O estudo do comportamento reológico de fluidos não Newtonianos tem grande importância em diversas áreas da engenharia. O aumento na demanda destes fluidos por exemplo, no uso doméstico, pessoal e em processos químicos acarreta em dificuldades que vão desde o processo de sua mistura ao seu manuseio. Dentre os fluidos não Newtonianos estão os viscoelásticos, os quais exibem deformação aparente quando os níveis de tensões são inferiores ao limite de escoamento do material e, dentro desta classe, alguns ainda podem apresentam comportamento elástico quando submetidos à baixa taxa de cisalhamento. Juntamente com os efeitos elasto-viscoplásticos, os materiais podem apresentar comportamento tixotrópico, onde devido as tensões sua reestruturação não é instantânea. Na presente Tese, fez-se um estudo numérico a fim de analisar o problema especifico de escoamentos de fluidos elasto-viscoplásticos com comportamento tixotrópico em uma expansão planar abrupta na razão de aspecto 4 : 1. O modelo mecânico aplicado consiste de uma equação viscoelástica para o campo de tensões, uma evolutiva para o parâmetro de estrutura do material, bem como as equações de conservação de massa e momento. O modelo mecânico aplicado mostrou-se capaz de prever o comportamento tixotrópico. A aproximação numérica do modelo aplicado foi feita através do método estabilizado de elementos finitos, especificamente o método Galerkin Mínimos-Quadrados (GLS), o qual foi implementado no código de elementos finitos para fluidos não Newtonianos em desenvolvimento no Laboratório de Mecânica dos Fluidos Aplicada e Computacional (LAMAC) da Universidade Federal do Rio Grande do Sul. Os fenômenos reológicos presentes no problema foram analisados a partir da influência da cinemática do escoamento, da elasticidade e da tixotropia, no nível de estruturação do material, na posição e tamanho das regiões não escoadas e na deformação elástica do material. Os resultados mostraram-se satisfatórios, pois condizem com os apresentados na literatura, apontando a boa predição do modelo mecânico aplicado bem como a robustez de sua implementação computacional. / The rheological behavior of non-Newtonian uids study is of great then importance in many areas of engineering. The increase in demand of these uids - for example, domestic use, personal and processes chemical - causes di culties ranging from the process of mixing it to handling. Among the non-Newtonian uids are viscoelastic, which exhibit apparent deformation when stress levels are lower than the yield limit of the material and, within this class, some still have elastic behavior when subjected to low shear rates. Together with the elastic-viscoplastic e ects, materials may exhibit thixotropic behavior, ie, due the restructuring strain is not instantaneous. In this thesis was made a numerical study to simulate the speci c problem of ow of elastic-viscoplastic uids with thixotropic behavior in abrupt planar expansion { common geometry in industrial systems associated with elastic-viscoplastic uids, whose ratio the aspect is 4:1. The mechanical behavior of most of these structured materials, are highly non-Newtonian, with this, there is need to make them more pseudoplastic, causing undesirable behaviors such as thixotropic which is a phenomenon somewhat characterized and modeled in the literature. The mechanical model applied is able to predict thixotropic behavior and is composed of a viscoelastic equation for the stress eld and an evolutionary to the material structure parameter in addition to the mass and momentum conservation equations. This mechanical model is approximated by a stabilized nite element model, called the Galerkin method of least squares (GLS). In order to study the rheological phenomena present in the problem is analyzed the in uence of the ow kinematics, elasticity and thixotropy in the level of structure of the material, in position and size of unyielded regions and the elastic deformation of the material. The results were satisfactory, since the the study of intensity the U the results agree with those reported in the literature pointing to good prediction of the mechanical model applied well as the hardiness of their computational implementation.The results of the elasticity showed quite spectacular behavior of unyielded regions of the material since for high values of relaxation time, the unyielded region of channel larger has the form of " ngers"because of the high exibility of the material along the line of symmetry of the channel. The results of the analysis of the elastic deformation show, that model correctly dosing the elasticity in the unyielded regions. The thixotropic e ects reported a slower structuring of the material with increasing characteristic time, in response to strain change caused by the expansion, results indicate that for high values of relaxation time, higher the distance between the unyielded regions of smaller channel and larger.

