This thesis is in support of my Master of Fine Arts exhibition in the Carroll Reece Museum at East Tennessee State University. It describes a body of work that strives for a balance of form and aesthetics. Chapter 1 charts how I refined my personal stylistic qualities. Chapter 2 gives a brief overview of the history of ceramics techique. Chapter 3 discusses the surface treatment and construction of the pieces in the show. Chapter 4 presents each piece in chronological order.

No description available.

We propose an exact filtered backprojection algorithm for inversion of the cone beam data in the case when the trajectory is composed of a distorted circle and a line segment. The length of the scan is determined by the region of interest , and it is independent of the size of the object. With few geometric restrictions on the curve, we show that we have an exact reconstruction. Numerical experiments demonstrate good image quality.

Characterization and assessment of liquefaction potential of gravelly soil in a reliable cost-effective manner has always been a great challenge for the geotechnical engineers. The typical laboratory investigation techniques have proven to be ineffective for characterizing gravelly soil due to the cost and difficulty of extracting undisturbed sample from gravelly deposits. The traditional in-situ tests like SPT or CPT are not very suitable for gravelly soil because of interference with large size gravel particles which can artificially increase the penetration resistance. The Becker Penetration Test, well known for gravelly soil characterization, is cost-prohibitive for routine projects and is not available in most of the world. The Chinese dynamic cone penetration test (DPT) with a larger diameter probe compared to the SPT or CPT, can be economically performed with conventional drilling equipment. Besides the penetration testing, in-situ measurement of shear wave velocity (Vs) is another alternative of characterizing gravel liquefaction. Probabilistic liquefaction triggering curves were developed by performing both DPT and shear wave velocity test at the Chengdu Plain of China where massive gravel liquefaction took place during 2008 Wenchuan earthquake. These curves have significant uncertainty as they were developed from a single event database. As a part of this study, both DPT and Vs tests have been performed at various sites around the world where gravelly soil did or did not liquefy in various past earthquakes. These newly collected data have been added to the existing Chinese dataset to form a large database on gravel liquefaction case histories for both DPT and Vs. Based on this larger database, new magnitude dependent probabilistic liquefaction triggering procedures have been developed for both DPT and Vs. The larger database has significantly improved the triggering curves by reducing the spread and constraining the curves at both the higher and lower end. New Magnitude Scaling Factor (MSF) curves have been developed for both DPT and Vs which were found to be consistent with existing MSF curves. Further, an instructive comparison has been drawn between the performance of CPT and newly developed DPT triggering procedure the liquefaction potential of gravelly deposits CentrePort in Wellington. Results showed that both DPT and CPT performed reasonably well in liquefaction assessment of the gravelly fill. However, the CPT-based CRR profiles contain intermittent spikes due to the interaction with gravel particles whereas the DPT resistance appear to be relatively smooth. Similar comparison has been presented between the DPT and BPT in performing liquefaction assessment of gravelly soil at the Borah Peak sites in Idaho. It is found that both DPT and BPT successfully evaluate the liquefaction potential of the loose critical layers but the medium dense to dense layers are identified as non-liquefiable by the DPT whereas the same deposits are identified as liquefiable by the BPT. Lastly, an investigation has been carried out to observe the effect of hydraulic conductivity and in-situ drainage on the liquefaction triggering in gravelly soils based on field data along with a group of numerical analyses. It is found that the hydraulic conductivity of gravelly soil reduces with sand content which eventually may cause liquefaction during earthquake shaking. Low permeability cap layer may also impede the drainage path to generate excess pore pressure to trigger liquefaction in the gravelly strata.

Snow is a complex material that is difficult to characterize especially due to its high compressibility and temperature-sensitive nonlinear viscoelasticity. Snow mechanics has been intensively investigated by avalanche and army researchers for decades. However, fewer research studies have been published for compacted snow, commonly defined as snow with a density in the range of 370-560 kg/m3. From a mobility perspective, the tires are the primary point of force and motion generation and their interaction with the terrain causes an increased reliance on the skill of the driver for safer mobility. Thus, standards like ASTM F1805 are implemented for the evaluation of winter tires which can be used in harsh conditions like ice and snow. This work focuses on evaluating the prior efforts performed for the measurement of snow properties. In addition, analysis using regression models and principal component analysis is performed to understand the extent to which specific measurements related to snow affect the traction of the tire. It was found that the compressive and shear properties of snow contribute more than 90% to the variation in the traction coefficient of a tire when evaluated on a compacted snow domain per ASTM F1805. Identification of this phenomenon allowed the enhancement of an existing device that can be used for measuring the compaction and shear properties of snow. The device hence conceptualized was manufactured in-house and tested at the Smithers Winter Test Center to benchmark against existing devices available commercially. Further, a more analytical method for evaluating the resistive pressure for the penetration of the device was formulated. Based on this, a possible framework for the determination of the bevameter parameters using measurements of the new device has been proposed which needs to be validated experimentally and computationally. / Doctor of Philosophy / Winter tires sold in North America require prior evaluation according to a standard namely the ASTM F1805 to bear the 'mountain-snowflake symbol' for severe snow usage. The standard specifies the conditions for evaluating a prototype winter tire and the necessary track preparation methodologies. However, the computational model of a track used for such a certification is not found in the literature causing the manufacturing of such winter tires to be more of a 'trial-and-error' process. The main objective of this investigation is to assess earlier studies of snow characteristics. Additionally, analysis employing regression models and principal component analysis was conducted to comprehend the extent to which particular measurements connected to snow affect the traction of the tire. When tested using an ASTM F1805-compliant compacted snow domain, it was discovered that the compressive and shear properties of snow account for more than 90% of the variation in the traction coefficient of a tire. The discovery of this phenomenon made it possible to improve a tool for assessing the compaction and shear characteristics of snow. The device that was conceptualized was manufactured internally and put to the test at Smithers Winter Test Center to compare it to other devices that were already on the market. Further, a new analytical method for evaluation of the resistive pressure to the device was developed. Using measurements from the new device, a method to utilize the devised output parameters as inputs and for the validation of a computational snow model is proposed.

