Thesis (M. S.)--Mechanical Engineering, Georgia Institute of Technology, 2009. / Committee Chair: Singhose, William; Committee Member: Costello, Mark; Committee Member: Whiteman, Wayne

Orientador: Renato Pavanello / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-18T00:16:08Z (GMT). No. of bitstreams: 1 Bispo_RafaelSantana_M.pdf: 3924997 bytes, checksum: eedf2e79312457d2dfe0c4ce2418dfe4 (MD5) Previous issue date: 2011 / Resumo: Atualmente, existe uma grande tendência no incremento da produção de energia elétrica através de fontes renováveis. Em especial, a geração de energia elétrica produzida através de parques eólicos tem sido bastante adotada. O projeto desses equipamentos envolve a modelagem dinâmica acoplada solo-fluido-estrutura que pode ser estudada usando-se a formulação particionada, onde o problema da interação entre os meios é tratado de maneira iterativa. Nesse tipo de técnica, é possível que modelos fisicamente heterogêneos, chamados de partições, possam utilizar diferentes técnicas de discretização, como por exemplo o domínio do fluido ser baseado em uma formulação de Elementos de Contorno e o domínio estrutura baseado em uma formulação em Elementos Finitos. Neste trabalho, é realizado um estudo dinâmico de turbinas eólicas, utilizando tratamento particionado e Multiplicadores de Lagrange afim de se obter as frequências características e as curvas de resposta em frequência do sistema em análise. A discretização do problema é realizada através do Método dos Elementos Finitos (FEM) utilizando elemento de pórtico e quadrilateral de Wilson. Desta forma, a resolução de problemas de interação, utilizando a formulação particionada, é estudada com a finalidade de avaliar a convergência e a viabilidade da técnica em problemas harmônicos estruturais / Abstract: Currently, there is a great tendency in increasing the production of electricity through renewable sources. In this context, the generation of electric energy produced by wind farms has been widely adopted. The design of these devices involves the dynamic modeling of coupling fluid-structure-soil that can be studied using the partitioned formulation, where the problem of interaction between the parties is iterative manner. In this type of technique, it is possible that physically heterogeneous models, called partitions, can use different discretization techniques, such as the domain of fluid is based on a formulation of boundary element or based on a finite element formulation. In this paper a harmonic simulation of wind turbines, using partitioned treatment and Lagrange multipliers is studied in order to obtain the characteristic frequencies and frequency response function of the system under analysis. The discretization of the problem is performed using the Finite Element Method (FEM), as well as beam elements and the quadrilateral Wilson element. Thus, the resolution of elastodynamics problems, using the partitioned formulation is studied with the aim of assessing the feasibility and convergence of this technique, applied to dynamic harmonic analysis / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestre em Engenharia Mecânica

Orientadores: Vera Lucia da Rocha Lopes, Jose Mario Martinez / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica / Made available in DSpace on 2018-08-06T05:45:54Z (GMT). No. of bitstreams: 1 Mendonca_LuzianeFerreirade_D.pdf: 3091539 bytes, checksum: 7c40c0932b2056dbe8dfb8e1f7da8401 (MD5) Previous issue date: 2006 / Resumo: Sistemas de Otimalidade (ou Sistemas KKT) são sistemas formados pelas condições primais-duais estacionárias para a solução de problemas de otimização. Sob hipóteses adequadas (condições de qualificação), os minimizadores locais de um problema de minimização satisfarão as equações e inequações KKT; entretanto, infelizmente, muitos outros pontos estacionários (incluindo maximizadores) também são soluções desse sistema não linear. Por essa razão, os métodos destinados à resolução de problemas de programação não-linear fazem uso constante da estrutura de minimização, e o uso simples de métodos destinados à resolução de sistemas não-lineares podem gerar soluções espúrias. Todavia, caso o método destinado à resolução do sistema KKT tenha um ponto inicial situado na região de atração para um minimizador, esse método pode vir a ser muito eficiente. Neste trabalho, os métodos quase-Newton para a resolução de sistemas não-lineares são usados como aceleradores de algoritmos de programação não-linear (Lagrangiano Aumentado) com restrições de igualdade, desigualdade e caixa. Utilizamos como acelerador o método simétrico inverso de correção de posto um (ISR1), o qual realiza reínicios periódicos e faz uso das estruturas esparsas das matrizes para armazenamento. São demonstrados resultados de convergência e são realizados vários experimentos numéricos que comprovam a eficiência desta estratégia para problemas de minimização com restrições de igualdade, e indicam outros caminhos para problemas de minimização com restrições gerais (igualdade, desigualdade e caixa) / Abstract: Optimality (or KKT) systems arise as primal-dual stationarity conditions for constrained optimization problems. Under suitable constraint qualifications, local minimizers satisfy KKT equations but, unfortunately, many other stationary points (including, perhaps, maximizers) may solve these nonlinear systems too. For this reason, nonlinear-programming solvers make strong use of the minimization structure and the naive use of nonlinear-system solvers in optimization may lead to spurious solutions. Nevertheless, in the basin of attraction of a minimizer, nonlinear-system solvers may be quite efficient. In this work quasi-Newton methods for solving nonlinear systems are used as accelerators of nonlinear-programming (augmented Lagrangian) algorithms. A periodically-restarted memoryless symmetric rank-one (SRI) correction method is introduced for that purpose. Convergence results are given. For problems with only equality constraints, numerical experiments that confirm that the acceleration is effective are presented. A bunch of problems with equalities, inequalities and box constraints is tested and several comments and suggestions for further work are presented / Doutorado / Doutor em Matemática Aplicada

