Bruce Ronald Musicus. / Originally published as thesis (Dept. of Electrical Engineering and Computer Science, Ph.D., 1982). / Includes bibliographies. / Supported in part by the Advanced Research Projects Agency monitored by ONR under Contract N00014-81-K-0742 NR-049-506 Supported in part by the National Science Foundation under Grant ECS80-07102

Estimating the structural weight of a Hybrid Wing Body (HWB) aircraft during conceptual design has proven to be a significant challenge due to its unconventional configuration. Aircraft structural weight estimation is critical during the early phases of design because inaccurate estimations could result in costly design changes or jeopardize the mission requirements and thus degrade the concept's overall viability. The tools and methods typically employed for this task are inadequate since they are derived from historical data generated by decades of tube-and-wing style construction. In addition to the limited applicability of these empirical models, the conceptual design phase requires that any new tools and methods be flexible enough to enable design space exploration without consuming a significant amount of time and computational resources. This thesis addresses these challenges by developing a parametric and physics-based modeling and simulation (M&S) environment for the purpose of HWB structural weight estimation. The tools in the M&S environment are selected based on their ability to represent the unique HWB geometry and model the physical phenomena present in the centerbody section. The new M&S environment is used to identify key design parameters that significantly contribute to the variability of the HWB centerbody structural weight and also used to generate surrogate models. These surrogate models can augment traditional aircraft sizing routines and provide improved structural weight estimations.

This thesis is concerned with simulation output analysis. In particular, we are inter- ested in estimating the variance parameter of a steady-state output process. The estimation of the variance parameter has immediate applications in problems involving (i) the precision of the sample mean as a point estimator for the steady-state mean and #956;X, and (ii) confidence intervals for and #956;X. The thesis focuses on new variance estimators arising from Schrubens method of standardized time series (STS). The main idea behind STS is to let such series converge to Brownian bridge processes; then their properties are used to derive estimators for the variance parameter. Following an idea from Shorack and Wellner, we study different levels of folded Brownian bridges. A folded Brownian bridge is obtained from the standard Brownian bridge process by folding it down the middle and then stretching it so that it spans the interval [0,1]. We formulate the folded STS, and deduce a simplified expression for it. Similarly, we define the weighted area under the folded Brownian bridge, and we obtain its asymptotic properties and distribution. We study the square of the weighted area under the folded STS (known as the folded area estimator ) and the weighted area under the square of the folded STS (known as the folded Cram??von Mises, or CvM, estimator) as estimators of the variance parameter of a stationary time series. In order to obtain results on the bias of the estimators, we provide a complete finite-sample analysis based on the mean-square error of the given estimators. Weights yielding first-order unbiased estimators are found in the area and CvM cases. Finally, we perform Monte Carlo simulations to test the efficacy of the new estimators on a test bed of stationary stochastic processes, including the first-order moving average and autoregressive processes and the waiting time process in a single-server Markovian queuing system.

A general multidisciplinary design problem features coupling and feedback between contributing analyses. This feedback may lead to convergence issues requiring significant iteration in order to obtain a feasible design. This work casts the multidisciplinary design problem as a dynamical system in order to leverage the benefits of dynamical systems theory in a new domain. Three areas from dynamical system theory are chosen for investigation: stability analysis, optimal control, and estimation theory. Stability analysis is used to investigate the existence of a solution to the design problem and how that solution can be found. Optimal control techniques allow consideration of contributing analysis output and design variables constraints at the same level of the optimization hierarchy. Finally, estimation methods are employed to rapidly evaluate the robustness of the multidisciplinary design. These three dynamical system techniques are then combined in a methodology for the rapid robust design of linear multidisciplinary systems. While inherently linear, the developed robust design methodology is shown to be extensible to nonlinear systems. The applicability and performance of the developed technique is demonstrated through linear and nonlinear test problems including the design of a hypersonic aerodynamic surface for a system in which an increase in range or improvement in landed accuracy is sought. In addition, it is shown that the developed robust design methodology scales well compared to other methods.

