Implementação de uma arquitetura fuzzy neural em hardware com treinamento online

Computational Intelligence Methods have been expanding to industrial applications motivated by their ability to solve problems in engineering. Therefore, the embedded systems follow the same idea of using computational intelligence tools embedded on machines. There are several works in the area of embedded systems and intelligent systems. However, there are a few papers that have joined both areas. The aim of this study was to implement an adaptive fuzzy neural hardware with online training embedded on Field Programmable Gate Array – FPGA. The system adaptation can occur during the execution of a given application, aiming online performance improvement. The proposed system architecture is modular, allowing different configurations of fuzzy neural network topologies with online training. The proposed system was applied to: mathematical function interpolation, pattern classification and selfcompensation of industrial sensors. The proposed system achieves satisfactory performance in both tasks. The experiments results shows the advantages and disadvantages of online training in hardware when performed in parallel and sequentially ways. The sequentially training method provides economy in FPGA area, however, increases the complexity of architecture actions. The parallel training method achieves high performance and reduced processing time, the pipeline technique is used to increase the proposed architecture performance. The study development was based on available tools for FPGA circuits.

Contribuições à estratégia de controle sem detecção de harmônicos aplicada a um filtro ativo de potência paralelo

The conventional control schemes applied to Shunt Active Power Filters (SAPF) are Harmonic extractor-based strategies (HEBSs) because their effectiveness depends on how quickly and accurately the harmonic components of the nonlinear loads are identified. The SAPF can be also implemented without the use of the load harmonic extractors. In this case, the harmonic compensating term is obtained from the system active power balance. These systems can be considered as balanced-energy-based schemes (BEBSs) and their performance depends on how fast the system reaches the equilibrium state. In this case, the phase currents of the power grid are indirectly regulated by double sequence controllers with two degrees of freedom, where the internal model principle is employed to avoid reference frame transformation. Additionally the DSC controller presents robustness when the SAPF is operating under unbalanced conditions. Furthermore, SAPF implemented without harmonic detection schemes compensate simultaneously harmonic distortion and reactive power of the load. Their compensation capabilities, however, are limited by the SAPF power converter rating. Such a restriction can be minimized if the level of the reactive power correction is managed. In this work an estimation scheme for determining the filter currents is introduced to manage the compensation of reactive power. Experimental results are shown for demonstrating the performance of the proposed SAPF system.

Metodologia númerico-experimental para avaliação da espessura de filmes de nitreto de titânio obtidos por processamento a plasma

In recent decades, changes in the surface properties of materials have been used to improve their tribological characteristics. However, this improvement depends on the process, treatment time and, primarily, the thickness of this surface film layer. Physical vapor deposition (PVD) of titanium nitrate (TiN) has been used to increase the surface hardness of metallic materials. Thus, the aim of the present study was to propose a numerical-experimental method to assess the film thickness (l) of TiN deposited by PVD. To reach this objective, experimental results of hardness (H) assays were combined with a numerical simulation to study the behavior of this property as a function of maximum penetration depth of the indenter (hmax) into the film/substrate conjugate. Two methodologies were adopted to determine film thickness. The first consists of the numerical results of the H x hmax curve with the experimental curve obtained by the instrumental indentation test. This methodology was used successfully in a TiN-coated titanium (Ti) conjugate. A second strategy combined the numerical results of the Hv x hmax curve with Vickers experimental hardness data (Hv). This methodology was applied to a TiN-coated M2 tool steel conjugate. The mechanical properties of the materials studied were also determined in the present study. The thicknesses results obtained for the two conjugates were compatible with their experimental data.

Caracterização e separação física de placas de circuito impresso de computadores obsoletos

The unbridled consumption of electronic equipment associated with fast immersion of new technologies on the market leads to the accelerated growth of electronic waste. Such waste mostly contains printed circuit boards in its structure. Printed circuit boards have many metals, including heavy metals, being highly toxic. Electronic waste is discarded improperly and indiscriminately, usually without any previous treatment and with other municipal waste, contaminating the environment and causing serious problems to human health. Beyond these metals, there are also precious metals and high value-added basis, that can be recovered and recycled, reducing the exploration of natural resources. Thus, due to the high growth potential and reuse of these waste treatment processes, characterization and separation were applied to the printed circuit boards. The printed circuit boards were subjected to physical treatments such as dismantling, crushing, sizing separation, magnetic separation and chemical treatments such as pyrolysis and leaching. Through characterization process (pyrolysis and leaching) the proportions of the components of the granulometric range were determined: 46,08% of metals; 23,32% of polymers and 30,60% of ceramics. It was also observed by particle size separation that metal components tend to concentrate in coarse fractions, while polymeric and ceramic components in fine fractions. From the magnetic separation process was obtained 12,08% of magnetic material and 82,33% of non-magnetic material.

