Identificação não linear usando uma rede fuzzy wavelet neural network modificada

In last decades, neural networks have been established as a major tool for the identification of nonlinear systems. Among the various types of networks used in identification, one that can be highlighted is the wavelet neural network (WNN). This network combines the characteristics of wavelet multiresolution theory with learning ability and generalization of neural networks usually, providing more accurate models than those ones obtained by traditional networks. An extension of WNN networks is to combine the neuro-fuzzy ANFIS (Adaptive Network Based Fuzzy Inference System) structure with wavelets, leading to generate the Fuzzy Wavelet Neural Network - FWNN structure. This network is very similar to ANFIS networks, with the difference that traditional polynomials present in consequent of this network are replaced by WNN networks. This paper proposes the identification of nonlinear dynamical systems from a network FWNN modified. In the proposed structure, functions only wavelets are used in the consequent. Thus, it is possible to obtain a simplification of the structure, reducing the number of adjustable parameters of the network. To evaluate the performance of network FWNN with this modification, an analysis of network performance is made, verifying advantages, disadvantages and cost effectiveness when compared to other existing FWNN structures in literature. The evaluations are carried out via the identification of two simulated systems traditionally found in the literature and a real nonlinear system, consisting of a nonlinear multi section tank. Finally, the network is used to infer values of temperature and humidity inside of a neonatal incubator. The execution of such analyzes is based on various criteria, like: mean squared error, number of training epochs, number of adjustable parameters, the variation of the mean square error, among others. The results found show the generalization ability of the modified structure, despite the simplification performed

Modelos de dispositivos de microondas e ópticos através de redes neurais artificiais de alimentação direta

This dissertation contributes for the development of methodologies through feed forward artificial neural networks for microwave and optical devices modeling. A bibliographical revision on the applications of neuro-computational techniques in the areas of microwave/optical engineering was carried through. Characteristics of networks MLP, RBF and SFNN, as well as the strategies of supervised learning had been presented. Adjustment expressions of the networks free parameters above cited had been deduced from the gradient method. Conventional method EM-ANN was applied in the modeling of microwave passive devices and optical amplifiers. For this, they had been proposals modular configurations based in networks SFNN and RBF/MLP objectifying a bigger capacity of models generalization. As for the training of the used networks, the Rprop algorithm was applied. All the algorithms used in the attainment of the models of this dissertation had been implemented in Matlab

Classificação de padrões através de um comitê de máquinas aprimorado por aprendizagem por reforço

Reinforcement learning is a machine learning technique that, although finding a large number of applications, maybe is yet to reach its full potential. One of the inadequately tested possibilities is the use of reinforcement learning in combination with other methods for the solution of pattern classification problems. It is well documented in the literature the problems that support vector machine ensembles face in terms of generalization capacity. Algorithms such as Adaboost do not deal appropriately with the imbalances that arise in those situations. Several alternatives have been proposed, with varying degrees of success. This dissertation presents a new approach to building committees of support vector machines. The presented algorithm combines Adaboost algorithm with a layer of reinforcement learning to adjust committee parameters in order to avoid that imbalances on the committee components affect the generalization performance of the final hypothesis. Comparisons were made with ensembles using and not using the reinforcement learning layer, testing benchmark data sets widely known in area of pattern classification

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

Utilização de RNA´s na construção do diagrama de vida constante de probabilidade de materiais compósitos

The static and cyclic assays are common to test materials in structures.. For cycling assays to assess the fatigue behavior of the material and thereby obtain the S-N curves and these are used to construct the diagrams of living constant. However, these diagrams, when constructed with small amounts of S-N curves underestimate or overestimate the actual behavior of the composite, there is increasing need for more testing to obtain more accurate results. Therewith, , a way of reducing costs is the statistical analysis of the fatigue behavior. The aim of this research was evaluate the probabilistic fatigue behavior of composite materials. The research was conducted in three parts. The first part consists of associating the equation of probability Weilbull equations commonly used in modeling of composite materials S-N curve, namely the exponential equation and power law and their generalizations. The second part was used the results obtained by the equation which best represents the S-N curves of probability and trained a network to the modular 5% failure. In the third part, we carried out a comparative study of the results obtained using the nonlinear model by parts (PNL) with the results of a modular network architecture (MN) in the analysis of fatigue behavior. For this we used a database of ten materials obtained from the literature to assess the ability of generalization of the modular network as well as its robustness. From the results it was found that the power law of probability generalized probabilistic behavior better represents the fatigue and composites that although the generalization ability of the MN that was not robust training with 5% failure rate, but for values mean the MN showed more accurate results than the PNL model

