• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 90
  • 20
  • 7
  • 5
  • 4
  • 3
  • 2
  • 1
  • 1
  • 1
  • Tagged with
  • 161
  • 161
  • 58
  • 35
  • 35
  • 34
  • 24
  • 23
  • 21
  • 20
  • 18
  • 18
  • 17
  • 16
  • 16
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
61

Application of Genetic Algorithm to a Forced Landing Manoeuvre on Transfer of Training Analysis

Tong, Peter, mail@petertong.com January 2007 (has links)
This study raises some issues for training pilots to fly forced landings and examines the impact that these issues may have on the design of simulators for such training. It focuses on flight trajectories that a pilot of a single-engine general aviation aircraft should fly after engine failure and how pilots can be better simulator trained for this forced landing manoeuvre. A sensitivity study on the effects of errors and an investigation on the effect of tolerances in the aerodynamic parameters as prescribed in the Manual of Criteria for the Qualification of Flight Simulators have on the performance of flight simulators used for pilot training was carried out. It uses a simplified analytical model for the Beech Bonanza model E33A aircraft and a vertical atmospheric turbulence based on the MIL-F-8785C specifications. It was found that the effect of the tolerances is highly sensitive on the nature of the manoeuvre flown and that in some cases, negative transfe r of training may be induced by the tolerances. A forced landing trajectory optimisation was carried out using Genetic Algorithm. The forced landing manoeuvre analyses with pre-selected touchdown locations and pre-selected final headings were carried out for an engine failure at 650 ft AGL for bank angles varying from banking left at 45° to banking right at 45°, and with an aircraft's speed varying from 75.6 mph to 208 mph, corresponding to 5% above airplane's stall speed and airplane's maximum speed respectively. The results show that certain pre-selected touchdown locations are more susceptible to horizontal wind. The results for the forced landing manoeuvre with a pre-selected location show minimal distance error while the quality of the results for the forced landing manoeuvre with a pre-selected location and a final heading show that the results depend on the end constraints. For certain pre-selected touchdown locations and final headings, the airplane may either touchdown very close to the pre-selected touchdown location but with greater final h eading error from the pre-selected final heading or touchdown with minimal final heading error from the pre-selected final heading but further away from the pre-selected touchdown location. Analyses for an obstacle avoidance forced landing manoeuvre were also carried out where an obstacle was intentionally placed in the flight path as found by the GA program developed for without obstacle. The methodology developed successfully found flight paths that will avoid the obstacle and touchdown near the pre-selected location. In some cases, there exist more than one ensemble grouping of flight paths. The distance error depends on both the pre-selected touchdown location and where the obstacle was placed. The distance error tends to increase with the addition of a specific final heading requirement for an obstacle avoidance forced landing manoeuvre. As with the case without specific final heading requirement, there is a trade off between touching down nearer to the pre-selected location and touching down with a smaller final heading error.
62

Frequentist Model Averaging For Functional Logistic Regression Model

Jun, Shi January 2018 (has links)
Frequentist model averaging as a newly emerging approach provides us a way to overcome the uncertainty caused by traditional model selection in estimation. It acknowledges the contribution of multiple models, instead of making inference and prediction purely based on one single model. Functional logistic regression is also a burgeoning method in studying the relationship between functional covariates and a binary response. In this paper, the frequentist model averaging approach is applied to the functional logistic regression model. A simulation study is implemented to compare its performance with model selection. The analysis shows that when conditional probability is taken as the focus parameter, model averaging is superior to model selection based on BIC. When the focus parameter is the intercept and slopes, model selection performs better.
63

Supervisory Control Optimization with Sequential Quadratic Programming for Parallel Hybrid Vehicle with Synchronous Power Sources

