Detecção e diagnostico de agarramento em válvulas posicionadoras

Valve stiction, or static friction, in control loops is a common problem in modern industrial processes. Recently, many studies have been developed to understand, reproduce and detect such problem, but quantification still remains a challenge. Since the valve position (mv) is normally unknown in an industrial process, the main challenge is to diagnose stiction knowing only the output signals of the process (pv) and the control signal (op). This paper presents an Artificial Neural Network approach in order to detect and quantify the amount of static friction using only the pv and op information. Different methods for preprocessing the training set of the neural network are presented. Those methods are based on the calculation of centroid and Fourier Transform. The proposal is validated using a simulated process and the results show a satisfactory measurement of stiction.

Influência de mudanças de fase no ciclo claro-escuro sobre o controle autonômico cardíaco de ratos

Introduction: The circadian system has neural projections for the Autonomic Nervous System (ANS), directly interfering with sympathetic-vagal modulation of the cardiovascular system. Disturbances in the circadian system, such as phase changes in light-dark cycle (LD), has been related to the risk of development of cardiovascular diseases due to increased sympathetic tone and reduction o Heart Rate Variability (HRV - RR intervals). Purpose: Investigate the interaction between Circadian Timing System and cardiac autonomic control in rats. Materials and methods: We used 18 Wistar rats (♀, age = 139.9 ± 32.1 days, weight = 219.5 ± 16.2 g), divided into three distinct groups: Control (CG), phase delay of 6h (GDe) and phase advance of 6h (GAd). Three animals were excluded during data collection (CG/GDe/GAd - n=5). Telemeters were surgically implanted in each animal for continuous acquisition of electrocardiographic (ECG) signals (duration of 21 days in the CG and 28 days in GDe/ GAd). A LD cycle was established 12h: 12h, beginning of light at18:00h and dark at 06:00h. The animals remained in the same CG LD cycle throughout the experimental period, while, on the 14th day of registration, the GDe and GAd underwent a delay and an advance in 6h, respectively. Throughout the experimental period, the locomotor activity (LA), the mean heart rate (mHR) and variables related to iRR [mean RR (mRR), SDNN, RMSSD, LF, HF and LF/ HF ratio ] were recorded. All data were analyzed in blocks of 3 and 7 days, for the presence of circadian rhythm, values of Cosinor - mesor, amplitude and acrophase (paired t test), phase relationship, differences between light and dark (t test independent), averages every 30 minutes along each time series (two-way ANOVA with post hoc Bonferroni). The data block B1,M1 and M2 in CG served as benchmarks for comparisons between series of analysis of the GAT/GAV. Results: We observed circadian rhythmicity in the variables LA, mRR and mFC(p<0.01). mRR and mFC showed phase relationship with the LA in all three groups, being less stable in GAd. In the CG, no significant differences between blocks were found in any of the analyzes(p>0.05). Among the 7 day blocks, there was a significant reduction in mRR(p=0.04) and mFC(p=0.03) in GDe and significant reduction in HF mean(p=0.02) in GAd; and between 3 day blocks, a significant increase of LF/HF(p= 0.04) in the GDe; besides mRR(p=0.03), SDNN(p=0.04), RMSSD (p=0.04), LF (p=0.01) and HF(p=0.02) significant increase in the GAd. It was found that the differences between the means of the mRR, LA and mFC in light and dark phases were not significant after phase changes in some of the blocks/moments (GDe and GAd). No significant results were found when comparing rhythmic variables means every 30 minutes over the blocks, except for a significant decrease in mRR at the middle of the dark phase (B2) and the start of light phase (B3) - (p<0.01). Conclusion: phase advances and delays (6h) altered cardiac autonomic control in the experimental groups by temporarily HRV decrease. Phase advances apparently had greater negative interference in this process, in relation to the phase delays.

A neuromodulação do córtex pré-frontal dorsolateral na percepção de tempo em contexto neutro ou emocionalmente ativo