Soil liquefaction can cause devastating damage and loss and is a serious concern in civil engineering practice. One method for evaluating liquefaction triggering potential is a risk-targeted probabilistic approach that has been shown to provide more consistent and accurate estimates of liquefaction risk than traditional methods. This approach is a “performance-based” procedure which is based off of the performance-based earthquake engineering (PBEE) framework developed by the Pacific Earthquake Engineering Research (PEER) Center. Unfortunately, due to its complexity, performance-based liquefaction assessment is not often used in engineering practice. However, previous researchers have developed a simplified performance-based procedure which incorporates the accuracy and benefits of a full performance-based procedure while maintaining a more simplistic and user-friendly approach. Until now, these simplified performance-based procedures have only been available for the SPT (Standard Penetration Test). With the increasing popularity of the CPT (Cone Penetration Test), a simplified procedure is needed for CPT-based liquefaction assessment. This thesis presents the derivation of a simplified performance-based procedure for evaluating liquefaction triggering using the Ku et al. (2012) and Boulanger and Idriss (2014) models. The validation study compares the results of the simplified and full performance-based procedures. The comparison study compares the accuracy of the simplified performance-based and traditional pseudo-probabilistic procedures. These studies show that the simplified performance-based procedure provides a better and more consistent approximation of the full performance-based procedure than traditional methods. This thesis also details the development of the liquefaction loading maps which are an integral part of the simplified method.

3D-printing is a new and upcoming manufacturing technique that can significantly reduce time and material losses in production. Fused deposition modeling (FDM) is one of the most commonly used 3D-printing methods for processing conventional thermoplastic polymers. To reduce the printing time and usage of material via FDM technology, a user typically specifies infill density. Therefore, it is important to understand how this printing parameter affects the fire and mechanical properties of the 3D-printed object.  This study aims to investigate the effect of various infill densities on mechanical and fire properties of polylactic acid (PLA) composites produced by FDM 3D-printing technology. PLA composites of five different infill densities were 3D-printed: 20%, 40%, 60%, 80% and 100%. The samples for all tests were designed in AutoCAD and then imported into the slicing software, Ultimaker Cura. The 3D-printer used for printing was the Ultimaker S3 which uses FDM technology. To test the fire and mechanical behavior of 3D-printed PLA composites three tests were conducted: cone calorimeter test, tensile test and UL-94 flammability test. The cone calorimeter testing was done using the incident radiation of 35 kW/m2. The results showed that the trend of HHR curves of all infill densities are akin to each other, though the peak heat release rate and total heat released increases with higher infill density. Time to ignition was also longer for samples with higher infill density. Tensile testing was conducted according to the ASTM D638 standard. The results showed that with increasing infill density mechanical properties improved, with 100% infill density having the highest tensile strength (58.15 MPa) and elastic modulus (1472.1 MPa). From the UL-94 test results no difference in flammability could be observed. Every sample had no rating, which implies that PLA specimens of all infill densities are very flammable, with long afterflame and heavy flammable dripping. The study concludes that among the examined infill densities, no ideal percentage of infill density could be found. Requirements based on application will determine what infill density is most appropriate. Nevertheless, the data collected can hopefully provide a useful reference in designing and manufacturing 3D-printed PLA composites.

