Metodologias estatísticas aplicadas a dados de análises químicas da água produzida em um campo maduro de petróleo

Produced water is characterized as one of the most common wastes generated during exploration and production of oil. This work aims to develop methodologies based on comparative statistical processes of hydrogeochemical analysis of production zones in order to minimize types of high-cost interventions to perform identification test fluids - TIF. For the study, 27 samples were collected from five different production zones were measured a total of 50 chemical species. After the chemical analysis was applied the statistical data, using the R Statistical Software, version 2.11.1. Statistical analysis was performed in three steps. In the first stage, the objective was to investigate the behavior of chemical species under study in each area of production through the descriptive graphical analysis. The second step was to identify a function that classify production zones from each sample, using discriminant analysis. In the training stage, the rate of correct classification function of discriminant analysis was 85.19%. The next stage of processing of the data used for Principal Component Analysis, by reducing the number of variables obtained from the linear combination of chemical species, try to improve the discriminant function obtained in the second stage and increase the discrimination power of the data, but the result was not satisfactory. In Profile Analysis curves were obtained for each production area, based on the characteristics of the chemical species present in each zone. With this study it was possible to develop a method using hydrochemistry and statistical analysis that can be used to distinguish the water produced in mature fields of oil, so that it is possible to identify the zone of production that is contributing to the excessive elevation of the water volume.

Algoritmos genéticos aplicados a um comitê de LS-SVM em problemas de classificação

The pattern classification is one of the machine learning subareas that has the most outstanding. Among the various approaches to solve pattern classification problems, the Support Vector Machines (SVM) receive great emphasis, due to its ease of use and good generalization performance. The Least Squares formulation of SVM (LS-SVM) finds the solution by solving a set of linear equations instead of quadratic programming implemented in SVM. The LS-SVMs provide some free parameters that have to be correctly chosen to achieve satisfactory results in a given task. Despite the LS-SVMs having high performance, lots of tools have been developed to improve them, mainly the development of new classifying methods and the employment of ensembles, in other words, a combination of several classifiers. In this work, our proposal is to use an ensemble and a Genetic Algorithm (GA), search algorithm based on the evolution of species, to enhance the LSSVM classification. In the construction of this ensemble, we use a random selection of attributes of the original problem, which it splits the original problem into smaller ones where each classifier will act. So, we apply a genetic algorithm to find effective values of the LS-SVM parameters and also to find a weight vector, measuring the importance of each machine in the final classification. Finally, the final classification is obtained by a linear combination of the decision values of the LS-SVMs with the weight vector. We used several classification problems, taken as benchmarks to evaluate the performance of the algorithm and compared the results with other classifiers

Metodologias estatísticas aplicadas a dados de análises químicas da água produzida em um campo maduro de petróleo

Produced water is characterized as one of the most common wastes generated during exploration and production of oil. This work aims to develop methodologies based on comparative statistical processes of hydrogeochemical analysis of production zones in order to minimize types of high-cost interventions to perform identification test fluids - TIF. For the study, 27 samples were collected from five different production zones were measured a total of 50 chemical species. After the chemical analysis was applied the statistical data, using the R Statistical Software, version 2.11.1. Statistical analysis was performed in three steps. In the first stage, the objective was to investigate the behavior of chemical species under study in each area of production through the descriptive graphical analysis. The second step was to identify a function that classify production zones from each sample, using discriminant analysis. In the training stage, the rate of correct classification function of discriminant analysis was 85.19%. The next stage of processing of the data used for Principal Component Analysis, by reducing the number of variables obtained from the linear combination of chemical species, try to improve the discriminant function obtained in the second stage and increase the discrimination power of the data, but the result was not satisfactory. In Profile Analysis curves were obtained for each production area, based on the characteristics of the chemical species present in each zone. With this study it was possible to develop a method using hydrochemistry and statistical analysis that can be used to distinguish the water produced in mature fields of oil, so that it is possible to identify the zone of production that is contributing to the excessive elevation of the water volume.