A meta principal desse trabalho foi comparar métodos de estimação para coeficientes de herdabilidade para os modelos inteiramente ao acaso e em blocos casualizados. Para os dois casos foram utilizadas as definições de coeficiente de herdabilidade (h2) no sentido restrito, dadas respectivamente, por h2=4 σ2t/(σ2+σ2t) e h2=4 σ2t/(σ2+σ2t+σ2b). . Portanto, é preciso estimar os componentes de variância relativos ao erro experimental (σ2) e ao efeito de tratamentos (σ2t) quando se deseja estimar h2 para o modelo inteiramente ao acaso. Para o modelo para blocos casualizados, além de estimar os últimos dois componentes, é necessário estimar o componente de variância relativo ao efeito de blocos (σ2b). Para atingir a meta estabelecida, partiu-se de um conjunto de dados cujo coeficiente de herdabilidade é conhecido, o que foi feito através da simulação de dados. Foram comparados dois métodos de estimação, o método da análise da variância e método da máxima verossimilhança. Foram feitas 80 simulações, 40 para cada ensaio. Para os dois modelos, as 40 simulações foram divididas em 4 casos contendo 10 simulações. Cada caso considerou um valor distinto para h2, esses foram: h2=0,10; 0,20; 0,30 e 0,40; para cada um desses casos foram fixados 10 valores distintos para o σ2, a saber: σ2=10; 20; 30; 40; 50; 60; 70; 80; 90; 100. Os valores relativos ao σ2 foram encontrados através da equação dada para os coeficientes de herdabilidade, sendo que, para o modelo em blocos casualizados, foi fixado σ2b=20 para todas os 40 casos. Após realizadas as 80 simulações, cada uma obtendo 1000 conjunto de dados, e por conseqüência 1000 estimativas para cada componente de variância e coeficiente de herdabilidade relativos a cada um dos casos, foram obtidas estatísticas descritivas e histogramas de cada conjunto de 1000 estimativas. A comparação dos métodos foi feita através da comparação dessas estatísticas descritivas e histogramas, tendo como referência os valores dos parâmetros utilizados nas simulações. Para ambos os modelos observou-se que os dois métodos se aproximam quanto a estimação de σ2. Para o delineamento inteiramente casualizado, o método da máxima verossimilhança forneceu estimativas que, em média, subestimaram os valores de σ2t, e por conseqüência, tendem a superestimar o h2, o que não acontece para o método da análise da variância. Para o modelo em blocos casualizados, ambos os métodos se assemelham, também, quanto à estimação de σ2t, porém o método da máxima verossimilhança fornece estimativas que tendem a subestimar o σ2b, e e por conseqüência, tendem a superestimar o h2, o que não acontece para o método da análise da variância. Logo, o método da análise da variância se mostrou mais confiável quando se objetiva estimar componentes de variância e coeficientes de herdabilidade para ambos os modelos considerados. / The main aim of this work was to compare methods of estimation of heritability for the 1- way classification and the 2-way crossed classification without interaction. For both cases the definition of heritability (h2) in the narrow sense was used, given respectively, by h2=4σ2t/(σ2+σ2t) e h2=4σ2t/(σ2+σ2t+σ2b). Therefore, there is a need to estimate the components of variance related to the residual (σ2) and the effect of treatments (σ2t) in order to estimate (h2) for the 1-way classification. For the 2-way classification without interaction, there is a need to estimate the component of variance related to the effect of blocks (σ2b) as well as the other two components. To achieve the established aim, a data set with known heritability was used, produced by simulation. Two methods of estimation were compared: the analysis of variance method and the maximum likelihood method. 80 simulations were made, 40 for each classification. For both models, the 40 simulations were divided into 4 different groups containing 10 simulations. Each group considered a different value for h2 (h2=0,10; 0,20; 0,30 e 0,40) and for each one of those cases there were 10 different values fixed for) σ2 (σ2=10; 20; 30; 40; 50; 60; 70; 80; 90; 100). The values for σ2t were found using the equations for the heritability, and for the 2-way crossed classification without interaction, σ2b=20 for all the 40 cases. After the 80 simulations were done, each one obtaining 1000 data sets, and therefore 1000 estimates of each component of variance and the heritability, descriptive statistics and histograms were obtained for each set of 1000 estimates. The comparison of the methods was made based on the descriptive statistics and histograms, using as references the values of the parameters used in the simulations. For both models, the estimates of σ2 were close to the true values. For the 1-way classification, the maximum likelihood method gave estimates that, on average, underestimated the values of σ2t, and therefore the values of h2. This did not happen with the analysis of variance method. For the 2-way crossed classification without interaction, both methods gave similar estimates of σ2t, although the maximum likelihood method gave estimates that tended to underestimate σ2b and therefore to overestimate h2. This did not happen with the analysis of variance method. Hence, the analysis of variance method proved to be more accurate for the estimation of variance components and heritability for both classifications considered in this work.