Recuperação e purificação de quitosanases usando adsorção em leito expandido com streamline DEAE com modelagem e simulação usando redes neurais

Expanded Bed Adsorption (EBA) is an integrative process that combines concepts of chromatography and fluidization of solids. The many parameters involved and their synergistic effects complicate the optimization of the process. Fortunately, some mathematical tools have been developed in order to guide the investigation of the EBA system. In this work the application of experimental design, phenomenological modeling and artificial neural networks (ANN) in understanding chitosanases adsorption on ion exchange resin Streamline® DEAE have been investigated. The strain Paenibacillus ehimensis NRRL B-23118 was used for chitosanase production. EBA experiments were carried out using a column of 2.6 cm inner diameter with 30.0 cm in height that was coupled to a peristaltic pump. At the bottom of the column there was a distributor of glass beads having a height of 3.0 cm. Assays for residence time distribution (RTD) revelead a high degree of mixing, however, the Richardson-Zaki coefficients showed that the column was on the threshold of stability. Isotherm models fitted the adsorption equilibrium data in the presence of lyotropic salts. The results of experiment design indicated that the ionic strength and superficial velocity are important to the recovery and purity of chitosanases. The molecular mass of the two chitosanases were approximately 23 kDa and 52 kDa as estimated by SDS-PAGE. The phenomenological modeling was aimed to describe the operations in batch and column chromatography. The simulations were performed in Microsoft Visual Studio. The kinetic rate constant model set to kinetic curves efficiently under conditions of initial enzyme activity 0.232, 0.142 e 0.079 UA/mL. The simulated breakthrough curves showed some differences with experimental data, especially regarding the slope. Sensitivity tests of the model on the surface velocity, axial dispersion and initial concentration showed agreement with the literature. The neural network was constructed in MATLAB and Neural Network Toolbox. The cross-validation was used to improve the ability of generalization. The parameters of ANN were improved to obtain the settings 6-6 (enzyme activity) and 9-6 (total protein), as well as tansig transfer function and Levenberg-Marquardt training algorithm. The neural Carlos Eduardo de Araújo Padilha dezembro/2013 9 networks simulations, including all the steps of cycle, showed good agreement with experimental data, with a correlation coefficient of approximately 0.974. The effects of input variables on profiles of the stages of loading, washing and elution were consistent with the literature

Modelagem do comportamento à fadiga de compósitos de fibra de vidro a partir de um modelo misto de RNA

This work consists basically in the elaboration of an Artificial Neural Network (ANN) in order to model the composites materials’ behavior when submitted to fatigue loadings. The proposal is to develop and present a mixed model, which associate an analytical equation (Adam Equation) to the structure of the ANN. Given that the composites often shows a similar behavior when subject to float loadings, this equation aims to establish a pre-defined comparison pattern for a generic material, so that the ANN fit the behavior of another composite material to that pattern. In this way, the ANN did not need to fully learn the behavior of a determined material, because the Adam Equation would do the big part of the job. This model was used in two different network architectures, modular and perceptron, with the aim of analyze it efficiency in distinct structures. Beyond the different architectures, it was analyzed the answers generated from two sets of different data – with three and two SN curves. This model was also compared to the specialized literature results, which use a conventional structure of ANN. The results consist in analyze and compare some characteristics like generalization capacity, robustness and the Goodman Diagrams, developed by the networks.