January 2017 (has links)
abstract: The thesis covers the development and modeling of the supervisory hybrid controller using two different methods to achieve real-world optimization and power split of a parallel hybrid vehicle with a fixed shaft connecting the Internal Combustion Engine (ICE) and Electric Motor (EM). The first strategy uses a rule based controller to determine modes the vehicle should operate in. This approach is well suited for real-world applications. The second approach uses Sequential Quadratic Programming (SQP) approach in conjunction with an Equivalent Consumption Minimization Strategy (ECMS) strategy to keep the vehicle in the most efficient operating regions. This latter method is able to operate the vehicle in various drive cycles while maintaining the SOC with-in allowed charge sustaining (CS) limits. Further, the overall efficiency of the vehicle for all drive cycles is increased. The limitation here is the that process is computationally expensive; however, with advent of the low cost high performance hardware this method can be used for the hybrid vehicle control. / Dissertation/Thesis / Masters Thesis Engineering 2017
64

The Cost and Benefit of Bulk Energy Storage in the Arizona Power Transmission System

January 2013 (has links)
abstract: This thesis addresses the issue of making an economic case for energy storage in power systems. Bulk energy storage has often been suggested for large scale electric power systems in order to levelize load; store energy when it is inexpensive and discharge energy when it is expensive; potentially defer transmission and generation expansion; and provide for generation reserve margins. As renewable energy resource penetration increases, the uncertainty and variability of wind and solar may be alleviated by bulk energy storage technologies. The quadratic programming function in MATLAB is used to simulate an economic dispatch that includes energy storage. A program is created that utilizes quadratic programming to analyze various cases using a 2010 summer peak load from the Arizona transmission system, part of the Western Electricity Coordinating Council (WECC). The MATLAB program is used first to test the Arizona test bed with a low level of energy storage to study how the storage power limit effects several optimization out-puts such as the system wide operating costs. Very high levels of energy storage are then added to see how high level energy storage affects peak shaving, load factor, and other system applications. Finally, various constraint relaxations are made to analyze why the applications tested eventually approach a constant value. This research illustrates the use of energy storage which helps minimize the system wide generator operating cost by "shaving" energy off of the peak demand. / Dissertation/Thesis / M.S. Electrical Engineering 2013
65

Programação quadratica sequencial e condições de qualificação / Sequential quadratic programming and constraint qualification

Nunes, Fernanda Téles 03 September 2009 (has links)
Orientador: Maria Aparecida Diniz Ehrhardt / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica / Made available in DSpace on 2018-08-13T08:54:50Z (GMT). No. of bitstreams: 1 Nunes_FernandaTeles_M.pdf: 2400651 bytes, checksum: 206dfad35642a33d2de362510094e78d (MD5) Previous issue date: 2009 / Resumo: Abordando problemas de minimização de funções com restrições nos deparamos com as condições de otimalidade e, ainda, com condições de qualificação das res­trições. Nosso interesse é o estudo detalhado de várias condições de qualificação, com destaque para a condição de dependência linear positiva constante, e sua influência na convergência de algoritmos de Programação Quadrática Sequencial. A relevância deste estudo está no fato de que resultados de convergência que têm, em suas hipóteses, condições de qualificação fracas são mais fortes que aqueles baseados em condições de qualificação fortes. Experimentos numéricos serão realizados tanto para investigar a eficiência destes métodos na resolução de problemas com diferentes condições de qualificação, quanto para comparar dois diferentes tipos de busca, monótona e não-monótona. Tentamos confirmar a hipótese de que algoritmos baseados em uma busca não-monótona atuam contra o Efeito: Maratos, de comum ocorrência na resolução de problemas de minimização através de métodos de Programação Quadrática Sequencial. / Abstract: In the context of constrained optimization problems, we face the optimality conditions and also constraint qualification. Our aim is to study with details several constraint qualification, highlighting the constant positive linear dependence condition, and its influence in Sequential Quadratic Programming algorithms convergence. The relevance of this study is in the fact that convergence results having as hypothesis weak constraints qualification are stronger than those based on stronger constraints qualification. Numerical experiments will be done with the purpose of investigating the efficiency of these methods to solve problems with different constraints qualification and to compare two diferent kinds of line search, monotone and nonmonotone. We want to confirm the hypothesis that algorithms based on a nonmonotone line search act against the Maratos Effect, very common while solving minimization problems through Sequential Quadratic Programming methods. / Mestrado / Mestre em Matemática Aplicada
66