Made available in DSpace on 2014-10-09T12:48:58Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:05:51Z (GMT). No. of bitstreams: 1 09624.pdf: 7500503 bytes, checksum: 0fbfc877328a1063a37709866b6cefdc (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

Orientador: Ronei Jesus Poppi / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Quimica / Made available in DSpace on 2018-08-08T04:12:09Z (GMT). No. of bitstreams: 1 Pataca_LuizCarlosMoutinho_D.pdf: 3563791 bytes, checksum: f2877b8f6fa77bd24a2a89eed93b78c4 (MD5) Previous issue date: 2006 / Doutorado / Quimica Analitica / Doutor em Ciências

Orientador: Gilmar Barreto / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-22T21:44:00Z (GMT). No. of bitstreams: 1 TrespalaciosPerez_ManuelGuillermo_M.pdf: 4849717 bytes, checksum: 481dc5ddc485327961948561d605b96c (MD5) Previous issue date: 2012 / Resumo: Esta tese apresenta uma contribuição computacional para o método de mínimos quadrados recursivo (MQR) e a sua aplicação às múltiplas séries temporais. O objetivo principal é obter estimativas com menor erro médio final, em comparação com as estimativas resultantes da aplicação do método MQR usual. Os resultados das análises foram obtidos aplicando diferentes modificações e ajustes no modelo de dados de entrada e saída. Estes ajustes procuram ter a quantidade suficiente de dados nas séries temporais de entrada que forneçam a melhor relação com a série temporal de saída. O método desenvolvido usa inicialmente, e de forma independente, o método MQR em duas matrizes geradas pelo modelo de dados de entrada e saída, obtendo assim dois estimadores do mesmo modelo; a primeira matriz é composta pela totalidade dos dados do modelo final e a segunda matriz é gerada por um filtro que utiliza a primeira matriz e a nova observação, filtro especificado no presente trabalho. Os dois estimadores calculados para cada nova observação são comparados aplicando-lhes critérios estabelecidos neste trabalho, gerando assim um terceiro estimador que _e o melhor estimador do modelo de dados de entrada e saída. O método proposto é chamado de método de mínimos quadrados recursivo 3 (MQR3). Finalmente o método MQR3 _e testado com um modelamento de dados de múltiplas séries temporais e seus resultados comparados com os resultados do método MQR. As estatísticas encontradas para diferentes dados de validação mostram estimativas com menor erro médio final para o método MQR3 / Abstract: This thesis presents a computational contribution to the recursive least squares method (RLS) and the application to multiple time series. The main objective is to obtain estimates with lower mean final error, when compared to the estimates resulting from the application of the normal MQR method. The analysis results were obtained by applying different modifications and adjustments to the input and output data model. These adjustments seek to have a sufficient amount of data in the incoming time series to provide the highest ratio with the output time series. The method initially uses, independently, the RLS method in two arrays generated by the input and output data model, thus obtaining two estimators of the same model, the first matrix is composed of all of the data for the final model and the second matrix is generated by a filter that uses the first array and the new observation, filter specified in the present work. The two estimators calculated for each new observation are compared by applying them with criteria established in this work, thereby generating a third which is the best estimator of the input and output model. The proposed method is called recursive least squares method 3 (MQR3). Finally, the method is tested with a MQR3 multiple time series data modeling and the results compared with the results of the RLS method. The statistics found for deferent validation data show estimates with a lower final average error for the MQR3 method / Mestrado / Automação / Mestre em Engenharia Elétrica

