The ability of radiation to cause transient faults in space borne as well as ground based computers is well known. with the density of VLSI circuits increasing every year, the probability of an upset by radiation is becoming more likely. However, research in this area has matured over the last decade, and the mechanisms which cause such faults are better understood. This understanding enables us to propose ideas to eliminate or lessen the effects of radiation on VLSI circuits. Most of the research to date has concentrated on the effect of transient faults on flip-flops rather than combinational logic. This is due to several reasons. First, transient faults, also known as Single Event Upsets (SEU), were first observed in memory circuits located on board satellites. Second, an SEU can leave a lasting effect on a circuit if it occurs in a flip-flop, and third, SEUs can cause the output of a flip-flop to change state more easily if it occurs directly in the flip-flop rather than in the combinational logic. In combinational logic, the node struck by the radiation is completely disjoint from the flip-flops output node. This in effect causes the SEU to satisfy more criteria in order to change the flip-flops output state. The criteria that the SEU must satisfy tend to be complex, and this complexity has caused many researchers to believe that SEUs that occur in combinational logic cause negligible errors in the state of flip-flops. Thus, in this thesis, the criteria for latching a SEU are discussed, and original methods are presented that can be used to determine the probability of an SEU occurring at any node in a circuit will cause a change in the output state of a flip-flop. The methods are then incorporated into a program, named SUPER II, that is able to evaluate the circuit to determine the nodes with the highest probability of having a SEU error latched. The results from the program show that SEUs that occur in combinational logic can have a significant probability of becoming latched. / Master of Science

Traits provide an informative approach to examine species-environment interactions. Often, species-by-species approaches are inefficient to generate generalizable ecological relationships and do not predict species responses to environmental changes based on specific traits species possess. Multiple lines of inquiry and multi-scale approaches are best for assessing environment-trait responses. This thesis examines important questions not specifically addressed before in traits-based research. Chapter one explores biological and ecological trait associations incorporating ontogenetic diet shifts for New River fishes. Niche shift analysis as a chapter one sub-objective quantitatively support where species-specific diet shifts likely occur. Strong biological-ecological trait associations, some intuitive and others not so intuitive, were found that relate biological structure to ecological function. Improved understanding of trait associations, including what factors influence others, supports inference of ecology of fishes. Chapters two and three examine spatial and intraspecific trait variability. Chapter two specifically examines large-scale life history trait variability along latitudinal gradients for twelve widely distributed fish species, including directionality of trait variation, and hypothesizing how optimal traits change with large-scale environmental factors. Strong positive and negative patterns found include average total length of newly hatched larvae, average total length at maturation, average spawning temperature, average egg diameter, and maximum length. These five traits are correlated with other adaptive attributes (i.e. growth rate, reproductive output, and longevity/population turnover rate). In contrast to latitudinal scale, Chapter three examines trait variability of white sucker (Catostomus commersonii) and fantail darter (Etheostoma flabellare) as a function of small-watershed scale spatial factors and anthropogenic disturbance. Toms Creek and Chestnut Creek white sucker and fantail darter displayed positive response to disturbance, contrary to past studies. Lower resource competition, and / or competitive exclusion of fishes with similar niche requirements are possible mechanisms. All three objectives support understanding of trait association and variability as a useful foundation in ecological applications and for formulating plans for conservation and management of species. / Master of Science

Modeling is an essential step in the design of digital circuits [71. The coding of behavioral models for complex devices is a labor intensive task. Even with the use of a to01like the "Modeler's Assistant" [4}, the development of behavioral models is time consuming and labor intensive. The use of re-usable code along with a tool like the Modeler's Assistant can speed up model development. This thesis defines a set of higher level primitives which can be used for this purpose. These primitives are built as a macro library into the tool. The Modeler's Assistant together with the modeling primitives provides us with a tool that can simplify the process of model development. / Master of Science

Green computing, an emerging field of research that seeks to reduce excess power consumption in high performance computing (HPC), is gaining popularity among researchers. Research in this field often relies on simulation or only uses a small cluster, typically 8 or 16 nodes, because of the lack of hardware support. In contrast, System G at Virginia Tech is a 2592 processor supercomputer equipped with power aware components suitable for large scale green computing research. DIRECT is a deterministic global optimization algorithm, implemented in the mathematical software package VTDIRECT95. This thesis explores the potential energy savings for the parallel implementation of DIRECT, called pVTdirect, when used with a large scale computational biology application, parameter estimation for a budding yeast cell cycle model, on System G. Two power aware approaches for pVTdirect are developed and compared against the CPUSPEED power saving system tool. The results show that knowledge of the parallel workload of the underlying application is beneficial for power management. / Master of Science