Identificação e comparação entre controle preditivo com modelo não linear e PI sintonizados com PSO em sistema de separação gravitacional de águia-óleo

The separation methods are reduced applications as a result of the operational costs, the low output and the long time to separate the uids. But, these treatment methods are important because of the need for extraction of unwanted contaminants in the oil production. The water and the concentration of oil in water should be minimal (around 40 to 20 ppm) in order to take it to the sea. Because of the need of primary treatment, the objective of this project is to study and implement algorithms for identification of polynomial NARX (Nonlinear Auto-Regressive with Exogenous Input) models in closed loop, implement a structural identification, and compare strategies using PI control and updated on-line NARX predictive models on a combination of three-phase separator in series with three hydro cyclones batteries. The main goal of this project is to: obtain an optimized process of phase separation that will regulate the system, even in the presence of oil gushes; Show that it is possible to get optimized tunings for controllers analyzing the mesh as a whole, and evaluate and compare the strategies of PI and predictive control applied to the process. To accomplish these goals a simulator was used to represent the three phase separator and hydro cyclones. Algorithms were developed for system identification (NARX) using RLS(Recursive Least Square), along with methods for structure models detection. Predictive Control Algorithms were also implemented with NARX model updated on-line, and optimization algorithms using PSO (Particle Swarm Optimization). This project ends with a comparison of results obtained from the use of PI and predictive controllers (both with optimal state through the algorithm of cloud particles) in the simulated system. Thus, concluding that the performed optimizations make the system less sensitive to external perturbations and when optimized, the two controllers show similar results with the assessment of predictive control somewhat less sensitive to disturbances

Teoria de carteiras e a alocação de parques eólicos offshore

The consumption of energy on the planet is currently based on fossil fuels. They are responsible for adverse effects on the environment. Renewables propose solutions for this scenario, but must face issues related to the capacity of the power supply. Wind energy offshore emerging as a promising alternative. The speed and stability are greater winds over oceans, but the variability of these may cause inconvenience to the generation of electric power fluctuations. To reduce this, a combination of wind farms geographically distributed was proposed. The greater the distance between them, the lower the correlation between the wind velocity, increasing the likelihood that together achieve more stable power system with less fluctuations in power generation. The efficient use of production capacity of the wind park however, depends on their distribution in marine environments. The objective of this research was to analyze the optimal allocation of wind farms offshore on the east coast of the U.S. by Modern Portfolio Theory. The Modern Portfolio Theory was used so that the process of building portfolios of wind energy offshore contemplate the particularity of intermittency of wind, through calculations of return and risk of the production of wind farms. The research was conducted with 25.934 observations of energy produced by wind farms 11 hypothetical offshore, from the installation of 01 simulated ocean turbine with a capacity of 5 MW. The data show hourly time resolution and covers the period between January 1, 1998 until December 31, 2002. Through the Matlab R software, six were calculated minimum variance portfolios, each for a period of time distinct. Given the inequality of the variability of wind over time, set up four strategies rebalancing to evaluate the performance of the related portfolios, which enabled us to identify the most beneficial to the stability of the wind energy production offshore. The results showed that the production of wind energy for 1998, 1999, 2000 and 2001 should be considered by the portfolio weights calculated for the same periods, respectively. Energy data for 2002 should use the weights derived from the portfolio calculated in the previous time period. Finally, the production of wind energy in the period 1998-2002 should also be weighted by 1/11. It follows therefore that the portfolios found failed to show reduced levels of variability when compared to the individual production of wind farms hypothetical offshore

Predição não-linear de curvas de produção de petróleo via redes neurais recursivas

