CoordenaÃÃo de AperfeiÃoamento de NÃvel Superior / The wind resource assessment is a key step in the development of wind power generation projects. Minimizing errors in this step brings significant reliability gains for the whole project. In this study we sought a reduction in the errors associated with the adjustment of the Weibull distribution with respect to data collected by an improvement of this setting. With this purpose, it was applied the optimization algorithm of Harmonic Search (HS) to find the Weibull distribution parameters with the best fit. The HS was used to find the Weibull distribution parameters for two sets of data from the Sistema de OrganizaÃÃo de Dados Ambientais (SONDA) of the cities of Petrolina-PE and SÃo Martinho da Serra-RS. The parameters of the HS were selected by two methods, one being a result of the novel combination of two other already presented in the literature. We therefore compared the errors for each one to determine which method provides better optimization. For evaluating the quality setting, the root mean square error (RMSE) and the correlation coefficient were used. The HS-PA method uses a selection of random parameters but, results showed more stable than the IHS. It was found for the IHS method one RMSE = 0.006418 for Petrolina and RMSE = 0.008303 for SÃo Martinho da Serra. The HS-PA method presented a RMSE = 0.006419 for Petrolina and RMSE = 0.008303 for SÃo Martinho da Serra. The RMSE values for the traditional methods applied to the same data set, there were two or more times greater than those found by employing the Harmonic search. / A anÃlise de recurso eÃlico à uma etapa fundamental no desenvolvimento de projetos de geraÃÃo de energia eÃlica. A minimizaÃÃo dos erros nesta etapa traz ganhos significativos de confiabilidade para o projeto como um todo. Neste trabalho foi buscada uma diminuiÃÃo dos erros associados ao ajuste da distribuiÃÃo de Weibull em relaÃÃo aos dados coletados atravÃs de uma melhora deste ajuste. Visando tal intento, foi aplicado o algoritmo de otimizaÃÃo da Busca HarmÃnica (HS) para encontrar os parÃmetros da distribuiÃÃo de Weibull com o melhor ajuste. A HS foi utilizada para encontrar os parÃmetros da distribuiÃÃo de Weibull para dois conjuntos de dados provenientes do Sistema de OrganizaÃÃo de Dados ambientais (SONDA) das cidades de Petrolina-PE e SÃo Martinho da Serra-RS. Os parÃmetros da HS foram selecionados atravÃs de duas metodologias, sendo uma delas inovadora por resultar da combinaÃÃo de outras duas jà apresentadas anteriormente em literatura. Foram, portanto, comparados os erros referentes a cada uma para determinar qual mÃtodo fornecia uma melhor otimizaÃÃo. Para a avaliaÃÃo da qualidade do ajuste, foram utilizados o erro mÃdio quadrÃtico (RMSE) e o coeficiente de correlaÃÃo. O mÃtodo HS-PA, embora utilize uma seleÃÃo de parÃmetros aleatÃria, apresentou resultados mais estÃveis do que o IHS. Foi encontrado para o mÃtodo IHS um RMSE = 0,006418 para Petrolina e RMSE = 0,008303 para SÃo Martinho da Serra. O mÃtodo HS-PA apresentou um RMSE = 0,006419 para Petrolina e RMSE = 0,008303 para SÃo Martinho da Serra. Os valores de RMSE para os mÃtodos tradicionais aplicados ao mesmo conjunto de dados, foram duas ou mais vezes maiores do que os encontrados, empregando a Busca HarmÃnica.

