No description available.

This thesis presents the development and implementation of a generalized optimization framework for use in sheet-stamping process simulation by finite element analysis. The generic framework consists of three main elements: a process simulation program, an optimization code, and a response filtering program. These elements can be filled by any combination of applicable software packages. Example sheet-stamping process simulations are presented to demonstrate the usage of the framework in various forming scenarios. Each of the example simulations is presented with a sensitivity analysis. These examples include analysis of a 2-dimensional single-stage forming, a 2-dimensional multi-stage forming, and two different 3-dimensional single-stage forming processes. A forming limit diagram is used to define failure in the 3-dimensional process simulations. Optimization results are presented using damage minimization, thinning minimization, and springback minimization with aluminum alloy 6061-T6 blanks.

Evaluating classifiers in controlled settings is essential for empirical applications, as extensive knowledge on model-behaviour is needed for accurate predictions. This thesis investigates robustness against non-normality of two prominent classifiers, LDA and QDA. Through simulation, errors in leave-one-out cross-validation are compared for data generated by different multivariate distributions, also controlling for covariance structures, class separation and sample sizes. Unexpectedly, the classifiers perform better on data generated by heavy-tailed symmetrical distributions than by the normal distribution. Possible explanations are proposed, but the cause remains unknown. There is need for further studies, investigating more settings as well as mathematical properties to verify and understand these results.

No description available.

[pt] Nesta tese, é apresentado e analisado em termos da propriedade de convergência global um novo algoritmo para problemas de otimização quadrática sujeitos ou a restrições de desigualdades matriciais bilineares (BMIs) ou a restrições de desigualdades matriciais lineares (LMIs) dependentes da freqüência, estes problemas são muito relevantes para a teoria de controle porque uma ampla classe de controladores (por exemplo, controladores H2/Hinf de ordem fixa, síntese de controladores descentralizados, análise do desempenho robusto H2 ou Hinf, etc.) pode ser computada a partir da solução de problemas de otimização desta natureza. Infelizmente, estes problemas são reconhecidamente de difícil solução, pois envolvem, entre outras coisas, não convexidade (restrição BMI), não diferenciabilidade da restrição, etc. em função dessa complexibilidade, algumas alternativas para a obtenção de soluções aproximadas têm sido adotada na literatura especializada recente (Safonov, 1994; Paganini, 1996). O algoritmo proposto neste trabalho de tese é uma alternativa para as abordagens atuais com vantagens no sentido de permitir a obtenção de melhores aproximações assim como a possibilidade de explorar a estrutura particular de cada problema de interesse e, com isso, viabilizar do ponto de vista computacional o projeto de controladores envolvendo plantas de ordem mais elevada. Este algoritmo pode ser visto como a generalização de um algoritmo anterior com boas propriedades proposto por Corrêa & Sales (1998) para problemas quadráticos sujeitos a restrições envolvendo LMIs canônicas. De forma bastante genética, a solução do problema original (um problema envolvendo um número infinito de restrições é substituído por uma única) onde, em cada passo do algoritmo, a restrição é interativamente modificada. Demonstrar-se-á que para problemas quadráticos envolvendo restrições BMIs (problemas não convexos) a seqüência de soluções geradas pelo algoritmo convergirá para a solução ótima global do problema original. Por outro lado, no caso dos problemas quadráticos envolvendo restrições LMIs dependentes da freqüência, a seqüência gerada de custos auxiliares é monótona crescente e, adicionalmente, se a seqüência de matrizes de ponderação for limitada superiormente (uma condição suficiente), demonstrar-se-á que a seqüência de soluções geradas pelo algoritmo convergirá para a solução ótima global do problema original. Finalmente, são apresentadas algumas aplicações a problemas de controle acompanhadas de alguns exemplos numéricos ilustrativos. / [en] In this thesis, it is proposed and analysed in terms of the global-convergence property a new algorithm for solving quadratic optimisation problems under either a BMI (bilinear matrix inequality) or a frequency-dependent LMI (linear matrix inequality) constraints. These problems are of special interest in the control literature a some very important control problems such as the H2/H(infinite) fixed-order controller, multiobjectives, H2 and H (infinite) robust performance analysis among others problems can be posed as problems of this kind for which does not still exist yet a reliable global convergent algorithm. Nowadays, approximate solutions to those problems are based upon grid and interpolation techniques as suggested by Paganini (1996) in the case of frequency- wise LMI constraints or branch and bound algorithms or branch and bound algorithms mainly and alternating LMIs as far as BMIs constraints are involved (Safonov, 1994). All of those approaches suffer, of course, from obvious numerical difficulties. In fact, those approaches were introduced as preliminary attempts in solving the problems just mentioned. The algorithm to presented here, which can be seen as a generalisation of an earlier algorithm proposed by Corrêa e Sales (1998) for solving standard feasibility LMIs problems, is a step forward in an attempt of handling difficulties not faced properly by those methodologies. In a broaden sense, the proposed algorithm solves the original problem (a problem subject to an infinite number of constraints is replaced by a single one properly chosen. It is worth noting that this basic idea was introduced by Lawson (1961) in a rather different context, namely, the problem of computing Tchebycheff approximations by means of sequences of weighted quadratic problems. It is pointed out here that in the case of quadratic problems under a BMI constraint (a nonconvex problem); it is proved that the sequence of auxiliary solutions generated by the algorithm converges to the global optimal solution of the original one. On the other hand, as for quadratic problems under a frequency-dependent LMI constraint (an infinite-dimensional problem) it is proved that the auxiliary cost-sequence values increases asymptotically and, If the weight updating sequence is bounded from above (a sufficient condition), the sequence of auxiliary solutions will converge to the optimal solution of the original problem as well. Finally, some applications to control problem are presented accompanied by some numerical examples.