Orientador: Madson Cortes de Almeida / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-23T14:17:23Z (GMT). No. of bitstreams: 1 Schmidt_Fabiano_M.pdf: 1011600 bytes, checksum: d02fbe7d42a25cfc2745ce47bff3b467 (MD5) Previous issue date: 2013 / Resumo: A estimação de estado em sistemas de energia elétrica pode ser descrita como um conjunto de funções que visam fornecer o modelo em tempo real do sistema elétrico. Fazem parte do modelo, a topologia da rede, seus parâmetros elétricos e as variáveis de estado, frequentemente definidas como os fasores de tensões nas barras do sistema. Na obtenção das variáveis de estado, embora os sistemas de medição possam ser projetados para que a rede seja sempre observável, eventualmente, falhas de comunicação, mudanças topológicas ou falhas em medidores podem tornar a rede temporariamente não observável. Nessas situações, as equações normais do modelo de estimação por mínimos quadrados tornam-se mal postas inviabilizando a estimação do estado. Convencionalmente, esse problema é contornado através da restauração da observabilidade, onde pseudomedidas são adicionadas à rede tornando-a novamente observável. Esta dissertação trata da estimação de estado regularizada, que é um método alternativo para lidar com os problemas associados a não observabilidade temporária da rede. Com o estimador regularizado é possível evitar a etapa de restauração da observabilidade, simplificando o processo de estimação de estado. O esquema de regularização adotado é baseado na regularização de Tikhonov, que consiste na introdução de um conjunto de informações a priori que tornam o problema sempre factível. Esse conjunto de informações a priori é obtido da melhor estimativa conhecida para o estado da rede ou a partir de previsões baseadas no conhecimento do comportamento histórico da rede. Ao regularizar o problema, é desejável que o conjunto de informações introduzidas não deteriore a qualidade do estado estimado. Para tanto, deve haver um compromisso entre a qualidade das informações a priori e a suas respectivas ponderações. O uso de ponderações inadequadas pode deteriorar o estado estimado ou piorar o condicionamento numérico das matrizes envolvidas, dificultando a convergência do estimador regularizado. Tais problemas são estudados nesta dissertação. O método de regularização foi desenvolvido nas versões completa e desacoplada que foram combinadas em dois algoritmos visando melhorar as características de convergência do estimador regularizado. As versões desenvolvidas foram testadas com as redes de 14 e 118 barras do Institute of Electrical and Electronics Engineers (IEEE) / Abstract: Power system state estimation can be described as a set of functions aimed to provide a real time power system electrical model. This model is composed of the network topology, its electrical parameters and a set of state variables, often defined as the bus voltage phasors. In obtaining of the state variables, although the measurement systems may be designed to ensure that the grid is always observable, eventually, communication failures, topological changes or meter failures can make the grid temporarily unobservable. In these situations the normal equations of the least squares estimation model becomes ill posed, invalidating the state estimation. Conventionally this problem is solved by the observability restoration, where pseudomeasurements are added to the network making it observable again. This thesis deals with the regularized state estimation, that is, an alternative method to deal with the problems associated to temporary network unobservability. With the regularized estimator is possible to avoid the observability restoration stage, simplifying the state estimation procedure. The regularization scheme used is based on the Tikhonov regularization, that consists in the introduction of an a priori information set that results in a problem always feasible. This set of a priori information is obtained from the best estimative for the state variables or by a forecast based on historical patterns of the grid. During regularization of the problem it is desirable that the information set does not deteriorate the quality of the state estimated. Therefore, must be a compromise between the quality of the a priori information and its weighing. The use of inadequate weighting can deteriorate the state estimated or worsen the numerical condition of the involved matrices, disturbing or even preventing the regularized estimator convergence. These problems are studied in this thesis. The regularization method was developed for both the complete and the decoupled variants. These variants are combined in two algorithms that aim to improve the convergence characteristics of the regularized state estimator. The developed variants are tested with the IEEEs 14 and 118 networks / Mestrado / Energia Eletrica / Mestre em Engenharia Elétrica