Orientadores: Luiz Koodi Hotta, Fernando Antônio da Silva Moura / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática, Estatística e Computação Científica / Made available in DSpace on 2018-08-20T15:43:54Z (GMT). No. of bitstreams: 1 LunaHernandez_Angela_M.pdf: 2230052 bytes, checksum: 7cee5005f5f1bd9bbdd565de481db0ad (MD5) Previous issue date: 2012 / Resumo: A dissertação apresenta o processo de estimação em pesquisas repetidas sob o enfoque de séries temporais, empregando Modelos de Espaço de Estados e o filtro dos mínimos quadrados generalizados ou filtro GLS. Este filtro permite o tratamento de modelos com erros de observação autocorrelacionados de forma mais simples do que utilizando o filtro de Kalman e, além disso, possibilita a modelagem conjunta de várias subpopulações ou domínios sob restrições de benchmark obtidas a partir da mesma pesquisa. Isto não só permite manter a coerência entre as estimativas obtidas pelo método, e estimativas agregadas baseadas no planejamento da amostra, como ajuda na proteção contra possíveis erros de especificação dos modelos. Considerando o caso de amostras com rotação, também é abordado o processo de estimação da estrutura de autocorrelação dos erros amostrais empregando o método dos pseudo-erros. Via simulação, é replicado todo o procedimento de estimação, comparando resultados obtidos empregando os filtros GLS e de Kalman. Adicionalmente, é ilustrada a aplicação do filtro sob restrições de benchmark empregando a série de taxa de desemprego da Pesquisa Mensal de Emprego do IBGE, no período de março de 2002 a fevereiro de 2012 / Abstract: This work presents the estimation process in repeated surveys using State Space Models and the generalized linear squares filter, GLS filter, under the time series approach. This filter deals with autocorrelated errors in the observation equation, in a simpler way than the well-known Kalman filter. Additionally, it allows for modeling jointly several domains under benchmark constraints obtained from the same survey. The benchmarking not only achieves coherence between the model-based estimates and the corresponding design-based aggregated estimates, but also provides protection against possible model failures. For the scenario of samples with a rotation scheme, the estimation of the autocorrelation structure of observational errors, using the pseudo-errors method, is also addressed. Simulation are used to compare the GLS and Kalman filter estimators. Moreover, the application of GLS filter under benchmark restrictions is illustrated, using the unemployment rate time serie from the Brazilian monthly labor force survey, from March 2002 to February 2012 / Mestrado / Estatistica / Mestre em Estatística

Orientador: Gilmar Barreto / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-04T17:28:20Z (GMT). No. of bitstreams: 1 Pucciarelli_AmilcarJose_M.pdf: 3241769 bytes, checksum: cfad3cee97a35c84df363b3a8578479d (MD5) Previous issue date: 2005 / Resumo: Este estudo primeiramente investiga fundamentos teóricos para análise, desenvolvimento e implementação de algoritmos para modelagem de dados de sistemas dinâmicos e de séries temporais com a finalidade de predição. As séries temporais utilizadas são baseadas em dados reais retirados da literatura. A grande vantagem de se modelar uma série temporal e de se prever um dado futuro é poder tomar ações antecipadas sobre ela o quem vem a ser muito útil, por exemplo em controle. O modelo nebuloso Takagi-Sugeno será utilizado na modelagem das séries temporais onde os conjuntos nebulosos do antecedente e os parâmetros do conseqüente são estimados via métodos de agrupamentos e identificação paramétrica, respectivamente / Abstract: This work firstly explores theoretical foundations for analisis, development and implementation of algorithms for data modelling dynamic systems and time series with a prediction goaI. The used time series are based on real data from the literature. The main advantage of time series modelling and forecasting is make antecipated decisions about it, and this becomes very useful, for example, in controI. The Takagi-Sugeno fuzzy model is used for time series modelling where the antecedent fuzzy partitions and the consequent parameters are estimated by clustering methods and parametric identification, respectively / Mestrado / Automação / Mestre em Engenharia Elétrica

This thesis is focused on practical application of two methods used in non-life insurance, Nested Monte Carlo and Least squares Monte Carlo. Best estimate and 99.5% quantile was calculated using both methods and results was compared. Both methods are similar in estimates and therefore can be used for computation of capital requirement. Least squares Monte Carlo seem more favourable, because it significantly reduces computation time.

As wireless devices become more common, the ability to position a wireless device has become a topic of importance. Accurate positioning through technologies such as the Global Positioning System is possible for outdoor environments. Indoor environments pose a different challenge, and research continues to position users indoors. Due to the prevalence of wireless local area networks (WLANs) in many indoor spaces, it is prudent to determine their capabilities for the purposes of positioning. Signal strength and time based positioning systems have been studied for WLANs. Direction or angle of arrival (AOA) based positioning will be possible with multiple antenna arrays, such as those included with upcoming devices based on the IEEE 802.11n standard. The potential performance of such a system is evaluated. The positioning performance of such a system depends on the accuracy of the AOA estimation as well as the positioning algorithm. Two different maximum-likelihood (ML) derived algorithms are used to determine the AOA of the mobile user: a specialized simple ML algorithm, and the space- alternating generalized expectation-maximization (SAGE) channel parameter estimation algorithm. The algorithms are used to determine the error in estimating AOAs through the use of real wireless signals captured in an indoor office environment. The statistics of the AOA error are used in a positioning simulation to predict the positioning performance. A least squares (LS) technique as well as the popular extended Kalman filter (EKF) are used to combine the AOAs to determine position. The position simulation shows that AOA- based positioning using WLANs indoors has the potential to position a wireless user with an accuracy of about 2 m. This is comparable to other positioning systems previously developed for WLANs. / Applied Science, Faculty of / Engineering, School of (Okanagan) / Graduate