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / In recent years wind energy is becoming increasingly competitive on the world stage, making their participation in the electricity generation matrix presents a strong growth expectation. This dissertation initially presents an analysis of the behavior of wind at three locations in Northeast Brazil (Maracanau-CE, Petrolina-PE e Parnaiba-PI). In a second step, statistical analyzes are researched to the most appropriate behavior patterns of the observed wind resource in the three localities. In conclusion, the impact of the statistical analyzes used in the production of electricity from wind turbines is identified. In this study,historicaldata of speed and direction of wind collectedare used, during periods of: February 2012 to January 2013, to Maracanau; August 2012 to July 2013, for Parnaiba; and May 2012 to March 2013, for Petrolina. The Weibull distribution is applied to approximate the histograms of wind speed using different horizons of applications (annual, semiannual) and four different numerical methods (Empirical, Momentum, Energy Pattern Factor and Equivalent Energy) for estimation of the form and scale parameters. In addition to evaluating the application of Weibull, other frequency distributions (Normal, Gamma and Log-Normal)are analyzed, in order to obtain the best possible fit. In a last step, with the aid of RETScreen program,annual production of electricity, delivered to the grid from wind turbines,is calculated. The optimum wind speed occurred in Parnaiba (10 and 11 m / s), followed by Petrolina (8 and 9 m / s). Among all different numerical methods that was evaluated, the Equivalent Energy method presented the best performance, unlike the Energy Pattern Factor method, that presented the worst. The Weibull distribution showed good potential for setting wind data in Maracanau and Parnaiba, both located along the coastline. However, based on the wind data recorded, in Petrolina, which is located further inland, the performance was inferior. Among all the different frequency distributions that were verified, only normal distribution had an fit as good as Weibull distribution. Based on the annual electricity production estimation, Parnaiba is the city that has the best potential for energy production. / Nos Ãltimos anos a energia eÃlica tem se tornando cada vez mais competitiva no cenÃrio mundial, fazendo com que sua participaÃÃo na matriz elÃtrica apresente uma forte expectativa de crescimento. A presente dissertaÃÃo apresenta inicialmente uma anÃlise do comportamento do vento em trÃs localidades no Nordeste do Brasil (Maracanaà (CE), Petrolina (PE) e ParnaÃba (PI)); numa segunda etapa, sÃo pesquisadas anÃlises estatÃsticas mais adequadas aos padrÃes de comportamento do recurso eÃlico observado nas trÃs localidades e, concluindo, à identificado o impacto das anÃlises estatÃsticas utilizadas na produÃÃo de eletricidade de aerogeradores. Neste estudo sÃo utilizados dados histÃricos de velocidade e direÃÃo do vento coletados durante os perÃodos de: fevereiro de 2012 - janeiro de 2013 para MaracanaÃ, Agosto de 2012 - Julho de 2013 para ParnaÃba, maio de 2012 - marÃo 2013 para Petrolina. A distribuiÃÃo de frequÃncia de Weibull à aplicada para aproximar os histogramas de velocidade do vento, utilizando diferentes horizontes de aplicaÃÃes (anual, semestral) e quatro diferentes mÃtodos numÃricos (EmpÃrico, Momento, Fator PadrÃo de Energia e Energia Equivalente) para a estimaÃÃo dos parÃmetros de forma e escala. AlÃm de avaliar a aplicaÃÃo de Weibull, sÃo analisadas outras distribuiÃÃes de frequÃncia (Normal, Gama e Log-Normal) objetivando obter o melhor ajuste possÃvel. Numa Ãltima etapa, com o auxÃlio do programa RETScreen,à calculada a produÃÃo de eletricidade anual entregue à rede a partir de aerogeradores. Os melhores valores de velocidade do vento ocorreram em ParnaÃba (10 e 11 m/s), seguido de Petrolina (8 e 9 m/s). Dos diferentes mÃtodos numÃricos avaliados, o mÃtodo de energia equivalente apresentou o melhor desempenho e o mÃtodo fator de padrÃo de energia foi o mÃtodo com o pior desempenho. A distribuiÃÃo de Weibull demonstrou bom potencial para o ajuste de dados de vento em Maracanaà e ParnaÃba, ambas localizadas ao longo do litoral. No entanto, em Petrolina, que està situada mais para o interior, foi verificado um desempenho limitado a partir dos dados de vento registrados. Das diferentes distribuiÃÃes de frequÃncias testadas, apenas a distribuiÃÃo normal apresenta um ajuste aproximado ao que Weibull permite desenvolver. Com base nas estimaÃÃes da produÃÃo de eletricidade anual, ParnaÃba à a cidade que apresenta o melhor potencial para o aproveitamento eolioelÃtrico.