One of the main activities in the petroleum engineering is to estimate the oil production in the existing oil reserves. The calculation of these reserves is crucial to determine the economical feasibility of your explotation. Currently, the petroleum industry is facing problems to analyze production due to the exponentially increasing amount of data provided by the production facilities. Conventional reservoir modeling techniques like numerical reservoir simulation and visualization were well developed and are available. This work proposes intelligent methods, like artificial neural networks, to predict the oil production and compare the results with the ones obtained by the numerical simulation, method quite a lot used in the practice to realization of the oil production prediction behavior. The artificial neural networks will be used due your learning, adaptation and interpolation capabilities

Origem física das curvas sigmoidais respiratórias pressão-volume: recrutamento alveolar e elasticidade não-linear

An important unsolved problem in medical science concerns the physical origin of the sigmoidal shape of pressure volume curves of healthy (and some unhealthy) lungs. Such difficulties are expected because the lung, which is the most important structure in the respiratory system, is extremely complex. Its rheological properties are unknown and seem to depend on phenomena occurring from the alveolar scale up to the thoracic scale. Conventional wisdom holds that linear response, i.e., Hooke s law, together with alveolar overdistention, play a dominant role in respiration, but such assumptions cannot explainthe crucial empirical sigmoidal shape of the curves. In this doctorate thesis, we propose an alternative theory to solve this problem, based on the alveolar recruitment together with the nonlinear elasticity of the alveoli. This theory suggests that recruitment may be the predominant factor shaping these curves in the entire range of pressures normally employed in experiments. The proposed model correctly predicts the observed sigmoidal pressure volume curves, allowing us to discuss adequately the importance of this result, as well as its implications for medical practice

Estado nutricional antropométrico, composição corporal e pressão arterial em adolescentes