Caracterização limnológica e determinação da capacidade suporte do reservatório mendubim, (Rio Grande do Norte) para o cultivo de peixes em tanques-rede

The aims of this study were: i) assessing the trophic state of the Mendubim reservoir (semi-arid, Rio Grande do Norte, Brazil; 05° 38 99,0 S 36°55 98,0 W) based on chlorophyll-a, total phosphorus and nitrogen concentrations and water transparency; ii) relating the patterns of temporal variation of zooplankton and phytoplankton to the trophic state of the reservoir and iii) investigating the carrying capacity of the reservoir for cage fish farming. The samplingwas done monthly from July 2006 to July 2007 in three stations at the reservoir: next to the dam (barrage), in the central region and in the mouth of the main tributary. The abiotic and biotic variables analyzed were: Secchi depth, volatiles and fixed suspended solids, chlorophyll-a, total phosphorus and nitrogen, TN:TP ratio and mesozooplankton and phytoplankton composition and biomass. The results showed that the reservoir can be considered as mesotrophic with mean concentrations of total nitrogen, phosphorus and chlorophyll-a equal to 1711, 1 μg.L-1, 30,8 μg.L-1 and 5,62 μg.L-1 respectively. The Cyanophyceae class was the most representative in terms of density, with the presence of potentially toxic species such as Microcystis aeruginosa, Planktothrix planctonica, Cylindrospermopsis raciborskii, Aphanizomenon sp. ,Aphanocapsa delicatissima and Pseudanabaena acicularis. Among the zooplankton, the genus Notodiaptomus presented the largest biomass values. Overall, our results show that the light limitation should explain the weak relationship between chlorophyll-a and total phosphorus and nitrogen concentrations. We concluded that the water of Mendubim reservoir is suitable for intensive fish cage aquaculture. Based on the carrying capacity calculations for this reservoir, we found that the maximum sustainable yield of tilapias in cages in the reservoir is 126 ton per year assuming a factor of food conversion of 1.5: 1.0 and a phosphorus content in the fish food of 1%

Eutrofização e capacidade de carga de fósforo de seis reservatórios a bacia do Rio Seridó, região semi-árida do Estado do RN

The eutrofization is a natural process of accumulation of nutrients in aquatic´s body that it has been accelerated for the human´s actives, mainly the related with the activities of camp, industrial and the inadequate disposition of the domestic sewage. The enrichment of the aquatic´s body with nutrients, mainly the nitrogen and the phosphorus, and the consequent proliferation of algae and Cyanobacteria can commit the quality of the water for the public provisioning, for the fish farming and for other ends. The problem becomes more critical when there is a shortage of water naturally as in the semi-arid area of the Brazilian northeast. Before that problem this work had as objective evaluates the trophic state of six reservoirs of the basin of River Seridó of Rio Grande of Norte and also estimate the capacity of load of match of the reservoir and risk probabilities based on the established limits by the resolution Conama 357/05. The results demonstrate that the six reservoirs are eutrofization, with concentration of total phosphorus and cloro a in the water upster to 50 e 12 μg l-1. The results show that space homogeneity exists in the state trophic of the reservoirs, but a significant variation interanual in function of the increase of the concentrations of nutrients and decrease of the transparency of the water with the reduction of the body of water accumulated in the reservoirs.The results of the simulation risk estocastic show that the reservoirs could receive annually from 72 to 216 Kg of P, assuming a risk of 10% of increasing in more than 30 μg l-1 the annual medium concentrations of total match in the water of these reservoirs. This load could be high in until 360 kg of P a year in case the managers assume a risk of 10% of increasing in more than 50 μg l-1 the annual medium concentrations of total phosphorus in the waters of these reservoirs

Um Modelo híbrido para previsão de curvas de produção de petróleo

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Influência da dor na capacidade funcional de idosos institucionalizados