The time perception is critical for environmental adaptation in humans and other species. The temporal processing, has evolved through different neural systems, each responsible for processing different time scales. Among the most studied scales is that spans the arrangement of seconds to minutes. Evidence suggests that the dorsolateral prefrontal (DLPFC) cortex has relationship with the time perception scale of seconds. However, it is unclear whether the deficit of time perception in patients with brain injuries or even "reversible lesions" caused by transcranial magnetic stimulation (TMS) in this region, whether by disruption of other cognitive processes (such as attention and working memory) or the time perception itself. Studies also link the region of DLPFC in emotional regulation and specifically the judgment and emotional anticipation. Given this, our objective was to study the role of the dorsolateral prefrontal cortex in the time perception intervals of active and emotionally neutral stimuli, from the effects of cortical modulation by transcranial direct current stimulation (tDCS), through the cortical excitation (anodic current), inhibition (cathode current) and control (sham) using the ranges of 4 and 8 seconds. Our results showed that there is an underestimation when the picture was presented by 8 seconds, with the anodic current in the right DLPFC, there is an underestimation and with cathodic current in the left DLPFC, there is an overestimation of the time reproduction with neutral ones. The cathodic current over the left DLPFC leads to an inverse effect of neutral ones, an underestimation of time with negative pictures. Positive or negative pictures improved estimates for 8 second and positive pictures inhibited the effect of tDCS in DLPFC in estimating time to 4 seconds. With this work, we conclude that the DLPFC plays a key role in the o time perception and largely corresponds to the stages of memory and decision on the internal clock model. The left hemisphere participates in the perception of time in both active and emotionally neutral contexts, and we can conclude that the ETCC and an effective method to study the cortical functions in the time perception in terms of cause and effect.

Ferramenta para avaliação e inferência de parâmetros de redes industriais sem fio

Wireless sensor networks (WSN) have gained ground in the industrial environment, due to the possibility of connecting points of information that were inaccessible to wired networks. However, there are several challenges in the implementation and acceptance of this technology in the industrial environment, one of them the guaranteed availability of information, which can be influenced by various parameters, such as path stability and power consumption of the field device. As such, in this work was developed a tool to evaluate and infer parameters of wireless industrial networks based on the WirelessHART and ISA 100.11a protocols. The tool allows quantitative evaluation, qualitative evaluation and evaluation by inference during a given time of the operating network. The quantitative and qualitative evaluation are based on own definitions of parameters, such as the parameter of stability, or based on descriptive statistics, such as mean, standard deviation and box plots. In the evaluation by inference uses the intelligent technique artificial neural networks to infer some network parameters such as battery life. Finally, it displays the results of use the tool in different scenarios networks, as topologies star and mesh, in order to attest to the importance of tool in evaluation of the behavior of these networks, but also support possible changes or maintenance of the system.

Critério de correntropia no treinamento de redes fuzzy wavelet neural networks para identificação de sistemas dinâmicos não lineares

The great interest in nonlinear system identification is mainly due to the fact that a large amount of real systems are complex and need to have their nonlinearities considered so that their models can be successfully used in applications of control, prediction, inference, among others. This work evaluates the application of Fuzzy Wavelet Neural Networks (FWNN) to identify nonlinear dynamical systems subjected to noise and outliers. Generally, these elements cause negative effects on the identification procedure, resulting in erroneous interpretations regarding the dynamical behavior of the system. The FWNN combines in a single structure the ability to deal with uncertainties of fuzzy logic, the multiresolution characteristics of wavelet theory and learning and generalization abilities of the artificial neural networks. Usually, the learning procedure of these neural networks is realized by a gradient based method, which uses the mean squared error as its cost function. This work proposes the replacement of this traditional function by an Information Theoretic Learning similarity measure, called correntropy. With the use of this similarity measure, higher order statistics can be considered during the FWNN training process. For this reason, this measure is more suitable for non-Gaussian error distributions and makes the training less sensitive to the presence of outliers. In order to evaluate this replacement, FWNN models are obtained in two identification case studies: a real nonlinear system, consisting of a multisection tank, and a simulated system based on a model of the human knee joint. The results demonstrate that the application of correntropy as the error backpropagation algorithm cost function makes the identification procedure using FWNN models more robust to outliers. However, this is only achieved if the gaussian kernel width of correntropy is properly adjusted.

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.

Modelagem da degradação de fenol em efluentes aquosos através do processo de foto-fenton