Maxwell's equations for the electromagnetic field are symmetrized by introducing magnetic charges into the formalism of electrodynamics. The symmetrized equations are solved for the fields and potentials of point particles. Those potentials, some of which are found to be singular along a line, are used to formulate the Lagrangian for a system of two dyons (particles with both electric and magnetic charge). The equations of motion are derived from the Lagrangian. It is shown that the dimensionality constants k and k * , which we r e introduced to define the units of the electromagnetic fields, have to be equal in order to avoid center of mass acceleration in the two dyon system.

Thesis. 1976. Ph.D.--Massachusetts Institute of Technology. Dept. of Earth and Planetary Sciences. / Microfiche copy available in Archives and Science. / Bibliography: leaves 122-123. / by Charles A. Moo. / Ph.D.

In this research effort, we consider nondifferentiable optimization (NDO) problems that arise in several applications in science, engineering, and management, as well as in the context of other mathematical programming approaches such as Dantzig-Wolfe decomposition, Benders decomposition, Lagrangian duality, penalty function methods, and minimax problems. The importance and necessity of having effective solution methods for NDO problems has long been recognized by many scientists and engineers. However, the practical use of NDO techniques has been somewhat limited, principally due to the lack of computationally viable procedures, that are also supported by theoretical convergence properties, and are suitable for solving large-scale problems. In this research, we present some new algorithms that are based on popular computationally effective strategies, while at the same time, do not compromise on theoretical convergence issues. First, a new variable target value method (VTVM) is introduced that has an e-convergence property, and that differs from other known methods in that it does not require any prior assumption regarding bounds on an optimum or regarding the solution space. In practice, the step-length is often calculated by using an estimate of the optimal objective function value. For general nondifferentiable optimization problems, however, this may not be readily available. Hence, we design an algorithm that does not assume the possibility of having an a prior estimate of the optimal objective function value. Furthermore, along with this new step-length rule, we present a new subgradient deflection strategy in which a selected subgradient is rotated optimally toward a point that has an objective function value less than the incumbent target value. We also develop another deflection strategy based on Shor’s space dilation algorithm, so that the resulting direction of motion turns out to be a particular convex combination of two successive subgradients and we establish suitable convergence results. In the second part of this dissertation, we consider Lagrangian dual problems. Our motivation here is the inadequacy of the simplex method or even interior point methods to obtain quick, near-optimal solutions to large linear programming relaxations of certain discrete or combinatorial optimization problems. Lagrangian dual methods, on the other hand, are quite well equipped to deal with complicating constraints and ill-conditioning problems. However, available optimization methods for such problems are not very satisfactory, and can stall far from the optimal objective function value for some problems. Also, there is no practical implementation strategy for recovering a primal optimal solution. This is a major shortcoming, even if the method is used only for bounding purposes in the context of a branch and bound scheme. With this motivation, we present new primal convergence theorems that generalize existing results on recovering primal optimal solutions, and we describe a scheme for embedding these results within a practical primal-dual Lagrangian optimization procedure. / Ph. D.