The aim of this thesis is to optimize hydro power plants with data generated from observations and field tests at the plants. The output is optimal production tables and curves in order to operate and plan hydro power plants in an optimized way concerning power output, efficiency and distribution of water. The thesis is performed in collaboration with Vattenfall AB, which currently use an internal optimization program called SEVAP. Two alternative methods have been selected, employed and compared with the current optimization program, these are Interior-Point Method and Sequential Quadratic Programming. Three start-point strategies are created to increase the probability of finding a global optima. A heuristic rule is used for selection of strategy in order to prevent rapid changes in load distribution for small variations in dispatched water. The optimization is performed at three plants in Sweden with different size and setup. The results of this evaluation showed marginally better results for the employed methods in comparison to the currently used optimization. Further, the developed program is more flexible and compatible to integrate with future digitalization projects. / Syftet med detta examensarbete är att optimera vattenkraftverk med data som genererats från indextester vid kraftverken. Resultatet är optimala produktionstabeller och kurvor för drift och planering av vattenkraftverk. Dessa är baserade på att optimalt fördela vattnet mellan aggregaten för att maximera uteffekt och verkningsgrad. Detta arbete har utförts i samarbete med Vattenfall AB, som för närvarande använder ett internt optimeringsprogram som heter SEVAP. Två optimeringsmetoder har valts, implementerats och jämförts med det nuvarande optimeringsprogrammet. Dessa metoder är inrepunktsmetoden (IPM) och sekventiell kvadratiskt programmering (SQP). Tre startpunktsstrategier har används för att öka sannolikheten att hitta ett globalt optima. För att förhindra hastiga förändringar i lastfördelning för små variationer av avsänt vatten har en heuristisk regel används. Optimeringen har utförts på tre stationer med olika uppsättning och storlek. Resultatet av detta examensarbete visar marginellt bättre resultat för de använda metoderna i jämförelse med den nuvarande optimeringen. Det utvecklade programmet är flexibelt och kompatibelt att integrera med framtida digitaliseringsprojekt.