Submitted by (ana.araujo@ufrpe.br) on 2016-07-05T18:47:06Z No. of bitstreams: 1 Felipe Ricardo Santos de Gusmao.pdf: 1373394 bytes, checksum: 39658b579d3dd694736ba6792319191b (MD5) / Made available in DSpace on 2016-07-05T18:47:06Z (GMT). No. of bitstreams: 1 Felipe Ricardo Santos de Gusmao.pdf: 1373394 bytes, checksum: 39658b579d3dd694736ba6792319191b (MD5) Previous issue date: 2008-12-19 / The distribution inverse Weibull is suitable for modeling failure rates which are quite common in reliability and biological studies. In this work a new three-parameter distribution generalized inverse Weibull with decreasing and unimodal failure rate is introduced.We provide a comprehensive treatment of the mathematical properties of the generalized inverse Weibull and derive expressions for its moment generating function and the rth generalized moment. We also discuss maximum likelihood estimation and we provide formulae for the elements of the Observed information matrix, we also made an bayesian approach for this new distribution and an applied was made for a real data set for the methods classicand bayes. / A distribuição Weibull inversa tem a habilidade de modelar funções de risco com forma unimodal que são bastante comuns em estudos biológicos e de confiabilidade. Uma nova distribuição Weibull inversa generalizada tri-paramétrica com taxa de falha decrescente e unimodal é proposta. Um compreensivo tratamento das propriedades matemáticas de Weibull inversa generalizada é provido e foi encontrado expressões para suas funções geradoras de momentos e o r-ésimo momento generalizado foi determinado. Também discutimos a estimação de máxima verossimilhança e as fórmulas para os elementos da matriz de informação observada. Uma abordagem bayesiana para esta nova distribuição foi proposta e exemplificada, modelando um conjunto de dados agrários pelos métodos clássico e bayesiano.

Errors of computational simulations in design of a high-speed civil transport (HSCT) are investigated. First, discretization error from a supersonic panel code, WINGDES, is considered. Second, convergence error from a structural optimization procedure using GENESIS is considered along with the Rosenbrock test problem. A grid converge study is performed to estimate the order of the discretization error in the lift coefficient (CL) of the HSCT calculated from WINGDES. A response surface (RS) model using several mesh sizes is applied to reduce the noise magnification problem associated with the Richardson extrapolation. The RS model is shown to be more efficient than Richardson extrapolation via careful use of design of experiments. A programming error caused inaccurate optimization results for the Rosenbrock test function, while inadequate convergence criteria of the structural optimization produced error in wing structural weight of the HSCT. The Weibull distribution is successfully fit to the optimization errors of both problems. The probabilistic model enables us to estimate average errors without performing very accurate optimization runs that can be expensive, by using differences between two sets of results with different optimization control parameters such as initial design points or convergence criteria. Optimization results with large errors, outliers, produced inaccurate RS approximations. A robust regression technique, M-estimation implemented by iteratively reweighted least squares (IRLS), is used to identify the outliers, which are then repaired by higher fidelity optimizations. The IRLS procedure is applied to the results of the Rosenbrock test problem, and wing structural weight from the structural optimization of the HSCT. A nonsymmetric IRLS (NIRLS), utilizing one-sidedness of optimization errors, is more effective than IRLS in identifying outliers. Detection and repair of the outliers improve accuracy of the RS approximations. Finally, configuration optimizations of the HSCT are performed using the improved wing bending material weight RS models. / Ph. D.

No description available.

