Estimação da porcentagem de flúor em alumina fluoretada proveniente de uma planta de tratamento de gases por meio de um sensor virtual neural

As indústrias têm buscado constantemente reduzir gastos operacionais, visando o aumento do lucro e da competitividade. Para alcançar essa meta, são necessários, dentre outros fatores, o projeto e a implantação de novas ferramentas que permitam o acesso às informações relevantes do processo de forma precisa, eficiente e barata. Os sensores virtuais têm sido aplicados cada vez mais nas indústrias. Por ser flexível, ele pode ser adaptado a qualquer tipo de medição, promovendo uma redução de custos operacionais sem comprometer, e em alguns casos até melhorar, a qualidade da informação gerada. Como estão totalmente baseados em software, não estão sujeitos a danos físicos como os sensores reais, além de permitirem uma melhor adaptação a ambientes hostis e de difícil acesso. A razão do sucesso destes tipos de sensores é a utilização de técnicas de inteligência computacional, as quais têm sido usadas na modelagem de vários processos não lineares altamente complexos. Este trabalho tem como objetivo estimar a qualidade da alumina fluoretada proveniente de uma Planta de Tratamento de Gases (PTG), a qual é resultado da adsorção de gases poluentes em alumina virgem, via sensor virtual. O modelo que emula o comportamento de um sensor de qualidade de alumina foi criado através da técnica de inteligência computacional conhecida como Rede Neural Artificial. As motivações deste trabalho consistem em: realizar simulações virtuais, sem comprometer o funcionamento da PTG; tomar decisões mais precisas e não baseada somente na experiência do operador; diagnosticar potenciais problemas, antes que esses interfiram na qualidade da alumina fluoretada; manter o funcionamento do forno de redução de alumínio dentro da normalidade, pois a produção de alumina de baixa qualidade afeta a reação de quebra da molécula que contém este metal. Os benefícios que este projeto trará consistem em: aumentar a eficiência da PTG, produzindo alumina fluoretada de alta qualidade e emitindo menos gases poluentes na atmosfera, além de aumentar o tempo de vida útil do forno de redução.

Correlação de poços com múltiplos perfis através da rede neural multicamadas

A correlação estratigráfica busca a determinação da continuidade lateral das rochas, ou a equivalência espacial entre unidades litológicas em subsuperfície, a partir de informações geológico-geofísicas oriundas de poços tubulares, que atravessam estas rochas. Normalmente, mas não exclusivamente, a correlação estratigráfica é realizada a partir das propriedades físicas registradas nos perfis geofísicos de poço. Neste caso, busca-se a equivalência litológica a partir da equivalência entre as propriedades físicas, medidas nos vários poços de um campo petrolífero. A técnica da correlação estratigráfica com perfis geofísicos de poço não é uma atividade trivial e sim, sujeita a inúmeras possibilidades de uma errônea interpretação da disposição geométrica ou da continuidade lateral das rochas em subsuperfície, em função da variabilidade geológica e da ambigüidade das respostas das ferramentas. Logo, é recomendável a utilização de um grande número de perfis de um mesmo poço, para uma melhor interpretação. A correlação estratigráfica é fundamental para o engenheiro de reservatório ou o geólogo, pois a partir da mesma, é possível a definição de estratégias de explotação de um campo petrolífero e a interpretação das continuidades hidráulicas dos reservatórios, bem como auxílio para a construção do modelo geológico para os reservatórios, a partir da interpretação do comportamento estrutural das diversas camadas em subsuperfície. Este trabalho apresenta um método de automação das atividades manuais envolvidas na correlação estratigráfica, com a utilização de vários perfis geofísicos de poço, através de uma arquitetura de rede neural artificial multicamadas, treinada com o algoritmo de retropropagação do erro. A correlação estratigráfica, obtida a partir da rede neural artificial, possibilita o transporte da informação geológica do datum de correlação ao longo do campo, possibilitando ao intérprete, uma visão espacial do comportamento do reservatório e a simulação dos possíveis paleoambientes. Com a metodologia aqui apresentada foi possível a construção automática de um bloco diagrama, mostrando a disposição espacial de uma camada argilosa, utilizando-se os perfis de Raio Gama (RG), Volume de Argila (Vsh), Densidade (ρ_b) e de Porosidade Neutrônica (φ_n) selecionados em cinco poços da região do Lago Maracaibo, na Venezuela.

