This study focuses on the effect of alkali adsorption on the agglomeration of particles of bauxite, kaolinite, emathlite, lime, and two types of coal ash. An agglomeration (adhesion) temperature is defined which characterizes the adhesion propensity of particles. Using a small fluidized bed, a unique experimental technique is developed to measure this agglomeration point in-situ. The effects of alkali adsorption on the agglomeration characteristics of the substrates are determined. The agglomeration temperature of all substrates decreases as the alkali content increases. At low alkali loadings, alkali adsorption enhances particle agglomeration by forming new compounds of lower melting points. At high alkali concentrations, adhesion and agglomeration are caused by a layer of molten alkali which covers the exterior of the particles. Alkali surface composition of particles is studied using a Scanning Auger Microprobe (SAM). Results indicate that the alkali surface concentration decreases as agglomeration temperature increases. SAM depth profiling data provides information on the variations of alkali loading across particles. These results show that an alkali surface product layer is formed where most of the alkali adsorbed is concentrated. The use of additives to scavenge alkali vapors is further studied in a pilot scale downflow combustor under more typical combustion conditions. SAM surface analyses of additive particles indicate three mechanisms of alkali capture. Alkali adsorption by reaction, alkali surface condensation, and alkali nucleation and coagulation with additive particles. These mechanisms may occur independently or simultaneously depending primarily on the alkali vapor concentration and the temperature profile along the combustion furnace. A mathematical model is developed to represent the kinetics and mechanisms of the alkali adsorption and agglomeration process. Modeling results indicate that the adsorption-reaction process is influenced by diffusion of alkali through the surface product layer. The model predictions of the alkali adsorbed as a function of minimum agglomeration temperature agree very well with the experimental results. Alkali-additive interactions in a downflow combustor are also modeled to predict the mechanisms of alkali capture and the overall alkali removal efficiency. Model predictions of the alkali capture agree well with the experimental results.

Manufacturing facilities employ various types of transfer lines and networks with workstations and buffers. This approach promotes the production and fabrication of multicomponent equipments and systems. Analysis of these lines requires the application of discrete time Markov Chain methods. These methods when computerized present certain problems concerning the data storage of large sparse transition matrices. Repetitive multiplication techniques were used to provide the general Markov Chain solution for a series transfer line. These solutions were then computerized to evaluate the series line's availability trajectory. The limiting (leveling off) point for each trajectory provided the steady state availability. From these solutions the work then focuses on the development of new computer algorithms for the series transfer line configuration. These algorithms employ advanced techniques to minimize the storage of large sparse vectors and matices while maintaining relatively fast computational times. The algorithms rely on the line's transition matrix decomposition via graph theoretic methods. A set of library functions written in the C language were specially written to manipulate the Markov Chain matrix and vector data. An extensive set of results were analyzed for the three and four workstation series transfer lines. This analysis employed linear model regression techniques. Results were also collected for the five workstation line. These results show a marked improvement in overall availability when the line's last workstation has a high reliability. In addition, preliminary results indicate that three and four workstation series lines' overall availability are linear combinations of each workstation's availability. Finally, proposed topics for future research are presented in eight major areas. These topics include the development of models for parallel series, series parallel, feedback control, assembly, and disassembly type lines. Also, approximation models and decomposition methods are described in detail.