In the past two decades, there have been significant advances in the development of self-centering (SC) seismic force resisting systems. However, examples of SC systems used in practice are limited due to unusual field construction practices, high initial cost premiums and deformation incompatibility with the gravity framing. A self-centering beam moment frame (SCB-MF) has been developed that virtually eliminates residual drifts and concentrates the majority of structural damage in replaceable fuse elements. The SCB consists of a I-shaped steel beam augmented with a restoring force mechanism attached to the bottom flange and can be shop fabricated. Additionally, the SCB has been designed to eliminate the deformation incompatibility associated with the self-centering mechanism. The SCB-MF system is investigated and developed through analytical, computational, and experimental means. The first phase of the work involves the development of the SCB concepts and the experimental program on five two-thirds scale SCB specimens. Key parameters were varied to investigate their effect on global system hysteretic response and their effect on system components. These large-scale experiments validated the performance of the system, allowed the investigation of detailing and construction methods, provided information on the behavior of the individual components of the system. The experimental results also provided data to confirm and calibrate computational models that can capable of capturing the salient features of the SCB-MF response on global and component level. As a part of the second phase, a set of archetype buildings was designed using the self-centering beam moment frame (SCB-MF) to conduct a non-linear response history study. The study was conducted on a set of 9 archetype buildings. Four, twelve and twenty story frames, each with three levels of self-centering ratios representing partial and fully self-centering systems, were subjected to 44 ground motions scaled to two hazard levels. This study evaluated the performance of SCB-MFs in multi-story structures and investigated the probabilities of reaching limit states for earthquake events with varying recurrence period. The experimental and computational studies described in this dissertation demonstrate that the SCB-MF for steel-framed buildings can satisfy the performance goals of virtually eliminating residual drift and concentrating structural damage in replaceable fuses even during large earthquakes. / Ph. D.

Simulation, design, and control of dynamical systems play an important role in numerous scientific and industrial tasks. The need for detailed models leads to large-scale dynamical systems, posing tremendous computational difficulties when employed in numerical simulations. In order to overcome these challenges, we perform model reduction, replacing the large-scale dynamics with high-fidelity reduced representations. There exist a plethora of methods for reduced order modeling of linear systems, including the Iterative Rational Krylov Algorithm (IRKA), Balanced Truncation (BT), and Hankel Norm Approximation. However, these methods generally target stable systems and the approximation is performed over an infinite time horizon. If we are interested in a finite horizon reduced model, we utilize techniques such as Time-limited Balanced Truncation (TLBT) and Proper Orthogonal Decomposition (POD). In this dissertation we establish interpolation-based optimality conditions over a finite horizon and develop an algorithm, Finite Horizon IRKA (FHIRKA), that produces a locally optimal reduced model on a specified time-interval. Nonetheless, the quantities being interpolated and the interpolant are not the same as in the infinite horizon case. Numerical experiments comparing FHIRKA to other algorithms further support our theoretical results. Next, we discuss model reduction for nonlinear dynamical systems. For models with unstructured nonlinearities, POD is the method of choice. However, POD is input dependent and not optimal with respect to the output. Thus, we use operator splitting to integrate the best features of system theoretic approaches with trajectory based methods such as POD in order to mitigate the effect of the control inputs for the approximation of nonlinear dynamical systems. We reduce the linear terms with system theoretic methods and the nonlinear terms terms via POD. Evolving the linear and nonlinear terms separately yields the reduced operator splitting solution. We present an error analysis for this method, as well as numerical results that illustrate the effectiveness of our approach. While in this dissertation we only pursue the splitting of linear and nonlinear terms, this approach can be implemented with Quadratic Bilinear IRKA or Balanced Truncation for Quadratic Bilinear systems to further diminish the input dependence of the reduced order modeling. / Doctor of Philosophy / Simulation, design, and control of dynamical systems play an important role in numerous scientific and industrial tasks such as signal propagation in the nervous system, heat dissipation, electrical circuits and semiconductor devices, synthesis of interconnects, prediction of major weather events, spread of fires, fluid dynamics, machine learning, and many other applications. The need for detailed models leads to large-scale dynamical systems, posing tremendous computational difficulties when applied in numerical simulations. In order to overcome these challenges, we perform model reduction, replacing the large-scale dynamics with high-fidelity reduced representations. Reduced order modeling helps us to avoid the outstanding burden on computational resources. Numerous model reduction techniques exist for linear models over an infinite horizon. However, in practice we usually are interested in reducing a model over a specific time interval. In this dissertation, given a reduced order, we present a method that finds the best local approximation of a dynamical system over a finite horizon. We present both theoretical and numerical evidence that supports the proposed method. We also develop an algorithm that integrates operator splitting with model reduction to solve nonlinear models more efficiently while preserving a high level of accuracy.

No description available.