Detecção de refletores sísmicos por rede neural discreta

As redes neurais artificiais têm provado serem uma poderosa técnica na resolução de uma grande variedade de problemas de otimização. Nesta dissertação é desenvolvida uma nova rede neural, tipo recorrente, sem realimentação (self-feedback loops) e sem neurônios ocultos, para o processamento do sinal sísmico, para fornecer a posição temporal, a polaridade e as amplitudes estimadas dos refletores sísmicos, representadas pelos seus coeficientes de reflexão. A principal característica dessa nova rede neural consiste no tipo de função de ativação utilizada, a qual permite três possíveis estados para o neurônio. Busca-se estimar a posição dos refletores sísmicos e reproduzir as verdadeiras polaridades desses refletores. A idéia básica desse novo tipo de rede, aqui denominada rede neural discreta (RND), é relacionar uma função objeto, que descreve o problema geofísico, com a função de Liapunov, que descreve a dinâmica da rede neural. Deste modo, a dinâmica da rede leva a uma minimização local da sua função de Liapunov e consequentemente leva a uma minimização da função objeto. Assim, com uma codificação conveniente do sinal de saída da rede tem-se uma solução do problema geofísico. A avaliação operacional da arquitetura desta rede neural artificial é realizada em dados sintéticos gerados através do modelo convolucional simples e da teoria do raio. A razão é para explicar o comportamento da rede com dados contaminados por ruído, e diante de pulsos fonte de fases mínima, máxima e misturada.

Tool Condition Monitoring of Single-Point Dresser Using Acoustic Emission and Neural Networks Models

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

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.

The role of clouds in improving the regression model for hourly values of diffuse solar radiation

The study introduces a new regression model developed to estimate the hourly values of diffuse solar radiation at the surface. The model is based on the clearness index and diffuse fraction relationship, and includes the effects of cloud (cloudiness and cloud type), traditional meteorological variables (air temperature, relative humidity and atmospheric pressure observed at the surface) and air pollution (concentration of particulate matter observed at the surface). The new model is capable of predicting hourly values of diffuse solar radiation better than the previously developed ones (R-2 = 0.93 and RMSE = 0.085). A simple version with a large applicability is proposed that takes into consideration cloud effects only (cloudiness and cloud height) and shows a R-2 = 0.92. (C) 2011 Elsevier Ltd. All rights reserved.

Application of discriminant analysis-based model for prediction of risk of low back disorders due to workplace design in industrial jobs

The occupational exposure limits of different risk factors for development of low back disorders (LBDs) have not yet been established. One of the main problems in setting such guidelines is the limited understanding of how different risk factors for LBDs interact in causing injury, since the nature and mechanism of these disorders are relatively unknown phenomena. Industrial ergonomists' role becomes further complicated because the potential risk factors that may contribute towards the onset of LBDs interact in a complex manner, which makes it difficult to discriminate in detail among the jobs that place workers at high or low risk of LBDs. The purpose of this paper was to develop a comparative study between predictions based on the neural network-based model proposed by Zurada, Karwowski & Marras (1997) and a linear discriminant analysis model, for making predictions about industrial jobs according to their potential risk of low back disorders due to workplace design. The results obtained through applying the discriminant analysis-based model proved that it is as effective as the neural network-based model. Moreover, the discriminant analysis-based model proved to be more advantageous regarding cost and time savings for future data gathering.

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.