The aim of this study was to evaluate the pain influence regarding location, intensity and duration over functional capacity in institutionalized elderly (Fundacao Leur Brito). This is a descriptive exploratory study with a transversal design and quantitative approach. The population of the study was composed of 60 elderly, being 50% males with 60 years-old minimum age and 104 years-old maximum age, with mean age 77,6 (?}11,64) years-old. Data was collected by a structured query formulary divided in four parts: 1) Social, health and demographic characterization; 2) Pain related aspects (time and location); 3) Pain evaluation by numeric scale and 4) Functional capacity measured by Barthel Index. It was observed after evaluation by Squared-chi test (x2) statistical significant difference between pain presence and Barthel Index activities: bath (pvalue=0,015), dressing (p-value= 0,041), intimal hygiene transference (p-value=0,001), chair and bed transference (p-value=0,032), walking (p-value=0,010) and go upstairs (p-value=0,008). It was also observed statistical difference between total Barthel score (dependent/independent) and pain presence, p-value<0, 000. Through data obtained by this study, a multiprofessional approach is necessary to proper pain control and maintenance and/or regain of functional capacity, leading to an increase in life quality with more independence and autonomy to elderly

Relação entre ambiente familiar e capacidade funcional de idosos residentes no município de Jequié-BA

(Objective) Assess the functional capacity and determine the difference between the means of functional capacity (basic and instrumental activities of daily living) and the age groups of elderly residents in an outlying area in the hinterland of Bahia/Northeast of Brazil. (Methods) Analytical study with cross-sectional design and a sample of 150 elderly individuals enrolled in four Health Units in the municipality of Jequié, Bahia, Brazil. The instrument consisted of sociodemographic and health data, the Barthel Index and the Lawton scale. (Results) In all, 78.00% of the elderly were classified as dependent in the basic activities and 65.33% in the instrumental activities of daily living. Using the Kruskal- Wallis test, we found a statistically significant difference between the means of instrumental activities and the age groups (p= 0.011). (Conclusion) An elevated number of elderly were classified as dependent in terms of functional capacity and increased age is related to greater impairment in the execution of instrumental activities of daily living

Impacto do déficit cognitivo na saúde e capacidade funcional da populção idosa do distrito oeste da cidade do Natal-RN

A incapacidade funcional e o déficit cognitivo são condições relacionadas com o processo de envelhecimento humano. O déficit cognitivo é considerado um importante fator preditor de prejuízos funcionais em populações idosas. Este estudo teve um caráter multidisciplinar na sua elaboração e execução, contando com profissionais de distintas áreas da saúde. Trata-se de um estudo transversal que analisou a associação do desempenho cognitivo na capacidade funcional de uma amostra populacional idosa residente em Natal-RN. Foram avaliados 213 idosos pelo Questionário Multidimensional BOAS e pelo Mini-Exame do Estado Mental (MEEM). Através de análise multivariada, mediante regressão linear, os resultados apontam a idade e o nível de escolaridade como fatores associados ao desempenho cognitivo nos idosos do estudo. O modelo explicativo final elaborado através de regressão logística apresentou o desempenho cognitivo como a única variável associada à incapacidade funcional para as atividades da vida diária, mesmo quando ajustado por variáveis sócio-demográficas. A identificação dos fatores, associados ao declínio do estado funcional dos idosos que vivem na comunidade, poderá reduzir o impacto dessas condições na capacidade funcional através da aplicação de medidas preventivas voltadas para este fim

Análise do óxido nítrico, perfil lipídico, resistência arterial e capacidade funcional de idosas submetidas a um treinamento aquático

