Aplicação de redes neuronais artificiais para aproximar o comportamento de estruturas

Dissertação para obtenção do Grau de Mestre em Engenharia Mecânica

Analyzing Multiple Outcomes: Is it Really Worth the use of Multivariate Linear Regression?

In health related research it is common to have multiple outcomes of interest in a single study. These outcomes are often analysed separately, ignoring the correlation between them. One would expect that a multivariate approach would be a more efficient alternative to individual analyses of each outcome. Surprisingly, this is not always the case. In this article we discuss different settings of linear models and compare the multivariate and univariate approaches. We show that for linear regression models, the estimates of the regression parameters associated with covariates that are shared across the outcomes are the same for the multivariate and univariate models while for outcome-specific covariates the multivariate model performs better in terms of efficiency.

Estudo comparativo de painéis reforçados em liga de alumínio ou em aço de construção naval

Dissertação para obtenção do Grau de Mestre em Engenharia Mecânica

In situ measured cross section geometry of old timber structures and its influence on structural safety

Old timber structures may show significant variation in the cross section geometry along the same element, as a result of both construction methods and deterioration. As consequence, the definition of the geometric parameters in situ may be both time consuming and costly. This work presents the results of inspections carried out in different timber structures. Based on the obtained results, different simplified geometric models are proposed in order to efficiently model the geometry variations found. Probabilistic modelling techniques are also used to define safety parameters of existing timber structures, when subjected to dead and live loads, namely self-weight and wind actions. The parameters of the models have been defined as probabilistic variables, and safety of a selected case study was assessed using the Monte Carlo simulation technique. Assuming a target reliability index, a model was defined for both the residual cross section and the time dependent deterioration evolution. As a consequence, it was possible to compute probabilities of failure and reliability indices, as well as, time evolution deterioration curves for this structure. The results obtained provide a proposal for definition of the cross section geometric parameters of existing timber structures with different levels of decay, using a simplified probabilistic geometry model and considering a remaining capacity factor for the decayed areas. This model can be used for assessing the safety of the structure at present and for predicting future performance.

Reliability analysis of a timber truss system subjected to decay

Assessing the safety of existing timber structures is of paramount importance for taking reliable decisions on repair actions and their extent. The results obtained through semi-probabilistic methods are unrealistic, as the partial safety factors present in codes are calibrated considering the uncertainty present in new structures. In order to overcome these limitations, and also to include the effects of decay in the safety analysis, probabilistic methods, based on Monte-Carlo simulation are applied here to assess the safety of existing timber structures. In particular, the impact of decay on structural safety is analyzed and discussed, using a simple structural model, similar to that used for current semi-probabilistic analysis.

Correcção de Scatter em imagens de mamografia por emissão de positrões (PEM) por método de dupla janela de energia

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

Caracterização de um mamógrafo Senographe 600T Senix da GE com ânodo de Mo. dosimetria das qualidades de radiação

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

Estimação em pequenos domínios

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

Metodologia Value-at-risk: aplicação a uma carteira de obrigações de tesouro portuguesas

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

Three essays on the term structure of interest rates

A PhD Dissertation, presented as part of the requirements for the Degree of Doctor of Philosophy from the NOVA - School of Business and Economics

Gas adsorption in the MIL-53(AI) metal organic framework. Experiments and molecular simulation

Dissertação para obtenção do Grau de Doutor em Engenharia Química

Segurança estrutural em geotecnia através de métodos de base probabilística

Dissertação elaborada no Laboratório Nacional de Engenharia Civil para obtenção do Grau de Mestre em Engenharia Civil no Ramo de Geotecnia pela Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa, no âmbito do protocolo de cooperação entre a FCT/UNL e o LNEC

Hedging in a discontinuous market: the barrier option

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Finance from the NOVA – School of Business and Economics

Modelos de previsão de preços de commodities

O principal objectivo deste trabalho assenta em desenvolver modelos de previsão de preços de commodities para assim comparar a capacidade preditiva da simulação de Monte Carlo com a das redes neuronais. A simulação de Monte Carlo é principalmente utilizada para avaliar as opções, já as redes neuronais são utilizadas para fazer previsões, classificações, clustering ou aproximação de funções. Os diversos modelos desenvolvidos foram aplicados na previsão do preço futuro do milho, petróleo, ouro e cobre. Sendo que os horizontes temporais testados neste trabalho foram 1 dia, 5 dias, 20 dias e 60 dias. Através da análise do erro absoluto médio percentual (MAPE) concluiu-se que no geral o modelo individual que apresentou um melhor desempenho preditivo foram as redes neuronais. Contudo, nas previsões a 1 e a 5 dias os resultados obtidos foram semelhantes para ambos os modelos. Para se tentar melhorar os resultados obtidos pelos modelos individuais foram aplicadas algumas técnicas de combinação de modelos. A combinação de modelos demonstrou no geral capacidade para melhorar os resultados dos modelos individuais, porém apenas para o horizonte a 60 dias é que os resultados melhoraram significativamente.

Molecular simulation of gas adsorption equilibria in nanoporous materials

This work is divided into two distinct parts. The first part consists of the study of the metal organic framework UiO-66Zr, where the aim was to determine the force field that best describes the adsorption equilibrium properties of two different gases, methane and carbon dioxide. The other part of the work focuses on the study of the single wall carbon nanotube topology for ethane adsorption; the aim was to simplify as much as possible the solid-fluid force field model to increase the computational efficiency of the Monte Carlo simulations. The choice of both adsorbents relies on their potential use in adsorption processes, such as the capture and storage of carbon dioxide, natural gas storage, separation of components of biogas, and olefin/paraffin separations. The adsorption studies on the two porous materials were performed by molecular simulation using the grand canonical Monte Carlo (μ,V,T) method, over the temperature range of 298-343 K and pressure range 0.06-70 bar. The calibration curves of pressure and density as a function of chemical potential and temperature for the three adsorbates under study, were obtained Monte Carlo simulation in the canonical ensemble (N,V,T); polynomial fit and interpolation of the obtained data allowed to determine the pressure and gas density at any chemical potential. The adsorption equilibria of methane and carbon dioxide in UiO-66Zr were simulated and compared with the experimental data obtained by Jasmina H. Cavka et al. The results show that the best force field for both gases is a chargeless united-atom force field based on the TraPPE model. Using this validated force field it was possible to estimate the isosteric heats of adsorption and the Henry constants. In the Grand-Canonical Monte Carlo simulations of carbon nanotubes, we conclude that the fastest type of run is obtained with a force field that approximates the nanotube as a smooth cylinder; this approximation gives execution times that are 1.6 times faster than the typical atomistic runs.