A New Prediction Model for Slope Stability Analysis

The instability of river bank can result in considerable human and land losses. The Po river is the most important in Italy, characterized by main banks of significant and constantly increasing height. This study presents multilayer perceptron of artificial neural network (ANN) to construct prediction models for the stability analysis of river banks along the Po River, under various river and groundwater boundary conditions. For this aim, a number of networks of threshold logic unit are tested using different combinations of the input parameters. Factor of safety (FS), as an index of slope stability, is formulated in terms of several influencing geometrical and geotechnical parameters. In order to obtain a comprehensive geotechnical database, several cone penetration tests from the study site have been interpreted. The proposed models are developed upon stability analyses using finite element code over different representative sections of river embankments. For the validity verification, the ANN models are employed to predict the FS values of a part of the database beyond the calibration data domain. The results indicate that the proposed ANN models are effective tools for evaluating the slope stability. The ANN models notably outperform the derived multiple linear regression models.

An insulin infusion advisory system based on autotuning nonlinear model-predictive control

This paper aims at the development and evaluation of a personalized insulin infusion advisory system (IIAS), able to provide real-time estimations of the appropriate insulin infusion rate for type 1 diabetes mellitus (T1DM) patients using continuous glucose monitors and insulin pumps. The system is based on a nonlinear model-predictive controller (NMPC) that uses a personalized glucose-insulin metabolism model, consisting of two compartmental models and a recurrent neural network. The model takes as input patient's information regarding meal intake, glucose measurements, and insulin infusion rates, and provides glucose predictions. The predictions are fed to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. An algorithm based on fuzzy logic has been developed for the on-line adaptation of the NMPC control parameters. The IIAS has been in silico evaluated using an appropriate simulation environment (UVa T1DM simulator). The IIAS was able to handle various meal profiles, fasting conditions, interpatient variability, intraday variation in physiological parameters, and errors in meal amount estimations.

Implementation of a phenomenological evaporation model into a porous network simulation for water management in low temperature fuel cells

A phenomenological transition film evaporation model was introduced to a pore network model with the consideration of pore radius, contact angle, non-isothermal interface temperature, microscale fluid flows and heat and mass transfers. This was achieved by modeling the transition film region of the menisci in each pore throughout the porous transport layer of a half-cell polymer electrolyte membrane (PEM) fuel cell. The model presented in this research is compared with the standard diffusive fuel cell modeling approach to evaporation and shown to surpass the conventional modeling approach in terms of predicting the evaporation rates in porous media. The current diffusive evaporation models used in many fuel cell transport models assumes a constant evaporation rate across the entire liquid-air interface. The transition film model was implemented into the pore network model to address this issue and create a pore size dependency on the evaporation rates. This is accomplished by evaluating the transition film evaporation rates determined by the kinetic model for every pore containing liquid water in the porous transport layer (PTL). The comparison of a transition film and diffusive evaporation model shows an increase in predicted evaporation rates for smaller pore sizes with the transition film model. This is an important parameter when considering the micro-scaled pore sizes seen in the PTL and becomes even more substantial when considering transport in fuel cells containing an MPL, or a large variance in pore size. Experimentation was performed to validate the transition film model by monitoring evaporation rates from a non-zero contact angle water droplet on a heated substrate. The substrate was a glass plate with a hydrophobic coating to reduce wettability. The tests were performed at a constant substrate temperature and relative humidity. The transition film model was able to accurately predict the drop volume as time elapsed. By implementing the transition film model to a pore network model the evaporation rates present in the PTL can be more accurately modeled. This improves the ability of a pore network model to predict the distribution of liquid water and ultimately the level of flooding exhibited in a PTL for various operating conditions.