A prática regular de exercício físico possibilita redução dos efeitos das disfunções hormonais e envelhecimento biológico natural que promovem desajustes hemodinâmicos, vasculares e músculo-esqueléticos, principalmente na população feminina no período pós-menopausa. Nesta fase da vida, o exercício aquático representa mais do que uma forma de ajuste funcional, é uma forma de manutenção de independência para as atividades da vida diária (AVD s) e melhoria na qualidade de vida. Este estudo teve como objetivo avaliar as modificações adquiridas pela prática regular de um programa de exercício aquático concorrente de intensidade moderada no nível de óxido nítrico (ON), no índice de resistividade arterial (IR), no perfil lipídico, na capacidade funcional e na qualidade de vida de idosas. A amostra foi formada por idosas (60 a 80 anos) selecionadas por randomização, por sorteio simples divididas em grupo controle e grupo de intervenção as quais foram submetidas a um programa de exercícios aquáticos proposto inicialmente em projeto piloto, por 12 semanas (n=34) e o ensaio clínico teve duração de 16 semanas (n=40). Foi coletada amostra sanguínea das idosas e avaliado o perfil lipídico pelo método enzimático com kit Labtest e o ON por medida indireta a partir da concentração de nitrito no sobrenadante das células em cultura em leitor de ELISA. A avaliação das artérias carótidas e vertebrais foi feita utilizando o método de ultra-som Doopler. A capacidade funcional foi avaliada por quatro testes que simulam atividades da vida diária que são: caminhar 10 metros (C10m), levantar da posição sentada (LPS), levantar da cadeira e locomover-se pela casa (LCLC) e o de levantar-se da posição decúbito ventral (LPDV). No estudo piloto foi avaliado o índice geral de autonomia funcional (IG) e a qualidade de vida através do questionário WHOQOL-100. Empregou-se a análise de variância (ANOVA) com medidas repetidas nos fatores grupo (GH e GC) e tempo (pré e pós-teste) para as comparações intra e intergrupos nas variáveis seguida do post hoc de Scheffé. Utilizou-se o teste de correlação de Pearson e adotado o valor de p<0,05 para a significância estatística. Houve aumento do ON circulante, redução nos índices de resistividade arterial, melhoria significativa nos níveis de colesterol e triglicérides e ganho na capacidade funcional das idosas após a intervenção. Existiu correlação inversa entre a resistividade da artéria vertebral direita (VERTD) e níveis de ON e entre a VERTD e os níveis plasmáticos de HDL, assim como entre estes e o teste C10m nas idosas em estudo. Não houve modificações significativas na qualidade de vida das idosas. Em conclusão, o programa de exercício proposto foi capaz de oferecer melhorias funcionais, aumentar o nível de óxido nítrico circulante, diminuindo a resistência arterial promovendo modificações no perfil lipídico de idosas

Limiar e tolerância à dor em mulheres com fibromialgia: correlação com a capacidade funcional, qualidade do sono e força muscular

O estudo tem como objetivo principal comparar o limiar e a tolerância à dor em mulheres com fibromialgia e mulheres sem sintomas de dor, correlacionando com capacidade funcional, qualidade do sono e força de preensão manual. Trata-se de um estudo descritivo de corte transversal realizado na Universidade Potiguar e na Universidade Federal do Rio Grande do Norte, envolvendo uma equipe multidisciplinar com fisioterapeutas, médicos e educadores físicos. Setenta e duas mulheres foram divididas em dois grupos: grupo com fibromialgia (GF, n = 40) e grupo controle (GC, n = 32) sem sintomas de dor, pareadas por idade e índice de massa corporal. Questionários foram usados para medir capacidade funcional (Fibromyalgia Impact Questionnaire FIQ) e qualidade do sono (Pittsburgh Sleep Quality Index - PSQI). Limiar e tolerância à dor foram medidas por um algômetro de pressão e para determinar força foi usado um dinamômetro de preensão manual. Na análise estatística o teste T de Student foi usado para comparar grupos e o coeficiente de correlação de Spearman foi usado no GF para correlacionar limiar e tolerância à dor com as demais medidas. O GF mostrou piores resultados nas medianas comparado ao GC: limiar de dor 2 kg/cm2 versus 5,5 kg/cm2 (p <0,001), tolerância à dor 2,8 kg/cm2 versus 7,2 kg/cm2 (p <0,001) e força de preensão manual 14,5 kgf versus 25,2 kgf (p <0,001). Adicionalmente os dois questionários apresentaram resultados piores também no GF em relação ao GC, respectivamente: FIQ 65 versus 14,3 (p <0,001) e PSQI 13 versus 5 (p <0,001). No GF foi observada correlação negativa tanto entre o limiar de dor quanto a tolerância à dor com os escores do FIQ e do PSQI. No GF foram observadas correlações positivas tanto entre o limiar de dor (p <0,01) quanto a tolerância à dor (p <0,01) com força de preensão manual. É possível concluir que limiar de dor e tolerância à dor estão correlacionados com capacidade funcional, qualidade do sono e força de preensão manual em mulheres com fibromialgia