The influence of discontinuity roughness on the shear strength and deformation behavior of rock joint is analyzed. The study is divided into three parts: laboratory direct shear test on rock samples having different roughness characteristics, characterization of roughness profiles using variogram and probability density distribution and the application of dynamical systems theory to analyze the stability condition of the sliding motion. The relative motion along the rough joint is erratic particularly at a low normal load. A steady motion develops as the normal load increases. The kinematics of translational motion has two distinct characteristics: the translation occurs as a result of a gross and uniform motion (sliding) and/or through localized inhomogeneous motion (slipping). Three modes of volumetric changes are observed during the tangential motion: a dilatant-contractant behavior with the overall volumetric change being strictly dilatant, a dilatant-contractant behavior with the overall volumetric change varying from dilatant to contractant and the strictly contractant behavior. The size of the sheared zones is a function of the distribution of the asperities and of the interface strength. The coefficient of friction decreases as the normal load increases. It may or may not increase when the normal load is decreased. The probability density distribution of the height of the interface asperities is not always Gaussian. The variation of the experimental distribution (histogram) indicates that the asperities are not necessarily sheared off in order of decreasing height but rather on the basis of the condition underlying the existence of contact. The slope of the initial portion of the variogram and the sill, when it exists, are used to characterize the surface morphology of the discontinuity. The lower the slope, the smoother the surface. Two types of anisotropy are observed: geometic anisotropy (elliptic shape) and zonal anisotropy. The rate of collapse of the boundary of the loop describing the roughness of the interface describes the deformation of the discontinuity. The location of the orbit with respect to the stagnation line depends on the normalized stiffness. As the normalized shear stiffness increases, the orbit tends to collapse towards the stagnation axis.

Because of the resolution limitations in remote sensing, the radiance recorded by the detector at each pixel is the integrated sum of the spectral radiance of all materials within the detector instantaneous-field-of-view (IFOV). If the detector IFOV covers more than one object class, the radiance detected is not characteristic of any single class but a mixture of all classes. These mixed pixels will have spectral signatures that fall within the convex hull formed by the signatures of all the classes. Traditional classifiers are therefore usually left with many misclassified or unclassified pixels. To remedy this problem, unmixing algorithms which decompose each pixel into a combination of several classes have been successfully applied to estimate the percentage of each class inside one pixel. In this dissertation, unmixing error of the least squares unmixing algorithm that is caused by the intrinsic data variance, system PSF blurring, detection noise, and band-to-band misregistration is analyzed and evaluated. For high unmixing accuracy, image restoration is proposed to remove the PSF blurring degradation. To objectively assess the restoration performance and expedite the design of our application-oriented restoration scheme, and objective criterion based on the measurement of spectral fidelity in frequency domain is suggested. Based on this criterion, a detailed comparison between the conventional Wiener filter and sampled Wiener filter is conducted, which highlights the significance of sampling aliasing and verifies the results obtained visually by other researchers. Our study shows that contrary to restoration for visual purposes, a partial restoration scheme, instead of full restoration, should be used for a better unmixing performance. Also, the sampling aliasing, which is an artifact and should be suppressed in traditional restoration application, is actually a signal component which needs to be restored for unmixing. Under fair SNR conditions ($\ge$30dB), the proposed restoration scheme can reduce the total unmixing error up to 40% to 70% depending on the scene complexity.

Super-resolution of imagery is an image restoration problem in which the goal is to recover attenuated spatial frequencies from diffraction-limited images. In this dissertation we consider the limits on super-resolution performance in terms of usable bandwidth of the restored frequency spectrum. Based on a characterization of the spectral extrapolation errors (viz. null objects), we derive an expression for an approximate bound on accurate bandwidth extension for the general class of super-resolution algorithms that incorporate a-priori assumptions of a non-negative, space-limited object. For super-resolution of sampled data, we show that it may be possible to achieve significant bandwidth extrapolation depending on the relationship between sampling rate, optical cutoff frequency, and object extent. Simulation results are presented which substantiate the derived bandwidth extrapolation bounds. Given the existence of oscillatory restoration artifacts, we present several artifact suppression techniques as adjuncts to a Poisson maximum a-posteriori (MAP) super-resolution algorithm developed by Hunt. It is shown that "resolution kernels" can be used to accomplish artifact-free restoration with limited bandwidth extension. To achieve increased bandwidth extension, a smoothness constrained MAP estimator is derived which demonstrates substantial artifact reduction in the restoration of natural scenes. It is shown that the constrained Poisson MAP estimator combines a Poisson image observation model with a specific form of Markov random field object prior. Several simulation results demonstrate the artifact suppression capabilities of the regularized algorithms.