Os objetivos deste trabalho foram: (1) estimar as prevalências de excesso de peso e de gordura corporal, obesidade central e pressão arterial elevada (PAE) em adolescentes beneficiários do Programa Nacional de Alimentação Escolar (PNAE) da rede municipal de ensino de Natal-RN; (2) verificar a associação entre variáveis antropométricas e de composição corporal com a pressão arterial, a maturação sexual e a história familiar positiva de fatores de risco para doença cardiovascular (FRDCV); (3) comparar dois padrões de referência para classificação do excesso de peso em adolescentes; e (4) propor equações preditivas de massa gorda (MG) e massa livre de gordura (MLG) baseadas nos perímetros corporais. Trata-se de um estudo transversal, com 526 adolescentes beneficiários do PNAE, em Natal, Brasil. O tamanho da população de estudo foi definido por amostragem aleatória, em dois estágios, e ponderada segundo número de alunos de cada escola. No primeiro estudo, o excesso de peso foi determinado por Índice de Massa Corporal (IMC), a gordura corporal estimada por dobras cutâneas e a obesidade central por perímetro abdominal. A pressão arterial elevada foi classificada conforme a American Academy of Pediatrics. As prevalências foram apresentadas em valores relativos e efeito do desenho. Realizou-se uma análise fatorial para sintetizar o conjunto de variáveis antropométricas visando identificar fatores comuns. Extraíram-se dois fatores: (1) padrão excesso de adiposidade e (2) padrão adiposidade central elevada. Para avaliar a associação entre os padrões de adiposidade corporal com pressão arterial elevada, faixa etária, maturação sexual e história familiar de FRDCV utilizou-se a Razão de Chances e respectivo intervalo de confiança de 95% e regressão logística. No segundo estudo, calculou-se a sensibilidade e a especificidade do excesso de peso classificado segundo o IOTF e a World Health Organization WHO em relação ao excesso de adiposidade corporal; e a estatística Kappa para medir a concordância entre os dois padrões de referência. No terceiro estudo, foram elaborados modelos preditivos de MG e MLG com base em nove perímetros corporais, utilizando a bioimpedância Byodinamics 450 como padrão de referência. Para tanto foram selecionados 218 adolescentes eutróficos, segundo o IMC a partir do estudo transversal. As equações foram estimadas por regressão linear múltipla, considerando a idade e os perímetros corporais. Os resultados apontaram que 14,1% dos meninos e 15,7% das meninas tinham excesso de peso; 15,3% dos meninos e 11,6% das meninas tinham excesso de gordura corporal e dentre os meninos 14,3% tinham pressão arterial elevada e as meninas, 21,4%. Todos os efeitos do desenho foram inferiores a 2,5%. Nos meninos, o padrão excesso de adiposidade foi associado à história familiar positiva de FRDCV (ORajust=2,60; 1,09-6,22), maturação sexual (ORajust=2,92; 1,04-8,22) e PAE (ORajust=3,66; 1,34-9,94). Os meninos com 12 anos e mais apresentaram 6,1 vezes mais chance de apresentar padrão adiposidade central elevada do que os adolescentes com 10 a 11 anos (IC95% 2,32-16,04), assim como os púberes apresentaram 3,2 vezes este mesmo padrão em relação aos pré-púberes (IC95%1,14-8,85). A partir da comparação entre os dois padrões de referencia de classificação do excesso de peso por meio do IMC, observou-se que a sensibilidade foi de 79,3% para o critério IOTF e de 88,9% para WHO e a especificidade foi de 94,7% e 89,9%, respectivamente. O nível de concordância foi maior para o critério IOTF (Kappa=0,70 x Kappa=0,64). Em relação à construção das equações preditivas de gordura corporal, do total de 106 meninos e 112 meninas, foram desenvolvidas duas equações para estimar MG e duas para MLG, considerando o sexo. No sexo masculino, a equação para estimar a MG incluiu as variáveis idade, punho, quadril e perímetro abdominal (R2=0,552; AIC=416,04) e MLG, idade, punho e antebraço (R2=0,869; AIC=578,24). Enquanto que no feminino, MG foi estimada pelas variáveis punho, perímetro do abdômen, do quadril, da coxa proximal e da panturrilha (R2=0,838; AIC=415,36); e a MLG por idade, punho, perímetro do abdômen, do quadril e da panturrilha (R2=0,878; AIC=512,48). Conclui-se que os adolescentes tinham elevada prevalência de excesso de adiposidade corporal e de pressão arterial elevada. Tanto o padrão excesso de adiposidade quanto adiposidade central elevada constituem-se em padrões de risco. O padrão excesso de adiposidade foi associado à pressão arterial, história familiar positiva de FRDCV e maturação sexual em meninos. O critério IOTF mostrou-se menos sensível, mais específico, com maior nível de concordância e maior probabilidade de identificar corretamente o excesso de gordura corporal nos adolescentes avaliados. Quatro equações foram desenvolvidas para a estimativa da MG e MLG em adolescentes. As equações desenvolvidas para estimar a MG no sexo feminino e MLG para ambos os sexos apresentaram valores elevados de coeficiente de determinação ajustados e, portanto, são as preferenciais. Este estudo foi realizado com a participação de equipe multidisciplinar composta por professores da área de Nutrição, Endocrinologia Pediátrica, Estatística, Educação Física, discentes do Curso de Graduação em Nutrição e residentes em Pediatria

Mortalidade infantil e condições socioeconômicas nas microrregiões do Nordeste Brasileiro