Orientador: Amauri Lopes / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-19T00:45:48Z (GMT). No. of bitstreams: 1 Krummenauer_Rafael_D.pdf: 3012176 bytes, checksum: 0650db76fe599c68547540bc1a973c9b (MD5) Previous issue date: 2011 / Resumo: Nesta tese tratamos do problema de estimação de direção de chegada de sinais digitalmente modulados, especificamente encontrado no enlace reverso de sistemas de comunicação semfiomultiusuário em que a estação rádio-base é composta por um arranjo de antenas, posicionadas de acordo com alguma geometria pré-definida. Os objetivos principais da tese são: esclarecer detalhes normalmente não tratados na descrição dos modelos de dados considerados na literatura existente, em especial, a respeito das condições de enlace e do sistema receptor necessárias para suportar a validade dos modelos empregados; agrupar alguns dos métodos de estimação existentes mais relevantes para o contexto tratado e avaliar suas vantagens e desvantagens; e introduzir o método de estimação desenvolvido nas pesquisas do período de doutorado, o qual representa a contribuição maior da tese em termos de metodologia. A solução proposta é fundamentada no critério da máxima verossimilhança e as análises de desempenho feitas através de simulações numéricas mostram que o estimador fornece resultados com qualidade próxima ao limite definido pelo limitante de Cramér-Rao. Algumas características importantes do método são: i) capacidade de operar com qualquer número de fontes se o número de antenas for maior ou igual a dois e o número de snapshots for suficientemente maior que a cardinalidade do espaço de sinal; ii) a qualidade da estimação não é afetada pela separação angular entre as fontes; e iii) possibilidade de operar com arranjos descalibrados / Abstract: This thesis deals with the problem of estimating the direction of arrival of digitally modulated signals, specifically found in the uplink of multiuser wireless communication systems where the base station uses an antenna array. The main objectives of the thesis are: to clarify some details normally not treated in the description of the data models assumed in the existing literature, specially, on link and receiver conditions that are necessary to validate the employed models; to present some of the most relevant estimation methods to the current context and to evaluate their advantages and disadvantages; and to introduce the estimation method developed in the researches during the doctoral period, which represents the greater contribution of the thesis in terms of metodology. The proposed solution is fundamented in the maximum likelihood criterion and the performance analysis done through numerical simulations show that the estimator yields results with a quality close to that defined by the Cramér-Rao bound. Some important characteristics of the method are: i) capacity to operate with any number of sources if the number of antennas is greater than or equal to two and the number of snapshots is sufficiently greater than the cardinality of the signal space; ii) the estimation quality does not depend on the angle separation between sources; and iii) possibility to operate with uncalibrated arrays / Doutorado / Telecomunicações e Telemática / Doutor em Engenharia Elétrica

