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

Identificação de modelos polinomiais narx utilizando algoritmos combinados de detecção de estrutura e estimação de parâmetros com aplicações práticas

A modelagem de processos industriais tem auxiliado na produção e minimização de custos, permitindo a previsão dos comportamentos futuros do sistema, supervisão de processos e projeto de controladores. Ao observar os benefícios proporcionados pela modelagem, objetiva-se primeiramente, nesta dissertação, apresentar uma metodologia de identificação de modelos não-lineares com estrutura NARX, a partir da implementação de algoritmos combinados de detecção de estrutura e estimação de parâmetros. Inicialmente, será ressaltada a importância da identificação de sistemas na otimização de processos industriais, especificamente a escolha do modelo para representar adequadamente as dinâmicas do sistema. Em seguida, será apresentada uma breve revisão das etapas que compõem a identificação de sistemas. Na sequência, serão apresentados os métodos fundamentais para detecção de estrutura (Modificado Gram- Schmidt) e estimação de parâmetros (Método dos Mínimos Quadrados e Método dos Mínimos Quadrados Estendido) de modelos. No trabalho será também realizada, através dos algoritmos implementados, a identificação de dois processos industriais distintos representados por uma planta de nível didática, que possibilita o controle de nível e vazão, e uma planta de processamento primário de petróleo simulada, que tem como objetivo representar um tratamento primário do petróleo que ocorre em plataformas petrolíferas. A dissertação é finalizada com uma avaliação dos desempenhos dos modelos obtidos, quando comparados com o sistema. A partir desta avaliação, será possível observar se os modelos identificados são capazes de representar as características estáticas e dinâmicas dos sistemas apresentados nesta dissertação

Avaliacao de dados de secagem de suspensoes de polpas de frutas em leito de jorro com alimentacao intermitente

The objective of this work was the development and improvement of the mathematical models based on mass and heat balances, representing the drying transient process fruit pulp in spouted bed dryer with intermittent feeding. Mass and energy balance for drying, represented by a system of differential equations, were developed in Fortran language and adapted to the condition of intermittent feeding and mass accumulation. Were used the DASSL routine (Differential Algebraic System Solver) for solving the differential equation system and used a heuristic optimization algorithm in parameter estimation, the Particle Swarm algorithm. From the experimental data food drying, the differential models were used to determine the quantity of water and the drying air temperature at the exit of a spouted bed and accumulated mass of powder in the dryer. The models were validated using the experimental data of drying whose operating conditions, air temperature, flow rate and time intermittency, varied within the limits studied. In reviewing the results predicted, it was found that these models represent the experimental data of the kinetics of production and accumulation of powder and humidity and air temperature at the outlet of the dryer

Avaliação e modelagem da absorção de H2S do gás natural em coluna de leito estagnado

Natural gas, although basically composed by light hydrocarbons, also presents contaminant gases in its composition, such as CO2 (carbon dioxide) and H2S (hydrogen sulfide). The H2S, which commonly occurs in oil and gas exploration and production activities, causes damages in oil and natural gas pipelines. Consequently, the removal of hydrogen sulfide gas will result in an important reduction in operating costs. Also, it is essential to consider the better quality of the oil to be processed in the refinery, thus resulting in benefits in economic, environmental and social areas. All this facts demonstrate the need for the development and improvement in hydrogen sulfide scavengers. Currently, the oil industry uses several processes for hydrogen sulfide removal from natural gas. However, these processes produce amine derivatives which can cause damage in distillation towers, can cause clogging of pipelines by formation of insoluble precipitates, and also produce residues with great environmental impact. Therefore, it is of great importance the obtaining of a stable system, in inorganic or organic reaction media, able to remove hydrogen sulfide without formation of by-products that can affect the quality and cost of natural gas processing, transport, and distribution steps. Seeking the study, evaluation and modeling of mass transfer and kinetics of hydrogen removal, in this study it was used an absorption column packed with Raschig rings, where the natural gas, with H2S as contaminant, passed through an aqueous solution of inorganic compounds as stagnant liquid, being this contaminant gas absorbed by the liquid phase. This absorption column was coupled with a H2S detection system, with interface with a computer. The data and the model equations were solved by the least squares method, modified by Levemberg-Marquardt. In this study, in addition to the water, it were used the following solutions: sodium hydroxide, potassium permanganate, ferric chloride, copper sulfate, zinc chloride, potassium chromate, and manganese sulfate, all at low concentrations (»10 ppm). These solutions were used looking for the evaluation of the interference between absorption physical and chemical parameters, or even to get a better mass transfer coefficient, as in mixing reactors and absorption columns operating in counterflow. In this context, the evaluation of H2S removal arises as a valuable procedure for the treatment of natural gas and destination of process by-products. The study of the obtained absorption curves makes possible to determine the mass transfer predominant stage in the involved processes, the mass transfer volumetric coefficients, and the equilibrium concentrations. It was also performed a kinetic study. The obtained results showed that the H2S removal kinetics is greater for NaOH. Considering that the study was performed at low concentrations of chemical reagents, it was possible to check the effect of secondary reactions in the other chemicals, especially in the case of KMnO4, which shows that your by-product, MnO2, acts in H2S absorption process. In addition, CuSO4 and FeCl3 also demonstrated to have good efficiency in H2S removal