Real-time image processing involves processing a wide spectrum of algorithms on huge data sets. Processing at the pixel data rate demands more powerful parallel machines than those developed for conventional image processing. This research takes advantage of current VLSI technology to examine a new approach for processing arbitrary algorithms at real-time data rate. It is based on embedding the algorithms, expressed by their dependency graphs, into two dimensional regularly connected processing arrays. Each node in a graph represents an operation which can be processed by an individual processor in the array. The embedding is performed such that data can be processed in a pipeline fashion as they are received. The result is a machine which exploits functional parallelism and data pipelining simultaneously. The presentation is divided into three parts: the first discusses graphical representation for general image processing algorithms, taking into account the nature of the data flow in real-time systems. The conditions for pipelining the processing of the graph are derived. Next the logical design of a class of VLSI arrays is considered. These arrays can be configured to embed arbitrary problem graphs. The discussion involves the architecture of the array, the architecture of its processing elements and an efficient programming scheme. Finally, static embedding of the dependency graphs into the proposed array is considered. Lower and upper bounds on the area needed to embed any graph are found. Three heuristic procedures to embed the graph at minimum cost are developed, implemented and tested. / Ph. D. / incomplete_metadata

Partial scan has served as an alternative solution for test generation for sequential circuits. As only a portion of flip-flops are incorporated into a scan chain in partial scan design, scan flip-flop selection constitutes a key procedure in partial scan. In this thesis, we propose a new way of selecting scan flip-flops, the Extended Tracking Algorithm (ETA). ETA is a test generation based method and aims to find the conditions that can lead to the detection of as many aborted faults as possible. The faults aborted by a sequential automatic test pattern generator (ATPG) are targeted in ETA and the requirements for the detection of the faults are used for the selection of scan flip-flops. The Extended Tracking Algorithm is realized in two different algorithms, optimal and heuristic, depending on the objectives. The optimal algorithm guarantees the minimal set of flip-flops for the detection of all of the aborted faults in a given circuit, but it has exponential worst case complexity. The heuristic algorithm, on the other hand, obtains a near optimal solution in shorter time. ETA provides a spectrum of accurate fault efficiency and/or fault coverage so that the designer can choose an affordable option. The method is simple and compatible with other scan flip-flop selection approaches. We implemented the Extended Tracking Algorithm in a program called BELLONA. Experiments have been conducted on ISCAS89 benchmark circuits with different specifications. Our experimental results show that ETA is an efficient solution for partial scan deign and only a small portion of scan flip-flops are necessary to obtain extremely high fault efficiency. / M.S.

A novel shear wall system for building structures has been developed that improves upon the performance of conventional steel plate shear walls by mitigating buckling. The new structural system, called the Ring Shaped - Steel Plate Shear Wall, was investigated and developed through experimental and computational methods. First, the plastic mechanism of the system was numerically derived and then analytically validated with finite element analyses. Next, five large-scale, quasi-static, cyclic experimental tests were conducted in the Thomas M. Murray Structures Laboratory at Virginia Tech. The large-scale experiments validated the system performance and provided data on the boundary frame forces, infill panel shear deformation modes, buckling mode shapes, and buckling magnitudes. Multiple computational modeling techniques were employed to reproduce different facets of the system behavior. First, detailed finite element models were constructed to accurately reproduce the cyclic performance, yielding pattern, and buckling mode shapes. The refined finite element models were utilized to further study the boundary element forces and ultra-low cycle fatigue behavior of the system. Second, reduced-order computational models were constructed that can accurately reproduce the hysteretic performance of the web plates. The reduced-order models were then utilized to study the nonlinear response history behavior of four prototype building structures using Ring Shaped - Steel Plate Shear Walls and conventional steel plate shear walls. The nonlinear response history analyses investigated the application of the system to a short period and a long period building configuration. In total 176 nonlinear response history analyses were conducted and statistically analyzed. Lastly, a practical design methodology for the Ring Shaped - Steel Plate Shear Wall web plates was presented. The experimental tests and computational simulations reported in this dissertation demonstrate that Ring Shaped - Steel Plate Shear Walls are capable of improving seismic performance of buildings by drastically reducing buckling and improving cyclic energy dissipation. / Ph. D. / A novel shear wall system for building structures has been developed that improves the performance of of buildings subjected to seismic loads. The new structural system, called the Ring Shaped - Steel Plate Shear Wall, was investigated and developed through experimental and computational methods. Five large-scale, cyclic experimental tests were conducted in the Thomas M. Murray Structures Laboratory at Virginia Tech. The large-scale experiments validated the system performance and provided data on the design forces and modes of failure. Multiple modeling techniques were employed to reproduce different facets of the system behavior. Refined finite element models were utilized to further study the system forces and failure modes. Other computational models were constructed to accurately reproduce the cyclic performance of the system. These models were then utilized to study the seismic behavior of four prototype building structures using the Ring Shaped - Steel Plate Shear Walls and conventional steel shear walls. Lastly, a practical design methodology for the Ring Shaped - Steel Plate Shear Wall web plates was presented. The experimental tests and computational simulations reported in this dissertation demonstrate that Ring Shaped - Steel Plate Shear Walls are capable of improving seismic performance of buildings. Additionally, the presented design methodology allows designers and researchers to continue exploring the RS-SPSW system.