des conditions d’irrigation et de drainage. Toutefois, il a été estimé que 75% des champs de canneberges du Centre-du-Québec ont des problèmes de drainage. Les problèmes de conception du système de drainage entraînent des réductions de 25% de rendement des canneberges tandis que le problème de colmatage des drains induit une réduction de 39%. De ce constat, le présent a pour objectif de prédire le degré de colmatage des drains à partir des hauteurs de la nappe à la mi-écartement, au-dessus du drain ainsi que du temps de rabattement de la nappe. Pour se faire, des modèles de prédiction par apprentissage supervisé et par régression linéaire multiple sont utilisés. Les expériences de drainage ont été menées en laboratoire à l’aide d’un simulateur de drainage afin de modéliser de façon analogique le degré de colmatage du drain. Les données expérimentales collectées à la fin de chaque expérience ont permis, entre autres, de générer les variables d’entrée des modèles prédictifs. Les résultats montrent qu’une augmentation du degré de colmatage du drain entraîne : (i) des augmentations du temps de rabattement de la nappe et de la résistance à l’entrée du drain, (ii) une diminution de la conductivité hydraulique équivalente du système de drainage et (iii) l’accumulation d’eau au voisinage du drain. Le modèle prédictif d’apprentissage supervisé ajusté explique 98,9% de la variation du degré de colmatage tandis que la régression n’explique que 58,9% de cette variation. Avec l’introduction de la conductivité hydraulique équivalente du système de drainage dans le modèle, la qualité de l’ajustement du modèle d’apprentissage supervisé passe de 98,9 à 99,8% et celle de régression linéaire multiple passe de 58,9 à 83,8%. / Quebec is one of the largest cranberry producer in the world. The substantial yields in cranberry production in Quebec is due in part to optimized management of irrigation and drainage conditions. However, it has been estimated that 75% of the cranberry fields in Central Quebec have drainage problems. Drainage system design problems result in 25% reduction in cranberry yields drain clogging issues are responsible for a 39% reduction. The present project proposes to predict the degree of clogging based on water table heights at the mid-spacing and above the drain in addition to the water table drawdown time. To do this, predictive models using supervised learning and multiple linear regressions are used. The drainage experiments were conducted in the laboratory using a drainage simulator to analogically reproduce the degree of drain clogging. The experimental data collected at the end of each experiment served, among other things, to generate the input variables of the predictive models. The results show that an increase in the degree of clogging leads: (i) to an increase in water table drawdown time and the drain entrance resistance, (ii) a decrease in the equivalent hydraulic conductivity of the drainage system and (iii) the accumulation of water in the vicinity of the drain. The adjusted supervised learning predictive model explains 98.9% of the variation of the clogging degree, while the regression model explains only 58.9%. With the introduction of the equivalent hydraulic conductivity of the drainage system, the quality of the supervised learning model goes from 98.9 to 99.8% and that of multiple linear regressions from 58.9 to 83.8%.

Finite element modeling of laminar shear box testing that consisted of loose sand treated with large diameter prefabricated vertical drains (PVDs), was performed. The objective of the modeling was to evaluate the reliability of the computer program FEQDrain for predicting excess pore pressure ratios (Ru) at sites treated with prefabricated drains. FEQDrain was found to be capable of successfully modeling measured excess pore pressure ratio time histories from the laminar shear box experiment, as long as an appropriate combination of €˜number of equivalent cycles and €˜shaking duration was chosen, and sensitive parameters were in the range of measured values. Hydraulic conductivity, soil compressibility, and cycles to liquefaction are sensitive parameters and govern the computed Ru values.Modeling shows that the loading rate in the laminar shear box (15 cycles at 2 Hz) likely induced higher Ru values than would be expected in a typical earthquake event with a longer duration. The longer duration allows the drains to dissipate pore pressures and prevent liquefaction. The number of equivalent cycles and duration of shaking combinations recommended for various moment magnitudes in the FEQDrain user manual predict lower, but similar Ru versus time curves. Thus, suggesting that PVDs would be equally effective for any size earthquake. However, drains are most effective at preventing liquefaction when earthquake ground motions are long and uniform, rather than short and intense.Results from models in this study compare favorably with those from computer modeling performed by Howell et al. (2014). The individual hydraulic conductivity and compressibility values were different they were somewhat compensating. Similar Ru values can be modeled with different combinations of these parameters.Based on computer analyses, wick drains and 2€ diameter PVDs were found to be relatively ineffective for preventing liquefaction. However, 3€ diameter PVDs are fairly effective but can be overwhelmed during intense shaking. In contrast, 4€ diameter and larger PVDs are significantly more effective.