Advanced Oxidation Processes (AOP) are techniques involving the formation of hydroxyl radical (HO•) with high organic matter oxidation rate. These processes application in industry have been increasing due to their capacity of degrading recalcitrant substances that cannot be completely removed by traditional processes of effluent treatment. In the present work, phenol degrading by photo-Fenton process based on addition of H2O2, Fe2+ and luminous radiation was studied. An experimental design was developed to analyze the effect of phenol, H2O2 and Fe2+ concentration on the fraction of total organic carbon (TOC) degraded. The experiments were performed in a batch photochemical parabolic reactor with 1.5 L of capacity. Samples of the reactional medium were collected at different reaction times and analyzed in a TOC measurement instrument from Shimadzu (TOC-VWP). The results showed a negative effect of phenol concentration and a positive effect of the two other variables in the TOC degraded fraction. A statistical analysis of the experimental design showed that the hydrogen peroxide concentration was the most influent variable in the TOC degraded fraction at 45 minutes and generated a model with R² = 0.82, which predicted the experimental data with low precision. The Visual Basic for Application (VBA) tool was used to generate a neural networks model and a photochemical database. The aforementioned model presented R² = 0.96 and precisely predicted the response data used for testing. The results found indicate the possible application of the developed tool for industry, mainly for its simplicity, low cost and easy access to the program.

Modelagem da degradação de fenol em efluentes aquosos através do processo de foto-fenton

Advanced Oxidation Processes (AOP) are techniques involving the formation of hydroxyl radical (HO•) with high organic matter oxidation rate. These processes application in industry have been increasing due to their capacity of degrading recalcitrant substances that cannot be completely removed by traditional processes of effluent treatment. In the present work, phenol degrading by photo-Fenton process based on addition of H2O2, Fe2+ and luminous radiation was studied. An experimental design was developed to analyze the effect of phenol, H2O2 and Fe2+ concentration on the fraction of total organic carbon (TOC) degraded. The experiments were performed in a batch photochemical parabolic reactor with 1.5 L of capacity. Samples of the reactional medium were collected at different reaction times and analyzed in a TOC measurement instrument from Shimadzu (TOC-VWP). The results showed a negative effect of phenol concentration and a positive effect of the two other variables in the TOC degraded fraction. A statistical analysis of the experimental design showed that the hydrogen peroxide concentration was the most influent variable in the TOC degraded fraction at 45 minutes and generated a model with R² = 0.82, which predicted the experimental data with low precision. The Visual Basic for Application (VBA) tool was used to generate a neural networks model and a photochemical database. The aforementioned model presented R² = 0.96 and precisely predicted the response data used for testing. The results found indicate the possible application of the developed tool for industry, mainly for its simplicity, low cost and easy access to the program.

Implementação de classificador de tarefas mentais baseado em EEG

BARBOSA, André F. ; SOUZA, Bryan C. ; PEREIRA JUNIOR, Antônio ; MEDEIROS, Adelardo A. D.de, . Implementação de Classificador de Tarefas Mentais Baseado em EEG. In: CONGRESSO BRASILEIRO DE REDES NEURAIS, 9., 2009, Ouro Preto, MG. Anais... Ouro Preto, MG, 2009

Desenvolvimento de um sistema Inteligente para a análise de cartas dinamométricas no método de elevação por bombeio mecânico

The artificial lifting of oil is needed when the pressure of the reservoir is not high enough so that the fluid contained in it can reach the surface spontaneously. Thus the increase in energy supplies artificial or additional fluid integral to the well to come to the surface. The rod pump is the artificial lift method most used in the world and the dynamometer card (surface and down-hole) is the best tool for the analysis of a well equipped with such method. A computational method using Artificial Neural Networks MLP was and developed using pre-established patterns, based on its geometry, the downhole card are used for training the network and then the network provides the knowledge for classification of new cards, allows the fails diagnose in the system and operation conditions of the lifting system. These routines could be integrated to a supervisory system that collects the cards to be analyzed

Estudo e implementação de algoritmos inteligentes para detecção e classificação de falhas na medição de gás natural

This master dissertation presents the study and implementation of inteligent algorithms to monitor the measurement of sensors involved in natural gas custody transfer processes. To create these algoritmhs Artificial Neural Networks are investigated because they have some particular properties, such as: learning, adaptation, prediction. A neural predictor is developed to reproduce the sensor output dynamic behavior, in such a way that its output is compared to the real sensor output. A recurrent neural network is used for this purpose, because of its ability to deal with dynamic information. The real sensor output and the estimated predictor output work as the basis for the creation of possible sensor fault detection and diagnosis strategies. Two competitive neural network architectures are investigated and their capabilities are used to classify different kinds of faults. The prediction algorithm and the fault detection classification strategies, as well as the obtained results, are presented

Reconfiguração dinâmica de estratégias distribuídas em dispositivos foundation fieldbus para a otimização de processos na indústria do petróleo