Otimização de canteiros : quadriláteros de perímetro constante e área máxima / Optimization of grounds : quadrilaterals of constant perimeter and maximum area

Souza, Marília Franceschinelli de, 1984- 25 August 2018 (has links)
Orientador: Sandra Augusta Santos / Dissertação (mestrado profissional) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-25T19:53:49Z (GMT). No. of bitstreams: 1 Souza_MariliaFranceschinellide_M.pdf: 3298789 bytes, checksum: cd2737cf2f9747a3856c42aee3ac5d83 (MD5) Previous issue date: 2014 / Resumo: O currículo de Matemática do Ensino Médio está atualmente muito denso e quase não permite ao professor explorar outros trabalhos que fujam das aulas expositivas, nas quais o papel do aluno é o de apenas escutar, anotar e reproduzir. Esse esquema antiquado desperta pouco interesse dos alunos pelas disciplinas, especialmente pela Matemática, tida por muitos como a vilã, bastante difícil de ser compreendida. Neste trabalho apresentamos uma proposta de projeto para ser trabalhado no Ensino Médio. O problema de otimização da área de quadriláteros de perímetro constante é abordado, utilizando essencialmente o conteúdo do Ensino Básico. O ponto de partida é um problema simples, presente na maioria dos livros-texto. A análise é ampliada gradativamente por situações mais próximas da realidade, oferecendo ao aluno a oportunidade de utilizar diversos conceitos estudados em uma aplicação da Matemática. Os problemas são abordados tanto de forma algébrica quanto geométrica, oferecendo elementos para que o aluno processe informações, anteveja possibilidades, analise o caso geral, exemplifique situações específicas e de fato possa compreender os problemas e interpretar as soluções obtidas / Abstract: The mathematics curriculum of the Brazilian High School is currently very dense. As a result, it is hard to explore alternative ways of teaching that allow the students to effectively participate in lessons, instead of just listening to the teacher and taking notes. The old fashioned expositive method, in general, does not encourage the interest of the students, especially in Mathematics, considered as a villain by many, because of it is intrinsic difficult. In this work it is presented the proposal of a project for the High School level. The problem of maximizing the area of quadrilaterals with constant perimeter is approached, using essentially the content of Basic Education. The starting point is a simple problem, present in most textbooks. The analysis is extended for situations closer to reality, offering students the opportunity to use many concepts already studied, in an application of mathematics. The problems are treated algebraically and geometrically, providing elements for the student to process information, anticipates possibilities, consider the general case, exemplify specific situations, so that they might indeed understand the problems and interpret the obtained solutions / Mestrado / Matemática em Rede Nacional / Mestra em Matemática em Rede Nacional
67

kernlab - An S4 Package for Kernel Methods in R

Karatzoglou, Alexandros, Smola, Alex, Hornik, Kurt, Zeileis, Achim 11 1900 (has links) (PDF)
kernlab is an extensible package for kernel-based machine learning methods in R. It takes advantage of R's new S4 object model and provides a framework for creating and using kernel-based algorithms. The package contains dot product primitives (kernels), implementations of support vector machines and the relevance vector machine, Gaussian processes, a ranking algorithm, kernel PCA, kernel CCA, and a spectral clustering algorithm. Moreover it provides a general purpose quadratic programming solver, and an incomplete Cholesky decomposition method.
68