Finite element modeling of laminar shear box testing that consisted of loose sand treated with large diameter prefabricated vertical drains (PVDs), was performed. The objective of the modeling was to evaluate the reliability of the computer program FEQDrain for predicting excess pore pressure ratios (Ru) at sites treated with prefabricated drains. FEQDrain was found to be capable of successfully modeling measured excess pore pressure ratio time histories from the laminar shear box experiment, as long as an appropriate combination of ˜number of equivalent cycles and ˜shaking duration was chosen, and sensitive parameters were in the range of measured values. Hydraulic conductivity, soil compressibility, and cycles to liquefaction are sensitive parameters and govern the computed Ru values.Modeling shows that the loading rate in the laminar shear box (15 cycles at 2 Hz) likely induced higher Ru values than would be expected in a typical earthquake event with a longer duration. The longer duration allows the drains to dissipate pore pressures and prevent liquefaction. The number of equivalent cycles and duration of shaking combinations recommended for various moment magnitudes in the FEQDrain user manual predict lower, but similar Ru versus time curves. Thus, suggesting that PVDs would be equally effective for any size earthquake. However, drains are most effective at preventing liquefaction when earthquake ground motions are long and uniform, rather than short and intense.Results from models in this study compare favorably with those from computer modeling performed by Howell et al. (2014). The individual hydraulic conductivity and compressibility values were different they were somewhat compensating. Similar Ru values can be modeled with different combinations of these parameters.Based on computer analyses, wick drains and 2 diameter PVDs were found to be relatively ineffective for preventing liquefaction. However, 3 diameter PVDs are fairly effective but can be overwhelmed during intense shaking. In contrast, 4 diameter and larger PVDs are significantly more effective.

In the design of tank foundations several design techniques are considered. This study focuses on grade supported tanks constructed under an extensive preload and instrumentation program. Settlement estimation methods were performed and compared to field instrumentation data taken at the project sites. Three project sites were selected for this study. The geotechnical investigations were performed by Eustis Engineering Services, L.L.C. and included both undisturbed soil borings and cone penetrometer tests. Conclusions were made about the accuracy of the calculations and how assumptions affect the settlement computation results.

This thesis describes the flow of ground water with a free water table, in an isotropic and homogeneous soil overlying a horizontal impermeable layer, from a surface-applied rainfall or irrigation supply to a system of horizontal tube drains. The idealized representation of this situation is in terms of a potential flow towards line sinks. The tube drains (sinks) may be at atmospheric pressure (free flow) or under back-pressure (submerged flow). After describing the flow to systems of drains at uniform depth and equal spacings, the thesis shows the effect of horizontal, vertical and combined misalignments of a single line sink on the free surface, the flow pattern and the rates of flow to each sink. The study was carried out by viscous-flow and electrical analogues, by a numerical method (Relaxation), and by the exact solution of equations based on physical approximations. In theory the extent of the effect of any misalignment is infinite. Experimentally it was found that the vertical misalignment had a significant effect on the free surface, discharge and flow net over the whole range of the experiment, while the horizontal one had a noticeable effect only extending over one spacing on either side of the misaligned drain. An approximate mathematical solution for the horizontal displacement has been obtained. Experiments on the viscous-flow model revealed the importance of the effect on the free surface etc. of the back pressure. The distribution of the effect along the free surface and the correlation between the free surface height and the back pressure are also presented. A new approach was used to elucidate the effect of negative pressure (capillary rise) on the free surface and obtain the surface of atmospheric pressure. An equation, independent of the geometrical scale ratio was derived to relate the various parameters for the unsteady flow to the scaling laws. The introduction of a reference potential at a fixed point in the field provides a method for the correction of the change with time in the position of equipotentials and represents a new technique. The variations of free surface heights with discharge, as observed on the viscous-flow model, were compared with the results of s some recent theories and showed discrepancies between the experiment and theories.

Thesis (Ph. D.)--University of Missouri-Columbia, 2003. / Typescript. Vita. Includes bibliographical references (leaves 198-204). Also available on the Internet.

Thesis (Ph. D.)--University of Missouri-Columbia, 2003. / Typescript. Vita. Includes bibliographical references (leaves 198-204). Also available on the Internet.

Thesis (Ph.D.)--University of Wollongong, 2004. / Typescript. Includes bibliographical references: leaf 243-252.

Thesis (Ph. D.)--University of Wollongong, 2005. / Typescript. Includes bibliographical references (leaf 241-251).