Adaptive optics is a new and growing research area aimed at creating high-quality imagery by correcting aberrations in optical systems caused by turbulence in the earth's atmosphere. This paper concentrates on the basics of physical optics leading into the design of an adaptive optics test bed to study the correction of aberrations using optical beam control. Adaptive optics requires the use of sophisticated optical equipment such as deformable mirrors and wavefront sensors. The experimental portion of the work focuses on using a deformable mirror to control the aberrations in a system using laser light. By using a combination of lenses, deformable mirror, and wavefront sensor, the test bed will correct aberrations induced into a plane wave. In addition, the mechanics and function these components was be explored, setting the building blocks for future studies concerning optical beam control.

The objective of this thesis is to describe the concept development, design, system integration, and operating procedures for the AUDASS II vehicle (Autonomous Docking and Spacecraft Servicing Simulator). The AUDASS II is an improved follow on design of AUDASS I, developed in September of 2002. The purpose of AUDASS II is to simulate a chaser spacecraft autonomously rendezvousing and docking with a target spacecraft for the purpose of conducting fluid transfer. This demonstration involves two vehicles elevated, via air pads, upon a smooth epoxy surface, thus allowing three near frictionless degrees of freedom. The ultimate goal of this thesis is to fabricate a vehicle and requisite documentation that will allow future students to conduct experiments using different control algorithms and/or sensors to conduct autonomous rendezvous and docking maneuvers.

"For acoustic frequencies of interest in the development of undersea wireless wide-area networks, wind noise is the dominant contribution to channel noise. The large dynamic range of wind noise forces the network designer to consider wide variations in link margin and effective node-to-node range. Previous reported correlations of acoustic communication performance and wind speed led to multiple hypotheses explaining degradation, including noise variability as well as wind-driven sea-surface effects such as roughness, entrained bubbles, mixing, and stratification. This theses examines operations of a 40-node wide-area acoustic network in varying noise conditions. This data set is unique in that the environment is strongly downward refracting. Based on theoretical models, wind noise is assumed to dominate all other noise sources. Considerations include dependence of link margin and range on wind speed. Although the experiment was not designed to examine the correlation between transmission range and wind speed, a weak correlation was observed in the limited data set available." p. i.

<p> The rapid growth in additive manufacturing technologies have brought various optimization techniques and methodologies to improve each phase that needs to be integrated and analyzed on system level to optimize the system performance. The challenges and limitations of each phase affect the system when integrated as a whole - creating a complex manufacturing environment that needs to be critically examined and managed. To have a better management of complex, emergent, and uncertain manufacturing system from design to recycling phase, a new way of thinking based on more holistic approach is necessary. In this paper, the system of systems paradigm (SoS) is introduced to treat additive manufacturing system as a whole and to present some SoS approaches that are based on holistic thinking. This paper provides a conceptual knowledge of SoS approach using systems principles, laws and approach emphasizing the characteristics and attributes of complex manufacturing system to the AM domain.</p>

<p> This project presents simulation results on the connectivity, data transfer protocols, reliability, and recovery from failure between two Local Area Networks (LANs), which can be at two geographically distant locations, and are connected through a Wide Area Network (WAN). Connectivity within the WAN is provided by the Border Gateway Protocol (BGP) whereas communication within each of the LANs is based on Enhanced Interior Gateway Routing Protocol (EIGRP). Simulation results show that the network is successful in: (1) establishing communication between the LANs using the BGP protocol, and (2) providing recovery from failure through redundancy by using two BGP links within the WAN and a third Internet Protocol Security (IPsec) link. If one of the BGP link fails, the second one will take over, and if the second link fails, the traffic automatically shifts over to the Internet Protocol Security (IPsec) link.</p>