[en] AN IMPROVED EXACT METHOD FOR THE UBQP / [pt] UM MÉTODO EXATO MELHORADO PARA O UBQP

DANIEL FLEISCHMAN 11 March 2011 (has links)
[pt] A Programação Quadrática Binária Irrestrita (UBQP) é amplamente estudada. Trata-se de uma ferramenta de modelagem poderosa, mas otimizar de um problema NP-difícil. Neste trabalho uma nova abordagem é apresentada, que pode ser usada para construir um algoritmo exato. Além disso, a ideia básica que fundamenta o trabalho pode ser usado em um espectro ainda mais amplo de problemas. O algoritmo exato derivado do novo método é altamente paralelizável, o que é uma característica desejável nos dias de hoje em que cloud computing já é uma realidade. Para instâncias razoavelmente grandes do UBQP, o novo método pode paralelizar a centenas, ou até milhares, de núcleos com facilidade, com um aumento de desempenho quase linear. / [en] Unconstrained Binary Quadratic Programming (UBQP) is widely studied. It is a powerful modeling tool and its associate problem is NP-hard. In this work a new approach is introduced, which can be used to build an exact algorithm. Also, the fundamental idea behind it can be used in an even wider family of problems. This exact algorithm derived from the new method is highly parallelizable, which is a desired feature nowadays, when the cloud computing is a reality. For reasonably large instances of UBQP, the new method can parallelize to hundreds, or even thousands, of cores easily, with a near-linear speedup.
69

Condições de otimalidade em programação multiobjetivo fracional quadrático / Multiobjective quadratic fractional programming problems

Oliveira, Washington Alves de, 1977- 18 August 2018 (has links)
Orientador: Antonio Carlos Moretti, Margarida Pinheiro Mello / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Matemática, Estatística e Computação Científica / Made available in DSpace on 2018-08-18T11:43:39Z (GMT). No. of bitstreams: 1 Oliveira_WashingtonAlvesde_D.pdf: 1534705 bytes, checksum: 351c92a12c85da49389a18880da92ee7 (MD5) Previous issue date: 2011 / Resumo: Existem na literatura diversos conceitos e definições que caracterizam e dão condições de otimalidade para as soluções de um problema de programação multiobjetivo. A mais importante é a condição necessária de primeira ordem, que generaliza a condição clássica do tipo Karush-Kuhn-Tucker em otimização não linear. Esta condição garante a existência de uma vizinhança arbitrária onde uma solução ótima está contida. No entanto, para se obter condições suficientes de otimalidade, tanto local como global, é necessário impor hipóteses adicionais sobre as funções objetivo e o conjunto de restrições, como convexidade ou as suas generalizações. Em determinados problemas tais hipóteses podem ser muito restritivas. Neste trabalho, introduzimos um conceito alternativo para identificar a vizinhança de uma solução ótima local em problemas de programação multiobjetivo. Em uma primeira etapa, usando este conceito, obtemos condições necessárias e suficientes de otimalidade para as soluções de um problema particular, onde cada função objetivo é constituída de um quociente de funções quadráticas e o conjunto de restrições é formado por desigualdades lineares. Então, mostramos como calcular o maior raio da região esférica centrada em uma solução ótima local na qual esta solução é ótima. Nesse processo, podemos concluir que esta solução também é globalmente ótima. Em uma segunda etapa, usando o gradiente e a Hessiana de cada função quadrática, caracterizamos as soluções ótimas locais. Em uma terceira etapa, obtemos condições suficientes de otimalidade global impondo algumas hipóteses adicionais, porém essas hipóteses não caracterizam nenhum tipo de convexidade generalizada sobre as funções objetivo. Finalizamos com alguns resultados de dualidade. Este problema particular, envolvendo otimização fracional, surge frequentemente em aplicações nos processos de tomada de decisão em Ciência da Gestão, por exemplo, quando se deseja otimizar razões como desempenho/custo, lucro/investimento, custo/tempo, etc. Por isso, também propomos ao longo do texto vários métodos computacionais derivados dos nossos resultados que podem ser usados na obtenção de soluções para esses tipos de aplicações / Abstract: In the literature there are several concepts and definitions that characterize and give optimality conditions for solutions of a multiobjective programming problem. The most important is the necessary first-order optimality condition that generalizes the Karush-Kuhn-Tucker conditions. This condition ensures the existence of an arbitrary neighborhood that contains an optimal solution. However, in order to obtain optimality sufficient conditions, both local and global, it is necessary to impose additional assumptions on the objective functions and on the feasible set such as convexity and its generalizations. Sometimes, in some problems, such assumptions are too restrictive. In this work, we introduce an alternative concept to identify the local optimal solution neighborhood in multiobjective programming problems. In a first step, using this concept, we obtain necessary and sufficient optimality conditions for the solutions of a particular problem, where each objective function consists of a ratio quadratic functions and the feasible set is defined by linear inequalities. Then we show how to calculate the largest radius of the spherical region centered on a local optimal solution in which the local solution is optimal. In this process we may conclude that the solution is also globally optimal. In a second step, using the gradient and the Hessian of each quadratic function, we characterize the local optimal solutions. In a third step, we obtain global optimality sufficient conditions by imposing some additional assumptions but these assumptions do not characterize any kind of generalized convexity on the objective functions. We conclude this work with some results of the duality. This particular problem, involving fractional optimization, arises frequently in the decision making of the management science applications, for example, if you want to otimize the performance/cost ratio, or profit/investment, or cost/time, etc.. Therefore, we also propose throughout the text various computational methods derived from our results. These methods can be used to obtain solutions to these types of applications / Doutorado / Matematica Aplicada / Doutor em Matemática Aplicada
70