Le mouvement des masses molles est à la fois une source d'erreur en analyse cinématique et une source d'information en analyse de la dynamique articulaire. Leur effet sur la cinématique peut être numériquement minimisé et leur dynamique estimée seulement par la simulation car aucune méthode numérique ne permet de distinguer la cinématique des masses molles de celle de l'os. Le travail présenté dans ce mémoire propose de développer une méthode numérique pour distinguer ces deux cinématiques. Une méthode d'optimisation locale a d'abord été utilisée pour évaluer le mouvement des masses molles et comparée à l'os pour valider celle-ci. Les résultats ont montré une inadaptation de la méthode locale à évaluer quantitativement et analyser le mouvement des masses molles. L'incapacité de cette méthode vient du fait qu'elle ne prend pas en compte l'ensemble des composantes du mouvement des masses molles. Un modèle numérique du membre inférieur a été développé dans la seconde étude pour considérer l'ensemble de ces composantes. Ce modèle assure le calcul de la cinématique articulaire du membre inférieur et estime un plus grand mouvement des masses molles à partir du déplacement total des marqueurs. Ce déplacement de marqueur est plus le fait d'une composante à l'unisson que d'une composante propre du mouvement des masses molles. Cette composante à l'unisson induit un mouvement commun des marqueurs par rapport à l'os. Ce mouvement commun permet ainsi de déduire la cinématique des masses molles autour des axes anatomiques des os modélisés. Cette méthode numérique permet ainsi de distinguer la cinématique de l'os de celle des masses molles offre une perspective d'étudier leur dynamique. / The movement of wobbling mass is the major source of error in kinematic analysis and a source of information in joint kinetic analysis. The effect on joint kinematic can numerically be minimized and their kinetic estimated using numerical model because there is no numerical method able to distinguish the wobbling masses kinematic from bones kinematic. The work presented in this thesis aims to develop a numerical method to distinguish those two kinematics. Firstly, a local optimisation method was used to assess the movement of wobbling mass and was compared to the bone in order to validate this numerical method. Results show maladjustment of the local method to assess quantitavely and analyze the movement of wobbling mass. The inability of this method is caused by the fact it cannot take in account all component of the movement of wobbling mass. A numerical model has been developing in the second part in order to consider all these components. This model insure similar joint kinematics, provides a bigger estimate of the movement of wobbling mass from marker displacements. This marker displacements is more induced by an unison component rather a own component of the movement of wobbling mass. The in unison component induce a common displacement of markers relative to the bones. This common movement allows to infer the kinematic of the wobbling mass in regard to anatomical axes of the modelled bones” This numerical method allows to distinguish the kinematic of bones and the kinematic of wobbling mass, and offers a perspective to investigate their kinetic.

Selection of variables and estimation of regression coefficients in datasets with the number of variables exceeding the number of observations consti- tutes an often discussed topic in modern statistics. Today the maximum penalized likelihood method with an appropriately selected function of the parameter as the penalty is used for solving this problem. The penalty should evaluate the benefit of the variable and possibly mitigate or nullify the re- spective regression coefficient. The SCAD and LASSO penalty functions are popular for their ability to choose appropriate regressors and at the same time estimate the parameters in a model. This thesis presents an overview of up to date results in the area of characteristics of estimates obtained by using these two methods for both small number of regressors and multidimensional datasets in a normal linear model. Due to the fact that the amount of pe- nalty and therefore also the choice of the model is heavily influenced by the tuning parameter, this thesis further discusses its selection. The behavior of the LASSO and SCAD penalty functions for different values and possibili- ties for selection of the tuning parameter is tested with various numbers of regressors on simulated datasets.