Evolução da biodegradabilidade da matéria orgânica em um sistema de lagoas de estabilização

Waste stabilization ponds (WSP) have been widely used for sewage treatment in hot climate regions because they are economic and environmentally sustainable. In the present study a WSP complex comprising a primary facultative pond (PFP) followed by two maturation ponds (MP-1 and MP-2) was studied, in the city of Natal-RN. The main objective was to study the bio-degradability of organic matter through the determination of the kinetic constant k throughout the system. The work was carried out in two phases. In the first, the variability in BOD, COD and TOC concentrations and an analysis of the relations between these parameters, in the influent raw sewage, pond effluents and in specific areas inside the ponds was studied. In the second stage, the decay rate for organic matter (k) was determined throughout the system based on BOD tests on the influent sewage, pond effluents and water column samples taken from fixed locations within the ponds, using the mathematical methods of Least Squares and the Thomas equation. Subsequently k was estimated as a function of a hydrodynamic model determined from the dispersion number (d), using empirical methods and a Partial Hydrodynamic Evaluation (PHE), obtained from tracer studies in a section of the primary facultative pond corresponding to 10% of its total length. The concentrations of biodegradable organic matter, measured as BOD and COD, gradually reduced through the series of ponds, giving overall removal efficiencies of 71.95% for BOD and of 52.45% for COD. Determining the values for k, in the influent and effluent samples of the ponds using the mathematical method of Least Squares, gave the following values respectively: primary facultative pond (0,23 day-1 and 0,09 day-1), maturation 1 (0,04 day-1 and 0,03 day-1) and maturation 2 (0,03 day-1 and 0,08 day-1). When using the Thomas method, the values of k in the influents and effluents of the ponds were: primary facultative pond (0,17 day-1 and 0,07 day-1), maturation 1 (0,02 day-1 and 0,01 day-1) and maturation 2 (0,01 day-1 and 0,02 day-1). From the Partial Hydrodynamic Evaluation, in the first section of the facultative pond corresponding to 10% of its total length, it can be concluded from the dispersion number obtained of d = 0.04, that the hydraulic regime is one of dispersed flow with a kinetic constant value of 0.20 day-1

Integração de redes GNSS: uma proposta metodológica de densificação da rede SIRGAS na América do Sul

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Migração e seletividade na região nordeste: um estudo a partir dos dados do censo demográfico de 2010

The aim of this study is to analyze the effect of migration on the income differential between northeastern migrants and nonmigrants and there by verify that the immigrants make up a group or not positively selected. The assumption that will be tested is that the presence of these immigrants affects income inequality in the region receptor, which may explain part of the high-stopping inequality in the Brazilian Northeast. The study is based on the literature selectivity migration introduced by Roy (1951), Borjas (1987) and Chiswick (1999). Does the estimated wage equation Mincer (1974) through the method of OLS, using information from the microdata sample of the 2010 Census, the Brazilian Institute of Geography and Statistics (IBGE). The results which correspond to the comparison of socioeconomic profile, showed that immigrants are more qualified and, on average, better paid than non-migrants. With the estimation of the model, it was found that, keeping all other variables constant, the income that immigrants earn is 14.43% higher than that of non-migrants. Thus, there was existence of positive selectivity in migration directed to the Northeast