Cone crushers are used in both the construction and mining industries for the production of aggregates and extraction of ores. Aggregates are used when building for example houses, roads and railways, hence the cone crushers are a vital part of modern society. To ensure the performance of the cone crusher, it is important to properly adjust the feeding conditions. Using computational methods to virtually analyze the performance of the crushers is a more time and cost efficient solution compared to physical testing. This thesis was divided into two parts, where the main objective of the first part was to use the discrete element method (DEM) to analyze the segregation in cone crushers. Three different methods were developed, which later were utilized to compare the segregation for four different feeding conditions. Two of the analysis methods only considered the segregation in the feed hopper, whilst the third method aimed to give an understanding ofthe segregation inside the crushing chamber. The two first methods could successfully be used to compare how segregated the feed material was for the four feeding conditions, however, the third method proved to be both hard to validate and highly dependent on proper material flow inside the crushing chamber. The main objective during the second part of the thesis was to investigate the possibility of running the DEM simulations coupled to a multibody simulation (MBS) software. The simulation routine was then used to compare the foundation loads for the same four feeding conditions as in the first part. The subframe was later modeled as a flexible body to analyze and compare the stresses the subframe was subject to during operation for the same four feeding conditions. Setting up and running the coupled simulation was successful. Different simulation settings were tested, anda general guideline on how those settings should be defined was set up. The actual impact the coupling had on the foundation loads and stresses in the subframe was however almost non-existent. This could probably be directly related to the fact that the crushing forces in EDEM are known to be many times smaller than what they have been measured to in experiments. This also meant that changing the feeding conditions to alter the segregation did not have a noticeable effect on the results.

This dissertation will focus on the forecasting and classification of time series. Specifically, the forecasting problem will focus on imbalanced time series (ITS) which contain a mix of a mix of low probability extreme observations and high probability normal observations. Two approaches are proposed to improve the forecasting of ITS. In the first approach proposed in chapter 2, an ITS will be modelled as a composition of normal and extreme observations, the input predictor variables and the associated forecast output will be combined into moving blocks, and the blocks will be categorized as extreme event (EE) or normal event (NE) blocks. Imbalance will be decreased by oversampling the minority EE blocks and undersampling the majority NE blocks using modifications of block bootstrapping and synthetic minority oversampling technique (SMOTE). Convolution neural networks (CNNs) and long-short term memory (LSTMs) will be selected for forecast modelling. In the second approach described in chapter 3, which focuses on improving the forecasting accuracies LSTM models, a training strategy called Circular-Shift Circular Epoch Training (CSET), is proposed to preserve the natural ordering of observations in epochs during training without any attempt to balance the extreme and normal observations. The strategy will be universal because it could be applied to train LSTMs to forecast events in normal time series or in imbalanced time series in exactly the same manner. The CSET strategy will be formulated for both univariate and multivariate time series forecasting. The classification problem will focus on the classification event-related potential neural time series by exploiting information offered by the cone of influence (COI) of the continuous wavelet transform (CWT). The COI is a boundary that is superimposed on the wavelet scalogram to delineate the coefficients that are accurate from those that are inaccurate due to edge effects. The features derived from the inaccurate coefficients are, therefore, unreliable. It is hypothesized that the classifier performance would improve if unreliable features, which are outside the COI, are zeroed out, and the performance would improve even further if those features are cropped out completely. Two CNN multidomain models will be introduced to fuse the multichannel Z-scalograms and the V-scalograms. In the first multidomain model, referred to as the Z-CuboidNet, the input to the CNN will be generated by fusing the Z-scalograms of the multichannel ERPs into a frequency-time-spatial cuboid. In the second multidomain model, referred to as the V-MatrixNet, the CNN input will be formed by fusing the frequency-time vectors of the V-scalograms of the multichannel ERPs into a frequency-time-spatial matrix.

<p>Recent experiments have highlighted the efficacy of post-installed reinforcement in enhancing the capacity of groups of  bonded anchors undergoing concrete breakout failure mode. This technique is particularly useful to enhance the performance of anchorages installed in members of limited dimensions such as beams and columns. This thesis presents the results of corresponding numerical investigations on bonded anchor groups in concrete strengthened with post-installed supplementary reinforcement subjected to tension loads. The study is conducted using the 3D Finite Element (FE) approach. The constitutive law of concrete is the microplane model with relaxed kinematic constraint. The interface between anchor or reinforcement and concrete is modeled with two-node bar elements, which are assigned with corresponding bond-stress slip characteristics. The proposed FE approach is validated against experimental results available in the literature by comparing load-displacement behavior and failure mode. </p> <p>The validation incorporates anchor groups with different configurations of post-installed supplementary reinforcing steel bars. The numerical investigations provide a deeper insight into the detailed behavior of anchor groups with post-installed reinforcement through the visualization of crack patterns, stress flows, and strain development. The results show that the post-installed rebars can lead to a significant increase in the performance of post-installed anchorages, and the load increase depends on the number and arrangement of rebars and the failure mode of the system. </p> <p>Lastly, the thesis presents a parametric study on strengthened anchor groups with post-installed rebars in narrow reinforced concrete (RC) members under various configurations. These simulations mimic anchorages used for seismic retrofitting beam-column joints in RC structures using a fully fastened haunch retrofit solution. Due to the limited width and depth of beams and columns, the capacity of the anchorages is often the weakest link in such retrofitting methods. The results from the FE study indicate that the post-installed supplementary reinforcement can be an efficient solution for upgrading the performance of post-installed anchorages in such retrofitting techniques.</p>