Nous étudions dans cette thèse l'influence d'un système d'authentification utilisant des codes graphiques 2D modifiés lors de l'impression par un procédé physique non-clônable. Un tel procédé part du principe qu'à très haute résolution le système d'impression acquisition peut être modélisé comme un processus stochastique, de part le caractère aléatoire de la disposition des fibres de papiers, de mélange des particules d'encre, de l'adressabilité de l'imprimante ou encore du bruit d'acquisition. Nous considérons un scénario où l'adversaire pourra estimer le code original et essaiera de le reproduire en utilisant son propre système d'impression. La première solution que nous proposons pour arriver à l'authentification est d'utiliser un test d'hypothèse à partir des modèles à priori connus et sans mémoire des canaux d'impression-acquisition de l'imprimeur légitime et du contrefacteur. Dans ce contexte nous proposons une approximation fiable des probabilités d'erreur via l'utilisation de bornes exponentiels et du principe des grandes déviations. Dans un second temps, nous analysons un scénario plus réaliste qui prends en compte une estimation a priori du canal du contrefacteur et nous mesurons l'impact de cette étape sur les performances du système d'authentification. Nous montrons qu'il est possible de calculer la distribution des probabilité de non-détection et d'en extraire par exemple ses performances moyennes. La dernière partie de cette thèse propose d'optimiser, au travers d'un jeu minimax, le canal de l'imprimeur. / We study in this thesis the impact of an authentication system based on 2D graphical codes that are corrupted by a physically unclonable noise such as the one emitted by a printing process. The core of such a system is that a printing process at very high resolution can be seen as a stochastic process and hence produces noise, this is due to the nature of different elements such as the randomness of paper fibers, the physical properties of the ink drop, the dot addressability of the printer, etc. We consider a scenario where the opponent may estimate the original graphical code and tries to reproduce the forged one using his printing process in order to fool the receiver. Our first solution to perform authentication is to use hypothesis testing on the observed memoryless sequences of a printed graphical code considering the assumption that we are able to perfectly model the printing process. The proposed approach arises from error exponent using exponential bounds as a direct application of the large deviation principle. Moreover, when looking for a more practical scenario, we take into account the estimation of the printing process used to generate the graphical code of the opponent, and we see how it impacts the performance of the authentication system. We show that it is both possible to compute the distribution of the probability of non-detection and to compute the average performance of the authentication system when the opponent channel has to be estimated. The last part of this thesis addresses the optimization problem of the printing channel.

M.Sc. (Mathematical Statistics) / Comparison of two distributions via use of the quantile comparison function is carried out specifically from possibly censored data. A semi-parametric method which assumes linearity of the quantile comparison function is examined thoroughly for non-censored data and then extended to incorporate censored data. A fully nonparametric method to construct confidence bands for the quantile comparison function is set out. The performance of all methods examined is tested using Monte Carlo Simulation.

M.Sc. (Mathematical Statistics) / Stochastic Differential Equations (SDE’s) are commonly found in most of the modern finance used today. In this dissertation we use SDE’s to model a random phenomenon known as the short-term interest rate where the explanatory power of a particular short-term interest rate model is largely dependent on the description of the SDE to the real data. The challenge we face is that in most cases the transition density functions of these models are unknown and therefore, we need to find reliable and accurate alternative estimation techniques. In this dissertation, we discuss estimating techniques for discretely sampled continuous diffusion processes that do not require the true transition density function to be known. Moreover, the reader is introduced to the following techniques: (i) continuous time maximum likelihood estimation; (ii) discrete time maximum likelihood estimation; and (iii) estimating functions. We show through a Monte Carlo simulation study that the parameter estimates obtained from these techniques provide a good approximation to the estimates obtained from the true transition density. We also show that the bias in the mean reversion parameter can be reduced by implementing the jackknife bias reduction technique. Furthermore, the data analysis carried out on South-African interest rate data indicate strongly that single factor models do not explain the variability in the short-term interest rate. This may indicate the possibility of distinct jumps in the South-African interest rate market. Therefore, we leave the reader with the notion of incorporating jumps into a SDE framework.