Thesis (Ph.D)--Civil and Environmental Engineering, Georgia Institute of Technology, 2009. / Committee Chair: Dr. Leroy Z. Emkin; Committee Co-Chair: Dr. Abdul-Hamid Zureick; Committee Member: Dr. Dewey H. Hodges; Committee Member: Dr. Kenneth M. Will; Committee Member: Dr. Rami M. Haj-ali. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Thesis (Ph.D.)--McMaster University, 2004. / Advisor: Ahmed Ghobarah. Includes bibliographical references (p. 341-351). Also available online.

CAPES / A materialização original do Paradigma Orientado a Notificações (PON) implementada na linguagem de programação C++ possibilitou a criação de aplicações sob o domínio desse paradigma. Apesar de tais contribuições para com o paradigma, o desenvolvimento de aplicações no PON ainda apresenta baixo nível de maturidade e certo nível de dificuldade. Tais dificuldades advêm principalmente da nova forma de estruturar os programas, onde os mesmos seguem um fluxo de notificações, o que difere da programação convencional. Ademais, até o momento, apenas algumas aplicações foram desenvolvidas, com escopos relativamente pequenos, o que não contribui efetivamente para a consolidação do paradigma e, particularmente, para o aprendizado via um conjunto efetivo de exemplos. Neste sentido, primeiramente, a própria estrutura geral do framework foi relida e reapresentada sob o viés de padrões de projeto, assim como houve contribuições na própria implementação desse. Subsequentemente, este trabalho teve como objetivo principal de estudo o nortear do desenvolvimento de aplicações baseadas no PON de maneira mais simplificada e eficiente, usando os avanços aqui citados. Isso se deu particularmente pela apresentação de padrões de desenvolvimento de software voltados para esse paradigma e infraestrutura, os quais buscam uma implementação mais purista envolvendo quesitos como redução do uso de implementações multiparadigmas, simplificação no desenvolvimento, garantia de determinismo, entre outras. De modo geral, as novas funcionalidades/recursos para a concepção de aplicações através do Framework PON facilitaram e melhoraram seu uso. / The original materialization of Notification Oriented Paradigm (NOP) previously implemented in the C++ programming language allowed the creation of applications under the domain of this paradigm. Although these contributions to the paradigm, the development of NOP applications still has a low level of maturity and a certain degree of difficulty. Such difficulties arise essentially from the new form of structuring programs, in which such applications follow a notification flow, which differs from the conventional programming. Moreover, so far, only a few applications have been developed, with relatively small scope, which does not contribute effectively to the consolidation of the paradigm, and particularly for learning through an effective set of examples. Henceforth, firstly, the general structure of the framework has been reread and reintroduced under the bias of design patterns, even as there were contributions in its implementation. Subsequently, the main goal of this paper is to guide the development of applications based on NOP in a more streamlined and efficient manner, using the advances mentioned here. This was achieved particularly by the presentation of technical software implementations aimed to this paradigm and its infrastructure, which pursues a manner to conceive more purist implementations, involving issues such as reducing the use of multiparadigms implementations, simplifying the development, ensuring determinism, and so far. In general, the new features for conceiving applications based on the NOP Framework made easier and better its use.

CAPES / A materialização original do Paradigma Orientado a Notificações (PON) implementada na linguagem de programação C++ possibilitou a criação de aplicações sob o domínio desse paradigma. Apesar de tais contribuições para com o paradigma, o desenvolvimento de aplicações no PON ainda apresenta baixo nível de maturidade e certo nível de dificuldade. Tais dificuldades advêm principalmente da nova forma de estruturar os programas, onde os mesmos seguem um fluxo de notificações, o que difere da programação convencional. Ademais, até o momento, apenas algumas aplicações foram desenvolvidas, com escopos relativamente pequenos, o que não contribui efetivamente para a consolidação do paradigma e, particularmente, para o aprendizado via um conjunto efetivo de exemplos. Neste sentido, primeiramente, a própria estrutura geral do framework foi relida e reapresentada sob o viés de padrões de projeto, assim como houve contribuições na própria implementação desse. Subsequentemente, este trabalho teve como objetivo principal de estudo o nortear do desenvolvimento de aplicações baseadas no PON de maneira mais simplificada e eficiente, usando os avanços aqui citados. Isso se deu particularmente pela apresentação de padrões de desenvolvimento de software voltados para esse paradigma e infraestrutura, os quais buscam uma implementação mais purista envolvendo quesitos como redução do uso de implementações multiparadigmas, simplificação no desenvolvimento, garantia de determinismo, entre outras. De modo geral, as novas funcionalidades/recursos para a concepção de aplicações através do Framework PON facilitaram e melhoraram seu uso. / The original materialization of Notification Oriented Paradigm (NOP) previously implemented in the C++ programming language allowed the creation of applications under the domain of this paradigm. Although these contributions to the paradigm, the development of NOP applications still has a low level of maturity and a certain degree of difficulty. Such difficulties arise essentially from the new form of structuring programs, in which such applications follow a notification flow, which differs from the conventional programming. Moreover, so far, only a few applications have been developed, with relatively small scope, which does not contribute effectively to the consolidation of the paradigm, and particularly for learning through an effective set of examples. Henceforth, firstly, the general structure of the framework has been reread and reintroduced under the bias of design patterns, even as there were contributions in its implementation. Subsequently, the main goal of this paper is to guide the development of applications based on NOP in a more streamlined and efficient manner, using the advances mentioned here. This was achieved particularly by the presentation of technical software implementations aimed to this paradigm and its infrastructure, which pursues a manner to conceive more purist implementations, involving issues such as reducing the use of multiparadigms implementations, simplifying the development, ensuring determinism, and so far. In general, the new features for conceiving applications based on the NOP Framework made easier and better its use.