Ferramentas quimiométricas aplicadas a classificação de amostras de leite em pó e quantificação de proteínas

In this work we used chemometric tools to classify and quantify the protein content in samples of milk powder. We applied the NIR diffuse reflectance spectroscopy combined with multivariate techniques. First, we carried out an exploratory method of samples by principal component analysis (PCA), then the classification of independent modeling of class analogy (SIMCA). Thus it became possible to classify the samples that were grouped by similarities in their composition. Finally, the techniques of partial least squares regression (PLS) and principal components regression (PCR) allowed the quantification of protein content in samples of milk powder, compared with the Kjeldahl reference method. A total of 53 samples of milk powder sold in the metropolitan areas of Natal, Salvador and Rio de Janeiro were acquired for analysis, in which after pre-treatment data, there were four models, which were employed for classification and quantification of samples. The methods employed after being assessed and validated showed good performance, good accuracy and reliability of the results, showing that the NIR technique can be a non invasive technique, since it produces no waste and saves time in analyzing the samples

Construção de modelos multivariados para determinação de lipídios totais e unidade em leite em pó comercial utilizando espectroscopia no infravermelho próximo

In this work calibration models were constructed to determine the content of total lipids and moisture in powdered milk samples. For this, used the near-infrared spectroscopy by diffuse reflectance, combined with multivariate calibration. Initially, the spectral data were submitted to correction of multiplicative light scattering (MSC) and Savitzsky-Golay smoothing. Then, the samples were divided into subgroups by application of hierarchical clustering analysis of the classes (HCA) and Ward Linkage criterion. Thus, it became possible to build regression models by partial least squares (PLS) that allowed the calibration and prediction of the content total lipid and moisture, based on the values obtained by the reference methods of Soxhlet and 105 ° C, respectively . Therefore, conclude that the NIR had a good performance for the quantification of samples of powdered milk, mainly by minimizing the analysis time, not destruction of the samples and not waste. Prediction models for determination of total lipids correlated (R) of 0.9955, RMSEP of 0.8952, therefore the average error between the Soxhlet and NIR was ± 0.70%, while the model prediction to content moisture correlated (R) of 0.9184, RMSEP, 0.3778 and error of ± 0.76%

Determinação do teor de diclofenaco sódico em comprimidos por espectroscopia no infravermelho próximo NIR com calibração multivariada PLS

This work is combined with the potential of the technique of near infrared spectroscopy - NIR and chemometrics order to determine the content of diclofenac tablets, without destruction of the sample, to which was used as the reference method, ultraviolet spectroscopy, which is one of the official methods. In the construction of multivariate calibration models has been studied several types of pre-processing of NIR spectral data, such as scatter correction, first derivative. The regression method used in the construction of calibration models is the PLS (partial least squares) using NIR spectroscopic data of a set of 90 tablets were divided into two sets (calibration and prediction). 54 were used in the calibration samples and the prediction was used 36, since the calibration method used was crossvalidation method (full cross-validation) that eliminates the need for a validation set. The evaluation of the models was done by observing the values of correlation coefficient R 2 and RMSEC mean square error (calibration error) and RMSEP (forecast error). As the forecast values estimated for the remaining 36 samples, which the results were consistent with the values obtained by UV spectroscopy

Espectroscopia no infravermelho próximo e métodos de calibração multivariada aplicados à determinação simultânea de parâmetros bioquímicos em plasma sanguíneo