Product reliability is an important characteristic for all manufacturers, engineers and consumers. Industrial statisticians have been planning experiments for years to improve product quality and reliability. However, rarely do experts in the field of reliability have expertise in design of experiments (DOE) and the implications that experimental protocol have on data analysis. Additionally, statisticians who focus on DOE rarely work with reliability data. As a result, analysis methods for lifetime data for experimental designs that are more complex than a completely randomized design are extremely limited. This dissertation provides two new analysis methods for reliability data from life tests. We focus on data from a sub-sampling experimental design. The new analysis methods are illustrated on a popular reliability data set, which contains sub-sampling. Monte Carlo simulation studies evaluate the capabilities of the new modeling methods. Additionally, Monte Carlo simulation studies highlight the principles of experimental design in a reliability context. The dissertation provides multiple methods for statistical inference for the new analysis methods. Finally, implications for the reliability field are discussed, especially in future applications of the new analysis methods. / Ph. D.

When a reliability experiment is used, practitioners can understand better what lifetimes to expect of a product under different operating conditions and what factors are important to designing reliability into a product. Reliability experiments, however, can be very challenging to analyze because often the reliability or lifetime data tend to follow distinctly non-normal distributions and the experiments typically involve censoring. Time and cost constraints may also lead to reliability experiments with experimental protocols that are not completely randomized. In many industrial experiments, for example, the split-plot structure arises when the randomization of the experimental runs is restricted. Additionally, for many reliability experiments, it is often cost effective to apply a treatment combination to a stand with multiple units on it as opposed to each unit individually, which introduces subsampling. The analysis of lifetime data assuming a completely randomized design has been well studied, but until recently analysis methodologies for more complex experimental designs with multiple error terms have not been a focus of the reliability field. This dissertation provides two analysis methods for analyzing right-censored Weibull distributed lifetime data from a split-plot experiment with subsampling. We evaluate the proposed methods through a simulation study. Companies also routinely perform life tests on their products to ensure that products meet requirements. Each of these life tests typically involves testing several units simultaneously with interest in the times to failure. Again, the fact that lifetime data tend to be nonnormally distributed and censored make the development of a control charting procedure more demanding. In this dissertation, one-sided lower and upper likelihood ratio based cumulative sum (CUSUM) control charting procedures are developed for right-censored Weibull lifetime data to monitor changes in the scale parameter, also known as the characteristic life, for a fixed value of the Weibull shape parameter. Because a decrease in the characteristic life indicates a decrease in the mean lifetime of a product, a one-sided lower CUSUM chart is the main focus. We illustrate the development and implementation of the chart and evaluate the properties through a simulation study. / Ph. D.

Yes / Time, cost and risk are the main elements that effect the operating margin of the oil and gas companies due to Turnaround Maintenance (TAM). Turnaround Maintenance (TAM) is a methodology for the total shutdown of plant facilities during a pre-defined period to execute inspection actions, replacement and repairs according to Scope of Work (SoW). This paper presents a new methodology for improving TAM scheduling of oil and gas plants. The methodology includes four stages: removing Non-critical Equipment (NE) from reactive maintenance to proactive maintenance, risk-based inspection of Critical Static Equipment (CSE), risk-based failure of Critical Rotating Equipment (CRE), and application of failure distributions. The results from improving TAM scheduling is associated with decreasing duration and increasing interval between TAM leading to improved availability, reliability, operation and maintenance costs and safety risks. The paper presents findings from the TAM model application. The methodology is fairly generic in its approach and can also be adapted for implementation in other oil and gas industries that operate under similar harsh conditions.