This investigation was undertaken to develop plastic design aids to be used in the direct design of optimum frames. It uses the concept of minimum weight of plastically designed steel frames, and the concept of linear programming to obtain general solutions. Among the special characteristics of this study are: A. The integration of both gravity and combined loading conditions into one linear programming problem. B. The application of the revised simplex method to the dual of a parametric original problem. C. The application of A and B above in the development of design aids for the optimum design of symmetrical single-bay, single-story portal frame. Specifically, design graphs for different height to span ratios and different vertical load to lateral load ratios are developed. The use of these graphs does not require the knowledge of linear programming or computers on the part of the designer.

In this investigation, the primary objective was to study the nonlinear behavior of unbraced two-bay concrete frames and to determine the extent to which ultimate load theory or limit design can be applied to these structures. The frame behavior was investigated analytically by two methods. In the first method the frame stability equation was derived assuming that members of the frame possess an elasto-plastic moment-curvature relationship. This stability analysis was also carried out by another model consisting of a column attached to a linear spring and carrying the total frame load. The second method was through a computer program which took material and geometric nonlinearities of concrete frames into account. A model concrete frame, with a scale factor of approximately one-third was considered. Variable parameters were loading condition, column reinforcement ratio, and beam to column load ratio. For each frame, the gravity loads were increased proportionally until 75% of the frame ultimate capacity under gravity loads was reached. Then; while these gravity loads were held constant, lateral load was applied and increased to failure. The overall geometry, 21-in high columns and 84-in long beam, were kept the same for all of model frames investigated. The computer study and the stability model analysis indicated that all frames remained stable until four plastic hinges (two in each bay) formed, thus producing a combined sway mechanism. Based on the scope of this study, it appears that limit design may be employed for unbraced reinforced concrete structures.

A well structured computer program has been developed to implement and compare two models for geometrically nonlinear finite element analysis of plane frames. One model is developed using classical beam-column theory, and the other is formulated using standard finite element techniques. The beam-column model is shown to converge to a more accurate equilibrium path using less elements to model a structure than the second model for frames. The models are also compared on the basis of their limitations and on the basis of the degree of difficulty for their implementation. The modified Riks/Wempner method is used to perform postbuckling analysis and to study the effect of varying the composition of the tangent stiffness matrix on its ability to detect instability. The importance of including all contributions in the incrementation of a model derived using convected coordinates is also studied. / M.S.

Scale model technology represents one method of investigating the behavior of advanced, weight-efficient composite structures under a variety of loading conditions. Testing of scale models can provide a cost effective alternative to destructive testing of expensive composite prototypes and can be used to verify predictions obtained from finite element analyses. It is necessary, however, to understand the limitations involved in testing scale model structures before the technique can be fully utilized. The objective of this research is to characterize these limitations, or scaling effects, in the large deflection response and failure of composite beams. Scale model beams were loaded with an eccentric axial compressive load designed to produce large bending deflections and global failure. / Ph. D.

This study compares energy minimization with direct stiffness for linear structural analysis. The energy minimization approach locates the generalized displacement vector by minimizing the total potential energy of the structure being analyzed. From the survey of variable metric and conjugate gradient algorithms included in this study, the Davidon-Fletcher-Powell variable metric algorithm and the FletcherReeves conjugate gradient algorithm were chosen to minimize the total potential energy. A description of both algorithms is presented. The direct stiffness method assembles the equilibrium equations of the structure being analyzed. These equations are solved by Gaussian elimination to determine the generalized displacement vector. Computer codes have been written for the energy minimization and direct stiffness methods. The comparison was based on computational effort, in terms of computer time, required for analysis. The results of this study show energy minimization is not competitive with direct stiffness for linear structural analysis. As the problem size increases by degree of freedom the direct stiffness method rapidly increases in superiority over the energy minimization method. / Master of Science

A study was performed to develop an interactive plane frame analysis program. The program was designed to run on Unix-based workstations and to be independent of the hardware platform. This was achieved by selecting the XWindow system as the underlying graphical user-interface. An event-driven, window-based program was developed using the XWindow Toolkit. The entire program was written in the C programming language. The program consists of two parts: a preprocessor and a processor. The preprocessor employs pull-down menus and pop-up dialog boxes and provides a convenient way of creating and modifying structural models. The processor performs linear elastic analysis of plane frame structures based on the matrix displacement method. The processor computes joint displacements, support reactions and local member-end forces of the structure. It also handles internal hinges, joint loads, member loads, prescribed displacements, multiple load cases and load combinations. Other features include automatic mesh generation for orthogonal frames, automatic computation of self-weight and specification of output options. To verify the accuracy of the program, analysis results from the program were compared with results obtained from a commercial structural analysis program. A comparison of joint displacements, support reactions and member-end forces indicate that the difference in the values obtained from the two programs is less than two percent and that the program provides accurate analysis results. / Master of Science