In this work, the quantitative analysis of glucose, triglycerides and cholesterol (total and HDL) in both rat and human blood plasma was performed without any kind of pretreatment of samples, by using near infrared spectroscopy (NIR) combined with multivariate methods. For this purpose, different techniques and algorithms used to pre-process data, to select variables and to build multivariate regression models were compared between each other, such as partial least squares regression (PLS), non linear regression by artificial neural networks, interval partial least squares regression (iPLS), genetic algorithm (GA), successive projections algorithm (SPA), amongst others. Related to the determinations of rat blood plasma samples, the variables selection algorithms showed satisfactory results both for the correlation coefficients (R²) and for the values of root mean square error of prediction (RMSEP) for the three analytes, especially for triglycerides and cholesterol-HDL. The RMSEP values for glucose, triglycerides and cholesterol-HDL obtained through the best PLS model were 6.08, 16.07 e 2.03 mg dL-1, respectively. In the other case, for the determinations in human blood plasma, the predictions obtained by the PLS models provided unsatisfactory results with non linear tendency and presence of bias. Then, the ANN regression was applied as an alternative to PLS, considering its ability of modeling data from non linear systems. The root mean square error of monitoring (RMSEM) for glucose, triglycerides and total cholesterol, for the best ANN models, were 13.20, 10.31 e 12.35 mg dL-1, respectively. Statistical tests (F and t) suggest that NIR spectroscopy combined with multivariate regression methods (PLS and ANN) are capable to quantify the analytes (glucose, triglycerides and cholesterol) even when they are present in highly complex biological fluids, such as blood plasma

Determinação de parâmetros (sólidos solúveis, pH e acidez titulável) em ameixas intactas usando espectroscopia no infravermelho próximo e seleção de comprimento de onda

The aim of this study was to evaluate the potential of near-infrared reflectance spectroscopy (NIRS) as a rapid and non-destructive method to determine the soluble solid content (SSC), pH and titratable acidity of intact plums. Samples of plum with a total solids content ranging from 5.7 to 15%, pH from 2.72 to 3.84 and titratable acidity from 0.88 a 3.6% were collected from supermarkets in Natal-Brazil, and NIR spectra were acquired in the 714 2500 nm range. A comparison of several multivariate calibration techniques with respect to several pre-processing data and variable selection algorithms, such as interval Partial Least Squares (iPLS), genetic algorithm (GA), successive projections algorithm (SPA) and ordered predictors selection (OPS), was performed. Validation models for SSC, pH and titratable acidity had a coefficient of correlation (R) of 0.95 0.90 and 0.80, as well as a root mean square error of prediction (RMSEP) of 0.45ºBrix, 0.07 and 0.40%, respectively. From these results, it can be concluded that NIR spectroscopy can be used as a non-destructive alternative for measuring the SSC, pH and titratable acidity in plums

Uso da espectroscopia do infravermelho próximo e técnicas multivariadas para diferenciar escherichia coli e salmonella enteritidis inoculadas em polpa de fruta (abacaxi)

Aiming to consumer s safety the presence of pathogenic contaminants in foods must be monitored because they are responsible for foodborne outbreaks that depending on the level of contamination can ultimately cause the death of those who consume them. In industry is necessary that this identification be fast and profitable. This study shows the utility and application of near-infrared (NIR) transflectance spectroscopy as an alternative method for the identification and classification of Escherichia coli and Salmonella Enteritidis in commercial fruit pulp (pineapple). Principal Component Analysis (PCA), Independent Modeling of Class Analogy (SIMCA) and Discriminant Analysis Partial Least Squares (PLS-DA) were used in the analysis. It was not possible to obtain total separation between samples using PCA and SIMCA. The PLS-DA showed good performance in prediction capacity reaching 87.5% for E. coli and 88.3% for S. Enteritides, respectively. The best models were obtained for the PLS-DA with second derivative spectra treated with a sensitivity and specificity of 0.87 and 0.83, respectively. These results suggest that the NIR spectroscopy and PLS-DA can be used to discriminate and detect bacteria in the fruit pulp

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

Métodos estatísticos recursivos aplicados ao problema de estimação de vazão de gás em plantas de Plunger-lift

This work has as main objective to find mathematical models based on linear parametric estimation techniques applied to the problem of calculating the grow of gas in oil wells. In particular we focus on achieving grow models applied to the case of wells that produce by plunger-lift technique on oil rigs, in which case, there are high peaks in the grow values that hinder their direct measurement by instruments. For this, we have developed estimators based on recursive least squares and make an analysis of statistical measures such as autocorrelation, cross-correlation, variogram and the cumulative periodogram, which are calculated recursively as data are obtained in real time from the plant in operation; the values obtained for these measures tell us how accurate the used model is and how it can be changed to better fit the measured values. The models have been tested in a pilot plant which emulates the process gas production in oil wells