Discrete models are used in describing various microscopic phenomena in many branches of science, ranging from biology through chemistry to physics. Arrays of evanescently coupled, equally spaced, identical waveguides are prime examples of optical structures in which discrete dynamics can be easily observed and investigated. As a result of discretization, these structures exhibit unique diffraction properties with no analogy in continuous systems. Recently nonlinear discrete optics has attracted a growing interest, triggered by the observation of discrete solitons in AlGaAs waveguide arrays reported by Eisenberg et al. in 1998. So far, the following experiments involved systems with third order nonlinearities. In this work, an experimental investigation of discrete nonlinear wave propagation in a second order nonlinear medium is presented. This system deserves particular attention because the nonlinear process involves two or three components at different frequencies mutually locked by a quadratic nonlinearity, and new degrees of freedom enter the dynamical process. In the first part of dissertation, observation of the discrete Talbot effect is reported. In contrast to continuous systems, where Talbot self-imaging effect occurs irrespective of the pattern period, in discrete configurations this process is only possible for a specific set of periodicities. The major part of the dissertation is devoted to the investigation of soliton formation in lithium niobate waveguide arrays with a tunable cascaded quadratic nonlinearity. Soliton species with different topology (unstaggered – all channels in-phase, and staggered – neighboring channels with a pi relative phase difference) are identified in the same array. The stability of the discrete solitons and plane waves (modulational instability) are experimentally investigated. In the last part of the dissertation, a phase-insensitive, ultrafast, all-optical spatial switching and frequency conversion device based on quadratic waveguide array is demonstrated. Spatial routing and wavelength conversion of milliwatt signals is achieved without pulse distortions.

The mathematical operation of correlation is a very simple concept, yet has a very rich history of application in a variety of engineering fields. It is essentially nothing but a technique to measure if and to what degree two signals match each other. Since this is a very basic and universal task in a wide variety of fields such as signal processing, communications, computer vision etc., it has been an important tool. The field of pattern recognition often deals with the task of analyzing signals or useful information from signals and classifying them into classes. Very often, these classes are predetermined, and examples (templates) are available for comparison. This task naturally lends itself to the application of correlation as a tool to accomplish this goal. Thus the field of Correlation Pattern Recognition has developed over the past few decades as an important area of research. From the signal processing point of view, correlation is nothing but a filtering operation. Thus there has been a great deal of work in using concepts from filter theory to develop Correlation Filters for pattern recognition. While considerable work has been to done to develop linear correlation filters over the years, especially in the field of Automatic Target Recognition, a lot of attention has recently been paid to the development of Quadratic Correlation Filters (QCF). QCFs offer the advantages of linear filters while optimizing a bank of these simultaneously to offer much improved performance. This dissertation develops efficient QCFs that offer significant savings in storage requirements and computational complexity over existing designs. Firstly, an adaptive algorithm is presented that is able to modify the QCF coefficients as new data is observed. Secondly, a transform domain implementation of the QCF is presented that has the benefits of lower computational complexity and computational requirements while retaining excellent recognition accuracy. Finally, a two dimensional QCF is presented that holds the potential to further save on storage and computations. The techniques are developed based on the recently proposed Rayleigh Quotient Quadratic Correlation Filter (RQQCF) and simulation results are provided on synthetic and real datasets.

In this thesis, analysis and comparison of different attitude control systems of a satelliteusing different reaction wheel configurations were investigated. Three different reactionwheel configurations (e.g. tetrahedron configuration, pyramid configuration, standardorthogonal 3-wheel configuration) and three control algorithms (Linear Quadratic Regulator,Sliding Mode, Integrator Backstepping) were analyzed and compared in terms of settlingtimes, power consumptions and actuator failure robustness. / <p>Validerat; 20121205 (global_studentproject_submitter)</p>