The study aims to answer the following question: what are the different profiles of infant mortality, according to demographic, socioeconomic, infrastructure and health care, for the micro-regions at the Northeast of Brazil? Thus, the main objective is to analyze the profiles or typologies associated mortality levels sociodemographic conditions of the micro-regions, in the year 2010. To this end, the databases of birth and death certificates of SIM and SINASC (DATASUS/MS), were taken from the 2010 population Census microdata and from SIDRA/IBGE. As a methodology, a weighted multiple linear regression model was used in the analysis in order to find the most significant variables in the explanation child mortality for the year 2010. Also a cluster analysis was performed, seeking evidence, initially, of homogeneous groups of micro-regions, from of the significant variables. The logit of the infant mortality rate was used as dependent variable, while variables such as demographic, socioeconomic, infrastructure and health care in the micro-regions were taken as the independent variables of the model. The Bayesian estimation technique was applied to the database of births and deaths, due to the inconvenient fact of underreporting and random fluctuations of small quantities in small areas. The techniques of Spatial Statistics were used to determine the spatial behavior of the distribution of rates from thematic maps. In conclusion, we used the method GoM (Grade of Membership), to find typologies of mortality, associated with the selected variables by micro-regions, in order to respond the main question of the study. The results points out to the formation of three profiles: Profile 1, high infant mortality and unfavorable social conditions; Profile 2, low infant mortality, with a median social conditions of life; and Profile 3, median and high infant mortality social conditions. With this classification, it was found that, out of 188 micro-regions, 20 (10%) fits the extreme profile 1, 59 (31.4%) was characterized in the extreme profile 2, 34 (18.1%) was characterized in the extreme profile 3 and only 9 (4.8%) was classified as amorphous profile. The other micro-regions framed up in the profiles mixed. Such profiles suggest the need for different interventions in terms of public policies aimed to reducing child mortality in the region

Emprego formal na região nordeste brasileira: seus efeitos na inclusão social

This work aims to study the problem of the formal job in the Brazilian Northeast region and its effect in the social inclusion, taking for base the analysis of variables defined in the Atlas of Social Exclusion, which is based on the 2000 Brazilian Census, choosing the county as unit of analysis. As methodological options, an exploratory data analysis was performed, followed by multivariate statistical techniques, such as weighted multiple regression analysis, cluster analysis and exploratory analysis of spatial data. The results pointed out to low rates of formal job for the active age population as well as low indexes of social inclusion in the Northeast region of Brazil. A strong association of the formal job with the indicators of social inclusion under investigation, was evidenced (schooling, inequality, poverty, youth and income form government transfers), as well as a strong association of the formal job with the new index of social inclusion (IIS), modified from the IES. At the Federative Units, in which better levels of formal job had been found, good indexes of social inclusion are also observed. Highlights for the state of the Rio Grande do Norte, with the best conditions of life, and for the states of the Maranhão and Piauí, with the worst conditions. The situation of the Northeast region, facing the indicators under study, is very precarious, claiming for the necessity of emphasizing programs and governmental actions, specially directed to the raise of formal job levels of the region, reflecting, thus, in improvements on the income inequality, as well as in the social inclusion of the population of Northeastern natives.

Análise não-linear de treliças pelo método dos elementos finitos posicional

This work presents the positional nonlinear geometric formulation for trusses using different strain measures. The positional formulation presents an alternative approach for nonlinear problems. This formulation considers nodal positions as variables of the nonlinear system instead of displacements (widely found in literature). The work also describes the arc-length method used for tracing equilibrium paths with snap-through and snap-back. Numerical applications for trusses already established in the literature and comparisons with other studies are provided to prove the accuracy of the proposed formulation

Classificação de imagens de ambientes coralinos: uma abordagem empregando uma combinação de classificadores e máquina de vetor de suporte

The use of the maps obtained from remote sensing orbital images submitted to digital processing became fundamental to optimize conservation and monitoring actions of the coral reefs. However, the accuracy reached in the mapping of submerged areas is limited by variation of the water column that degrades the signal received by the orbital sensor and introduces errors in the final result of the classification. The limited capacity of the traditional methods based on conventional statistical techniques to solve the problems related to the inter-classes took the search of alternative strategies in the area of the Computational Intelligence. In this work an ensemble classifiers was built based on the combination of Support Vector Machines and Minimum Distance Classifier with the objective of classifying remotely sensed images of coral reefs ecosystem. The system is composed by three stages, through which the progressive refinement of the classification process happens. The patterns that received an ambiguous classification in a certain stage of the process were revalued in the subsequent stage. The prediction non ambiguous for all the data happened through the reduction or elimination of the false positive. The images were classified into five bottom-types: deep water; under-water corals; inter-tidal corals; algal and sandy bottom. The highest overall accuracy (89%) was obtained from SVM with polynomial kernel. The accuracy of the classified image was compared through the use of error matrix to the results obtained by the application of other classification methods based on a single classifier (neural network and the k-means algorithm). In the final, the comparison of results achieved demonstrated the potential of the ensemble classifiers as a tool of classification of images from submerged areas subject to the noise caused by atmospheric effects and the water column

