Interpretação do gráfico de Hingle através de rede competitiva angular

A saturação de água é a principal propriedade petrofísica para a avaliação de reservatórios de hidrocarbonetos, pois através da análise dos seus valores é definida a destinação final do poço recém perfurado, como produtor ou poço seco. O cálculo da saturação de água para as formações limpas é, comumente, realizado a partir da equação de Archie, que envolve a determinação da resistividade da zona virgem, obtida a partir de um perfil de resistividade profunda e o cálculo de porosidade da rocha, obtida a partir dos perfis de porosidade. A equação de Archie envolve ainda, a determinação da resistividade da água de formação, que normalmente necessita de definição local e correção para a profundidade da formação e da adoção de valores convenientes para os coeficientes de Archie. Um dos métodos mais tradicionais da geofísica de poço para o cálculo da saturação de água é o método de Hingle, particularmente útil nas situações de desconhecimento da resistividade da água de formação. O método de Hingle estabelece uma forma linear para a equação de Archie, a partir dos perfis de resistividade e porosidade e a representa na forma gráfica, como a reta da água ou dos pontos, no gráfico de Hingle, com saturação de água unitária e o valor da resistividade da água de formação é obtido a partir da inclinação da reta da água. Independente do desenvolvimento tecnológico das ferramentas de perfilagem e dos computadores digitais, o geofísico, ainda hoje, se vê obrigado a realizar a interpretação de ábacos ou gráficos, sujeito a ocorrência de erros derivados da sua acuidade visual. Com o objetivo de mitigar a ocorrência deste tipo de erro e produzir uma primeira aproximação para a saturação de água em tempo real de perfilagem do poço, insere-se o trabalho apresentado nesta dissertação, com a utilização de uma conveniente arquitetura de rede neural artificial, a rede competitiva angular, capaz de identificar a localização da reta da água, a partir da identificação de padrões angulares presentes nos dados dos perfis de porosidade e resistividade representados no gráfico de Hingle. A avaliação desta metodologia é realizada sobre dados sintéticos, que satisfazem integralmente a equação de Archie, e sobre dados reais.

Método para determinação da viabilidade e qualidade de embriões

The invention described is a method for determining embryo viability and quality that makes a quick evaluation possible, with minimal interference in the development of the embryo, using a microscopy system associated with digital image capture, providing a set of values for each embryo which represents to what extent the embryo can be considered to belong to each of the four possible grades, using artificial neural network technology, with objectivity and reproducibility.

Previsão de Preço (LMP) por Redes GRNN

Pós-graduação em Engenharia Elétrica - FEIS

Detecção e localização de danos em materiais compósitos aplicado em aeronaves utilizando redes neurais artificiais

Pós-graduação em Engenharia Mecânica - FEIS

Previsão de demanda de cargas elétricas por seleção de variáveis stepwise e redes neurais artificiais

Pós-graduação em Engenharia Elétrica - FEIS

Comparing support vector machines and artificial neural networks in the recognition of steering angle for driving of mobile robots through paths in plantations

The use of mobile robots turns out to be interesting in activities where the action of human specialist is difficult or dangerous. Mobile robots are often used for the exploration in areas of difficult access, such as rescue operations and space missions, to avoid human experts exposition to risky situations. Mobile robots are also used in agriculture for planting tasks as well as for keeping the application of pesticides within minimal amounts to mitigate environmental pollution. In this paper we present the development of a system to control the navigation of an autonomous mobile robot through tracks in plantations. Track images are used to control robot direction by pre-processing them to extract image features. Such features are then submitted to a support vector machine and an artificial neural network in order to find out the most appropriate route. A comparison of the two approaches was performed to ascertain the one presenting the best outcome. The overall goal of the project to which this work is connected is to develop a real time robot control system to be embedded into a hardware platform. In this paper we report the software implementation of a support vector machine and of an artificial neural network, which so far presented respectively around 93% and 90% accuracy in predicting the appropriate route. (C) 2013 The Authors. Published by Elsevier B.V. Selection and peer review under responsibility of the organizers of the 2013 International Conference on Computational Science

Late-onset obsessive-compulsive disorder: Risk factors and correlates

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Monitoramento da condição da ferramenta de dressagem usando sinais de vibração e modelos neurais

Pós-graduação em Engenharia Elétrica - FEB

Uso de redes neurais com adaptação de pesos por modos deslizantes para controle de sistemas e aplicações em máquinas elétricas

Pós-graduação em Engenharia Elétrica - FEIS

Seleção de materiais e design: um método com base nas redes neurais artificiais

Pós-graduação em Design - FAAC

Use of uniform designs in combination with neural networks for viral infection process development