The petrochemical industry has as objective obtain, from crude oil, some products with a higher commercial value and a bigger industrial utility for energy purposes. These industrial processes are complex, commonly operating with large production volume and in restricted operation conditions. The operation control in optimized and stable conditions is important to keep obtained products quality and the industrial plant safety. Currently, industrial network has been attained evidence when there is a need to make the process control in a distributed way. The Foundation Fieldbus protocol for industrial network, for its interoperability feature and its user interface organized in simple configuration blocks, has great notoriety among industrial automation network group. This present work puts together some benefits brought by industrial network technology to petrochemical industrial processes inherent complexity. For this, a dynamic reconfiguration system for intelligent strategies (artificial neural networks, for example) based on the protocol user application layer is proposed which might allow different applications use in a particular process, without operators intervention and with necessary guarantees for the proper plant functioning

Implementação de um Módulo de Supervisão para um Sistema de Detecção de Vazamentos em Dutos de Petróleo

The transport of fluids through pipes is used in the oil industry, being the pipelines an important link in the logistics flow of fluids. However, the pipelines suffer deterioration in their walls caused by several factors which may cause loss of fluids to the environment, justifying the investment in techniques and methods of leak detection to minimize fluid loss and environmental damage. This work presents the development of a supervisory module in order to inform to the operator the leakage in the pipeline monitored in the shortest time possible, in order that the operator log procedure that entails the end of the leak. This module is a component of a system designed to detect leaks in oil pipelines using sonic technology, wavelets and neural networks. The plant used in the development and testing of the module presented here was the system of tanks of LAMP, and its LAN, as monitoring network. The proposal consists of, basically, two stages. Initially, assess the performance of the communication infrastructure of the supervisory module. Later, simulate leaks so that the DSP sends information to the supervisory performs the calculation of the location of leaks and indicate to which sensor the leak is closer, and using the system of tanks of LAMP, capture the pressure in the pipeline monitored by piezoresistive sensors, this information being processed by the DSP and sent to the supervisory to be presented to the user in real time

Controle preditivo neural aplicado à processos petroquímicos

A pesquisa tem como objetivo desenvolver uma estrutura de controle preditivo neural, com o intuito de controlar um processo de pH, caracterizado por ser um sistema SISO (Single Input - Single Output). O controle de pH é um processo de grande importância na indústria petroquímica, onde se deseja manter constante o nível de acidez de um produto ou neutralizar o afluente de uma planta de tratamento de fluidos. O processo de controle de pH exige robustez do sistema de controle, pois este processo pode ter ganho estático e dinâmica nãolineares. O controlador preditivo neural envolve duas outras teorias para o seu desenvolvimento, a primeira referente ao controle preditivo e a outra a redes neurais artificiais (RNA s). Este controlador pode ser dividido em dois blocos, um responsável pela identificação e outro pelo o cálculo do sinal de controle. Para realizar a identificação neural é utilizada uma RNA com arquitetura feedforward multicamadas com aprendizagem baseada na metodologia da Propagação Retroativa do Erro (Error Back Propagation). A partir de dados de entrada e saída da planta é iniciado o treinamento offline da rede. Dessa forma, os pesos sinápticos são ajustados e a rede está apta para representar o sistema com a máxima precisão possível. O modelo neural gerado é usado para predizer as saídas futuras do sistema, com isso o otimizador calcula uma série de ações de controle, através da minimização de uma função objetivo quadrática, fazendo com que a saída do processo siga um sinal de referência desejado. Foram desenvolvidos dois aplicativos, ambos na plataforma Builder C++, o primeiro realiza a identificação, via redes neurais e o segundo é responsável pelo controle do processo. As ferramentas aqui implementadas e aplicadas são genéricas, ambas permitem a aplicação da estrutura de controle a qualquer novo processo

Processamento Inteligente de Sinais de Pressão e Temperatura Adquiridos Através de Sensores Permanentes em Poços de Petróleo

Originally aimed at operational objectives, the continuous measurement of well bottomhole pressure and temperature, recorded by permanent downhole gauges (PDG), finds vast applicability in reservoir management. It contributes for the monitoring of well performance and makes it possible to estimate reservoir parameters on the long term. However, notwithstanding its unquestionable value, data from PDG is characterized by a large noise content. Moreover, the presence of outliers within valid signal measurements seems to be a major problem as well. In this work, the initial treatment of PDG signals is addressed, based on curve smoothing, self-organizing maps and the discrete wavelet transform. Additionally, a system based on the coupling of fuzzy clustering with feed-forward neural networks is proposed for transient detection. The obtained results were considered quite satisfactory for offshore wells and matched real requisites for utilization