Thesis (PhD (Mathematical Sciences. Applied Mathematics))--University of Stellenbosch, 2008. / This dissertation contains a detailed report on the results of a research project on the behaviour of a dynamical system consisting of a hoop to which a heavy particle is fixed at the rim. This loaded hoop rolls on a rough surface while remaining in the vertical plane. The motion of the hoop consists of various, possibly alternating, phases consisting of rolling without slipping, spinning or skidding motion and in some cases ends by hopping off the surface. A general mathematical model is developed, consisting of a system of second order ordinary differential equations, one for each of the three degrees of freedom. Analytic solutions are obtained in some cases; otherwise numerical solutions are used. Three specific applications of the general model are dealt with. In the first application the problem of massless hoops is investigated. The main emphasis is on the somewhat controversial question of what happens after the normal reaction becomes zero in a position where the particle is still moving downwards. A new result shows that the hoop can continue to move horizontally in a motion defined as skimming. The second application deals with rigid hoops and a large number of detailed results are presented. Classification schemes for the different types of behaviour are introduced and summarised in the form of phase diagrams. Some emphasis is placed on the rather amazing number of different patterns of motion that can be obtained by varying the parameters. In the third application two elastic models are analysed, with the primary purpose of explaining one aspect of the reported behaviour of experimental hoops, namely hopping while the particle is moving downwards. A chapter on experimental models rounds off the project.

published_or_final_version / Civil Engineering / Master / Master of Philosophy

A second-order Lagrangian system is a generalization of a classical mechanical system for which the Lagrangian action depends on the second derivative of the state variable. Recent work has shown that the dynamics of such systems c:an be substantially richer than for classical Lagrangian systems. In particular, topological properties of the planar curves obtained by projection onto the lower-order derivatives play a key role in forcing certain types of dynamics. However, the application of these techniques requires an analytic restriction on the Lagrangian that it satisfy a twist property. In this dissertation we approach this problem from the point of view of curve shortening in an effort to remove the twist condition. In classical curve shortening a family of curves evolves with a velocity which is normal to the curve and proportional to its curvature. The evolution of curves with decreasing action is more general, and in the first part of this dissertation we develop some results for curve shortening flows which shorten lengths with respect to a Finsler metric rather than a Riemannian metric. The second part of this dissertation focuses on analytic methods to accommodate the fact that the Finsler metric for second-order Lagrangian system has singularities. We prove the existence of simple periodic solutions for a general class of systems without requiring the twist condition. Further; our results provide a frame work in which to try to further extend the topological forcing theorems to systems without the twist condition. / Includes bibliography. / Dissertation (Ph.D.)--Florida Atlantic University, 2014. / FAU Electronic Theses and Dissertations Collection

This dissertation consists of three independent sections: In the first part, focusing on the auto industry we look at the challenges and solution strategies of employing build-to-order (BTO) with global supply. We consider some familiar tools for managing domestic supply and exploit them for managing international supply, and propose new methods. We study frequency of supply as a way to improve performance. We study the impact of forecast accuracy, and conclude that improvements there alone may not be sufficient to obtain desired savings. Within this perspective we look at a new shipping policy, 'Ship-to-Average", which prescribes sending a fixed quantity, based on the long term average forecast, with each shipment and making adjustments only if the inventory strays outside a prescribed range. In the second part we look at a Brownian control problem. When a manufacturer places repeated orders with a supplier to meet changing production requirements, he faces the challenge of finding the right balance between holding costs and the operational costs involved in adjusting the shipment sizes. Consider a storage system whose content fluctuates as a Brownian motion in the absence of control. A linear holding cost is incurred continuously. Inventory level can be adjusted by any quantity at a fixed plus proportional cost. We show control band policies are optimal for the average cost problem and calculate the optimal policy parameters. This form of policy is described by three parameters q, Q, S. When the inventory falls to 0 (rises to S), the controller expedites (curtails) shipments to return it to q (Q). Developing techniques based on Lagrangian relaxation we show that this type of policy is optimal even with constraints on the size of adjustments and on the maximum inventory level. The Brownian Control problem can be viewed as an idealization --without delivery delays, of the problem of supplying BTO operations, and provides some theoretical explanation for the Ship-to-Average policies. In fact, Ship-to-Average policies are a practical implementation of Control Band policies in the setting with delivery delays. Finally, we explore the power and applicability of the Lagrangian approach developed in the second part.