A conic optimization approach to variants of the trust region subproblem

Yang, Boshi 01 July 2015 (has links)
The Trust Region Subproblem (TRS), which minimizes a nonconvex quadratic function over the unit ball, is an important subproblem in trust region methods for nonlinear optimization. Even though TRS is a nonconvex problem, it can be solved in polynomial time using, for example, a semidefinite programming (SDP) relaxation. Different variants of TRS have been considered from both theoretical and practical perspectives. In this thesis, we study three variants of TRS and their SDP/conic relaxations. We first study an extended trust region subproblem (eTRS) in which the trust region equals the intersection of the unit ball with M linear cuts. When m = 0, when m = 1, or when m = 2 and the linear cuts are parallel, it is known that the eTRS optimal value equals the optimal value of a particular conic relaxation, which is solvable in polynomial time. However, it is also known that, when m ≥2 and at least two of the linear cuts intersect within the ball, i.e., some feasible point of the eTRS satisfies both linear constraints at equality, then the same conic relaxation may admit a gap with eTRS. We show that the conic relaxation admits no gap for arbitrary M as long as the linear cuts are non-intersecting. We then extend our result to a more general setting. We study an eTRS in which a quadratic function is minimized over a structured nonconvex feasible region: the unit ball with M linear cuts and R hollows. In the special case when m = 0 and r = 1, it is known that the eTRS has a tight polynomial-time solvable conic relaxation. We show that a certain conic relaxation is also tight for general R and M as long as the cuts and hollows satisfy some non-intersecting assumptions that generalize the previous paragraph. Finally, intersecting the feasible region of TRS with a second ellipsoid results in the two-trust-region subproblem (TTRS). Even though TTRS can also be solved in polynomial-time, existing approaches do not provide a concise conic relaxation. We investigate the use of conic relaxation for TTRS. Starting from the basic SDP relaxation of TTRS, which admits a gap, recent research has tightened the basic relaxation using valid second-order-cone (SOC) inequalities. For the special case of TTRS in dimension n=2, we fully characterize the remaining valid inequalities, which can be viewed as strengthened versions of the SOC inequalities just mentioned. We also demonstrate that these valid inequalities can be used computationally even when n > 2 to solve TTRS instances that were previously unsolved using techniques of conic relaxation.

Page generated in 0.3659 seconds