Controle preditivo não linear baseado no modelo de Hammerstein com prova de estabilidade

The Predictive Controller has been receiving plenty attention in the last decades, because the need to understand, to analyze, to predict and to control real systems has been quickly growing with the technological and industrial progress. The objective of this thesis is to present a contribution for the development and implementation of Nonlinear Predictive Controllers based on Hammerstein model, as well as to its make properties evaluation. In this case, in the Nonlinear Predictive Controller development the time-step linearization method is used and a compensation term is introduced in order to improve the controller performance. The main motivation of this thesis is the study and stability guarantee for the Nonlinear Predictive Controller based on Hammerstein model. In this case, was used the concepts of sections and Popov Theorem. Simulation results with literature models shows that the proposed approaches are able to control with good performance and to guarantee the systems stability

Identificação fuzzy-multimodelos para sistemas não lineares

This paper presents a new multi-model technique of dentification in ANFIS for nonlinear systems. In this technique, the structure used is of the fuzzy Takagi-Sugeno of which the consequences are local linear models that represent the system of different points of operation and the precursors are membership functions whose adjustments are realized by the learning phase of the neuro-fuzzy ANFIS technique. The models that represent the system at different points of the operation can be found with linearization techniques like, for example, the Least Squares method that is robust against sounds and of simple application. The fuzzy system is responsible for informing the proportion of each model that should be utilized, using the membership functions. The membership functions can be adjusted by ANFIS with the use of neural network algorithms, like the back propagation error type, in such a way that the models found for each area are correctly interpolated and define an action of each model for possible entries into the system. In multi-models, the definition of action of models is known as metrics and, since this paper is based on ANFIS, it shall be denominated in ANFIS metrics. This way, ANFIS metrics is utilized to interpolate various models, composing a system to be identified. Differing from the traditional ANFIS, the created technique necessarily represents the system in various well defined regions by unaltered models whose pondered activation as per the membership functions. The selection of regions for the application of the Least Squares method is realized manually from the graphic analysis of the system behavior or from the physical characteristics of the plant. This selection serves as a base to initiate the linear model defining technique and generating the initial configuration of the membership functions. The experiments are conducted in a teaching tank, with multiple sections, designed and created to show the characteristics of the technique. The results from this tank illustrate the performance reached by the technique in task of identifying, utilizing configurations of ANFIS, comparing the developed technique with various models of simple metrics and comparing with the NNARX technique, also adapted to identification

Estrutura ANFIS modificada para identificação e controle de plantas com ampla faixa de operação e não linearidade acentuada

In this work a modification on ANFIS (Adaptive Network Based Fuzzy Inference System) structure is proposed to find a systematic method for nonlinear plants, with large operational range, identification and control, using linear local systems: models and controllers. This method is based on multiple model approach. This way, linear local models are obtained and then those models are combined by the proposed neurofuzzy structure. A metric that allows a satisfactory combination of those models is obtained after the structure training. It results on plant s global identification. A controller is projected for each local model. The global control is obtained by mixing local controllers signals. This is done by the modified ANFIS. The modification on ANFIS architecture allows the two neurofuzzy structures knowledge sharing. So the same metric obtained to combine models can be used to combine controllers. Two cases study are used to validate the new ANFIS structure. The knowledge sharing is evaluated in the second case study. It shows that just one modified ANFIS structure is necessary to combine linear models to identify, a nonlinear plant, and combine linear controllers to control this plant. The proposed method allows the usage of any identification and control techniques for local models and local controllers obtaining. It also reduces the complexity of ANFIS usage for identification and control. This work has prioritized simpler techniques for the identification and control systems to simplify the use of the method