[Truncated abstract] Rocks, as the host to natural chains of coupled thermal, hydraulic and mechanical processes, are heterogeneous at a variety of length scales, and in their mechanical properties, as well as in the hydraulic and thermal transport properties. Rock heterogeneity affects the ultimate hydro-carbon recovery or geothermal energy production. This heterogeneity has been considered one important and difficult problem that needs to be taken into account for its effect on the coupled processes. The aim of this thesis is to investigate the effect of rock heterogeneity on multi-physical processes. A fully coupled finite element model, hereinafter referred to as a porosity-based model (PBM) was developed to characterise the thermal-hydraulic-mechanical (THM) coupling processes. The development of the PBM consists of a two-staged workflow. First, based on poromechanics, porosity, one of the inherent rock properties, was derived as a variant function of the thermal, hydraulic and mechanical effects. Then, empirical relations or experimental results, correlating porosity with the mechanical, hydraulic and thermal properties, were incorporated as the coupling effects. In the PBM, the bulk volume of the model is assumed to be changeable. The rate of the volumetric strain was derived as the difference of two parts: the first part is the change in volume per unit of volume and per unit of time (this part was traditionally considered the rate of volumetric strain); and the second is the product of the first part and the volumetric strain. The second part makes the PBM a significant advancement of the models reported in the literature. ... impact of the rock heterogeneity on the hydro-mechanical responses because of the requirement of large memory and long central processing unit (CPU) time for the 3D applications. In the 2D PBM applications, as the thermal boundary condition applied to the rock samples containing some fractures, the pore pressure is generated by the thermal gradient. Some pore pressure islands can be generated as the statistical model and the digital image model are applied to characterise the initial porosity distribution. However, by using the homogeneous model, this phenomenon cannot be produced. In the 3D PBM applications, the existing fractures become the preferential paths for the fluid flowing inside the numerical model. The numerical results show that the PBM is sufficiently reliable to account for the rock mineral distribution in the hydro-mechanical coupling processes. The applications of the statistical method and the digital image processing technique make it possible to visualise the rock heterogeneity effect on the pore pressure distribution and the heat dissipation inside the rock model. Monitoring the fluid flux demonstrates the impact of the rock heterogeneity on the fluid product, which concerns petroleum engineering. The overall fluid flux (OFF) is mostly overestimated when the rock and fluid properties are assumed to be homogeneous. The 3D PBM application is an example. As the rock is heterogeneous, the OFF by the digital core is almost the same as that by the homogeneous model (this is due to that some fractures running through the digital core become the preferential path for the fluid flow), and around 1.5 times of that by the statistical model.

Neste trabalho propomos um modelo de regress~ao utilizando a distribuição Weibull modificado, esta distribuição pode ser usada para modelar dados de sobrevivência quando a de função de risco tem forma de U ou banheira. Assumindo dados censurados, é considerado os estimadores de máxima verossimilhança e Jackknife para os parâmetros do modelo proposto. Foram derivadas as matrizes apropriadas para avaliar influiência local sobre os parâmetros estimados considerando diferentes peturbações e também é apresen- tada alguma medidas de influência global. Para diferentes parâmetros fixados, tamanhos de amostra e porcentagem de censuras, varia simulações foram feitas para avaliar a distribuição empírica do resíduo deviance modificado e comparado coma distribuição normal padrão. Esses estudos sugerem que a distribuição empírica do resíduo devianve modificado para o modelo de regressão log-Weibull modificado com dados censurados aproxima-se de uma dis- tribuição normal padrão. Finalmente analisamos um conjunto de dados utilizando o modelo de regressão log-Weibull modificado. Uma nova distribuição de quatro parâmetros é definida para modelar dados de tempo de vida. Algumas propriedades da distribuição é discutida, assim como ilustramos com exemplos a aplicação dessa nova distribuição. Palavras-chaves: Modelo de regressão; Distribuição Weibull modificada; Distribuição weibull modificada generalizada; Análise de sensibilidade; Dados censurados; Análise de resíduo / In this paperwork are proposed a regression model considering the modified Weibull distribution. This distribution can be used to model bathtub-shaped failure rate functions. Assuming censored data, we consider a classic and Jackknife estimator for the parameters of the model. We derive the appropriate matrices for assessing local influence on the parameter estimates under diferent perturbation schemes and we also present some ways to perform global influence. Besides, for diferent parameter settings, sample sizes and censoring percentages, various simulations are performed and the empirical distribution of the deviance modified residual is displayed and compared with the standard normal distribution. These studies suggest that the residual analysis usually performed in normal linear regression models can be straightforwardly extend for a martingale-type residual in log-modifiedWeibull regression models with censored data. Finally, we analyze a real data set under log-modified Weibull regression models. A diagnostic analysis and a model checking based on the deviance modified residual are performed to select an appropriate model. A new four-parameter distribution is introduced. Various properties the new distribution are discussed. Illustrative examples based on real data are also given.