This work aimed to compare the predictive capacity of empirical models, based on the uniform design utilization combined to artificial neural networks with respect to classical factorial designs in bioprocess, using as example the rabies virus replication in BHK-21 cells. The viral infection process parameters under study were temperature (34°C, 37°C), multiplicity of infection (0.04, 0.07, 0.1), times of infection, and harvest (24, 48, 72 hours) and the monitored output parameter was viral production. A multilevel factorial experimental design was performed for the study of this system. Fractions of this experimental approach (18, 24, 30, 36 and 42 runs), defined according uniform designs, were used as alternative for modelling through artificial neural network and thereafter an output variable optimization was carried out by means of genetic algorithm methodology. Model prediction capacities for all uniform design approaches under study were better than that found for classical factorial design approach. It was demonstrated that uniform design in combination with artificial neural network could be an efficient experimental approach for modelling complex bioprocess like viral production. For the present study case, 67% of experimental resources were saved when compared to a classical factorial design approach. In the near future, this strategy could replace the established factorial designs used in the bioprocess development activities performed within biopharmaceutical organizations because of the improvements gained in the economics of experimentation that do not sacrifice the quality of decisions.

Network-based high level data classification

Traditional supervised data classification considers only physical features (e. g., distance or similarity) of the input data. Here, this type of learning is called low level classification. On the other hand, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is, here, referred to as high level classification. In this paper, we propose a hybrid classification technique that combines both types of learning. The low level term can be implemented by any classification technique, while the high level term is realized by the extraction of features of the underlying network constructed from the input data. Thus, the former classifies the test instances by their physical features or class topologies, while the latter measures the compliance of the test instances to the pattern formation of the data. Our study shows that the proposed technique not only can realize classification according to the pattern formation, but also is able to improve the performance of traditional classification techniques. Furthermore, as the class configuration's complexity increases, such as the mixture among different classes, a larger portion of the high level term is required to get correct classification. This feature confirms that the high level classification has a special importance in complex situations of classification. Finally, we show how the proposed technique can be employed in a real-world application, where it is capable of identifying variations and distortions of handwritten digit images. As a result, it supplies an improvement in the overall pattern recognition rate.

Stochastic competitive learning in complex networks

Competitive learning is an important machine learning approach which is widely employed in artificial neural networks. In this paper, we present a rigorous definition of a new type of competitive learning scheme realized on large-scale networks. The model consists of several particles walking within the network and competing with each other to occupy as many nodes as possible, while attempting to reject intruder particles. The particle's walking rule is composed of a stochastic combination of random and preferential movements. The model has been applied to solve community detection and data clustering problems. Computer simulations reveal that the proposed technique presents high precision of community and cluster detections, as well as low computational complexity. Moreover, we have developed an efficient method for estimating the most likely number of clusters by using an evaluator index that monitors the information generated by the competition process itself. We hope this paper will provide an alternative way to the study of competitive learning.

Network-based stochastic semisupervised learning

Semisupervised learning is a machine learning approach that is able to employ both labeled and unlabeled samples in the training process. In this paper, we propose a semisupervised data classification model based on a combined random-preferential walk of particles in a network (graph) constructed from the input dataset. The particles of the same class cooperate among themselves, while the particles of different classes compete with each other to propagate class labels to the whole network. A rigorous model definition is provided via a nonlinear stochastic dynamical system and a mathematical analysis of its behavior is carried out. A numerical validation presented in this paper confirms the theoretical predictions. An interesting feature brought by the competitive-cooperative mechanism is that the proposed model can achieve good classification rates while exhibiting low computational complexity order in comparison to other network-based semisupervised algorithms. Computer simulations conducted on synthetic and real-world datasets reveal the effectiveness of the model.

THE SLOAN DIGITAL SKY SURVEY CO-ADD: A GALAXY PHOTOMETRIC REDSHIFT CATALOG

We present and describe a catalog of galaxy photometric redshifts (photo-z) for the Sloan Digital Sky Survey (SDSS) Co-add Data. We use the artificial neural network (ANN) technique to calculate the photo-z and the nearest neighbor error method to estimate photo-z errors for similar to 13 million objects classified as galaxies in the co-add with r < 24.5. The photo-z and photo-z error estimators are trained and validated on a sample of similar to 83,000 galaxies that have SDSS photometry and spectroscopic redshifts measured by the SDSS Data Release 7 (DR7), the Canadian Network for Observational Cosmology Field Galaxy Survey, the Deep Extragalactic Evolutionary Probe Data Release 3, the VIsible imaging Multi-Object Spectrograph-Very Large Telescope Deep Survey, and the WiggleZ Dark Energy Survey. For the best ANN methods we have tried, we find that 68% of the galaxies in the validation set have a photo-z error smaller than sigma(68) = 0.031. After presenting our results and quality tests, we provide a short guide for users accessing the public data.