An insulin infusion advisory system for type 1 diabetes patients based on non-linear model predictive control methods

In this paper, an Insulin Infusion Advisory System (IIAS) for Type 1 diabetes patients, which use insulin pumps for the Continuous Subcutaneous Insulin Infusion (CSII) is presented. The purpose of the system is to estimate the appropriate insulin infusion rates. The system is based on a Non-Linear Model Predictive Controller (NMPC) which uses a hybrid model. The model comprises a Compartmental Model (CM), which simulates the absorption of the glucose to the blood due to meal intakes, and a Neural Network (NN), which simulates the glucose-insulin kinetics. The NN is a Recurrent NN (RNN) trained with the Real Time Recurrent Learning (RTRL) algorithm. The output of the model consists of short term glucose predictions and provides input to the NMPC, in order for the latter to estimate the optimum insulin infusion rates. For the development and the evaluation of the IIAS, data generated from a Mathematical Model (MM) of a Type 1 diabetes patient have been used. The proposed control strategy is evaluated at multiple meal disturbances, various noise levels and additional time delays. The results indicate that the implemented IIAS is capable of handling multiple meals, which correspond to realistic meal profiles, large noise levels and time delays.

Multi-Model Data Fusion to Improve an Early Warning System for Hypo-/Hyperglycemic Events

Correct predictions of future blood glucose levels in individuals with Type 1 Diabetes (T1D) can be used to provide early warning of upcoming hypo-/hyperglycemic events and thus to improve the patient's safety. To increase prediction accuracy and efficiency, various approaches have been proposed which combine multiple predictors to produce superior results compared to single predictors. Three methods for model fusion are presented and comparatively assessed. Data from 23 T1D subjects under sensor-augmented pump (SAP) therapy were used in two adaptive data-driven models (an autoregressive model with output correction - cARX, and a recurrent neural network - RNN). Data fusion techniques based on i) Dempster-Shafer Evidential Theory (DST), ii) Genetic Algorithms (GA), and iii) Genetic Programming (GP) were used to merge the complimentary performances of the prediction models. The fused output is used in a warning algorithm to issue alarms of upcoming hypo-/hyperglycemic events. The fusion schemes showed improved performance with lower root mean square errors, lower time lags, and higher correlation. In the warning algorithm, median daily false alarms (DFA) of 0.25%, and 100% correct alarms (CA) were obtained for both event types. The detection times (DT) before occurrence of events were 13.0 and 12.1 min respectively for hypo-/hyperglycemic events. Compared to the cARX and RNN models, and a linear fusion of the two, the proposed fusion schemes represents a significant improvement.

The Optimal combination: Grammatical Swarm, Particle Swarm Optimization and Neural Networks.

Social behaviour is mainly based on swarm colonies, in which each individual shares its knowledge about the environment with other individuals to get optimal solutions. Such co-operative model differs from competitive models in the way that individuals die and are born by combining information of alive ones. This paper presents the particle swarm optimization with differential evolution algorithm in order to train a neural network instead the classic back propagation algorithm. The performance of a neural network for particular problems is critically dependant on the choice of the processing elements, the net architecture and the learning algorithm. This work is focused in the development of methods for the evolutionary design of artificial neural networks. This paper focuses in optimizing the topology and structure of connectivity for these networks.

Polynomial approximation using particle swarm optimization of lineal enhanced neural networks with no hidden layers.

This paper presents some ideas about a new neural network architecture that can be compared to a Taylor analysis when dealing with patterns. Such architecture is based on lineal activation functions with an axo-axonic architecture. A biological axo-axonic connection between two neurons is defined as the weight in a connection in given by the output of another third neuron. This idea can be implemented in the so called Enhanced Neural Networks in which two Multilayer Perceptrons are used; the first one will output the weights that the second MLP uses to computed the desired output. This kind of neural network has universal approximation properties even with lineal activation functions. There exists a clear difference between cooperative and competitive strategies. The former ones are based on the swarm colonies, in which all individuals share its knowledge about the goal in order to pass such information to other individuals to get optimum solution. The latter ones are based on genetic models, that is, individuals can die and new individuals are created combining information of alive one; or are based on molecular/celular behaviour passing information from one structure to another. A swarm-based model is applied to obtain the Neural Network, training the net with a Particle Swarm algorithm.