This thesis treats aspects of two fundamental problems in applied financial mathematics: calibration of a given stochastic process to observed marketprices on financial instruments (which is the topic of the first paper) and strategies for hedging options in financial markets that are possibly incomplete (which is the topic of the second paper). Calibration in finance means choosing the parameters in a stochastic process so as to make the prices on financial instruments generated by the process replicate observed market prices. We deal with the so called local volatility model which is one of the most widely used models in option pricing across all asset classes. The calibration of a local volatility surface to option marketprices is an ill-posed inverse problem as a result of the relatively small number of observable market prices and the unsmooth nature of these prices in strike and maturity. We adopt the practice advanced by some authors to formulate this inverse problem as a least squares optimization under the constraint that option prices follow Dupire’s partial differential equation. We develop two algorithms for performing the optimization: one based on techniques from optimal control theory and another in which a numerical quasi-Newton algorithmis directly applied to the objective function. Regularization of the problem enters easily in both problem formulations. The methods are tested on three months of daily option market quotes on two major equity indices.The resulting local volatility surfaces from both methods yield excellent replications of the observed market prices. Hedging is the practice of offsetting the risk in a financial instrument by taking positions in one or several other tradable assets. Quadratic hedging is a well developed theory for hedging contingent claims in incomplete markets by minimizing the replication error in a suitable L2-norm. This theory, though, is not widely used among market practitioners and relatively few scientific papers evaluate how well quadratic hedging works on real marketdata. We construct a framework for comparing hedging strategies, and use it to empirically test the performance of quadratic hedging of European call options on the Euro Stoxx 50 index modeled with an affine stochastic volatility model with and without jumps. As comparison, we use hedging in the standard Black-Scholes model. We show that quadratic hedging strategies significantly outperform hedging in the Black-Scholes model for out of the money options and options near the money of short maturity when only spot is used in the hedge. When in addition another option is used for hedging, quadratic hedging outperforms Black-Scholes hedging also for medium dated options near the money. / Den här avhandlingen behandlar aspekter av två fundamentala problem i tillämpad finansiell matematik: kalibrering av en given stokastisk process till observerade marknadspriser på finansiella instrument (vilket är ämnet för den första artikeln) och strategier för hedging av optioner i finansiella marknader som är inkompletta (vilket är ämnet för den andra artikeln). Kalibrering i finans innebär att välja parametrarna i en stokastisk process så att de priser på finansiella instrument som processen genererar replikerar observerade marknadspriser. Vi behandlar den så kallade lokala volatilitets modellen som är en av de mest utbrett använda modellerna inom options prissättning för alla tillgångsklasser. Kalibrering av en lokal volatilitetsyta till marknadspriser på optioner är ett illa ställt inverst problem som en följd av att antalet observerbara marknadspriser är relativt litet och att priserna inte är släta i lösenpris och löptid. Liksom i vissa tidigare publikationer formulerar vi detta inversa problem som en minsta kvadratoptimering under bivillkoret att optionspriser följer Dupires partiella differentialekvation. Vi utvecklar två algoritmer för att utföra optimeringen: en baserad på tekniker från optimal kontrollteori och en annan där en numerisk kvasi-Newton metod direkt appliceras på målfunktionen. Regularisering av problemet kan enkelt införlivas i båda problemformuleringarna. Metoderna testas på tre månaders data med marknadspriser på optioner på två stora aktieindex. De resulterade lokala volatilitetsytorna från båda metoderna ger priser som överensstämmer mycket väl med observerade marknadspriser. Hedging inom finans innebär att uppväga risken i ett finansiellt instrument genom att ta positioner i en eller flera andra handlade tillgångar. Kvadratisk hedging är en väl utvecklad teori för hedging av betingade kontrakt i inkompletta marknader genom att minimera replikeringsfelet i en passande L2-norm. Denna teori används emellertid inte i någon högre utsträckning av marknadsaktörer och relativt få vetenskapliga artiklar utvärderar hur väl kvadratisk hedging fungerar på verklig marknadsdata. Vi utvecklar ett ramverk för att jämföra hedgingstrategier och använder det för att empiriskt pröva hur väl kvadratisk hedging fungerar för europeiska köpoptioner på aktieindexet Euro Stoxx 50 när det modelleras med en affin stokastisk volatilitetsmodell med och utan hopp. Som jämförelse använder vi hedging i Black-Scholes modell.Vi visar att kvadratiska hedgingstrategier är signifikant bättre än hedging i Black-Scholes modell för optioner utanför pengarna och optioner nära pengarna med kort löptid när endast spot används i hedgen. När en annan option används i hedgen utöver spot är kvadratiska hedgingstrategier bättre än hedging i Black-Scholes modell även för optioner nära pengarna medmedellång löptid. / <p>QC 20141121</p>