Sistema inteligente para o processamento de imagens digitais intrabucais oclusais

Several are the areas in which digital images are used in solving day-to-day problems. In medicine the use of computer systems have improved the diagnosis and medical interpretations. In dentistry it’s not different, increasingly procedures assisted by computers have support dentists in their tasks. Set in this context, an area of dentistry known as public oral health is responsible for diagnosis and oral health treatment of a population. To this end, oral visual inspections are held in order to obtain oral health status information of a given population. From this collection of information, also known as epidemiological survey, the dentist can plan and evaluate taken actions for the different problems identified. This procedure has limiting factors, such as a limited number of qualified professionals to perform these tasks, different diagnoses interpretations among other factors. Given this context came the ideia of using intelligent systems techniques in supporting carrying out these tasks. Thus, it was proposed in this paper the development of an intelligent system able to segment, count and classify teeth from occlusal intraoral digital photographic images. The proposed system makes combined use of machine learning techniques and digital image processing. We first carried out a color-based segmentation on regions of interest, teeth and non teeth, in the images through the use of Support Vector Machine. After identifying these regions were used techniques based on morphological operators such as erosion and transformed watershed for counting and detecting the boundaries of the teeth, respectively. With the border detection of teeth was possible to calculate the Fourier descriptors for their shape and the position descriptors. Then the teeth were classified according to their types through the use of the SVM from the method one-against-all used in multiclass problem. The multiclass classification problem has been approached in two different ways. In the first approach we have considered three class types: molar, premolar and non teeth, while the second approach were considered five class types: molar, premolar, canine, incisor and non teeth. The system presented a satisfactory performance in the segmenting, counting and classification of teeth present in the images.

Sistema inteligente para o processamento de imagens digitais intrabucais oclusais

Several are the areas in which digital images are used in solving day-to-day problems. In medicine the use of computer systems have improved the diagnosis and medical interpretations. In dentistry it’s not different, increasingly procedures assisted by computers have support dentists in their tasks. Set in this context, an area of dentistry known as public oral health is responsible for diagnosis and oral health treatment of a population. To this end, oral visual inspections are held in order to obtain oral health status information of a given population. From this collection of information, also known as epidemiological survey, the dentist can plan and evaluate taken actions for the different problems identified. This procedure has limiting factors, such as a limited number of qualified professionals to perform these tasks, different diagnoses interpretations among other factors. Given this context came the ideia of using intelligent systems techniques in supporting carrying out these tasks. Thus, it was proposed in this paper the development of an intelligent system able to segment, count and classify teeth from occlusal intraoral digital photographic images. The proposed system makes combined use of machine learning techniques and digital image processing. We first carried out a color-based segmentation on regions of interest, teeth and non teeth, in the images through the use of Support Vector Machine. After identifying these regions were used techniques based on morphological operators such as erosion and transformed watershed for counting and detecting the boundaries of the teeth, respectively. With the border detection of teeth was possible to calculate the Fourier descriptors for their shape and the position descriptors. Then the teeth were classified according to their types through the use of the SVM from the method one-against-all used in multiclass problem. The multiclass classification problem has been approached in two different ways. In the first approach we have considered three class types: molar, premolar and non teeth, while the second approach were considered five class types: molar, premolar, canine, incisor and non teeth. The system presented a satisfactory performance in the segmenting, counting and classification of teeth present in the images.

Uma arquitetura distribuída de hardware e software para controle de um robô móvel autônomo

BRITTO, Ricardo S.; MEDEIROS, Adelardo A. D.; ALSINA, Pablo J. Uma arquitetura distribuída de hardware e software para controle de um robô móvel autônomo. In: SIMPÓSIO BRASILEIRO DE AUTOMAÇÃO INTELIGENTE,8., 2007, Florianópolis. Anais... Florianópolis: SBAI, 2007.

Processamento e análise de imagens na caracterização da superfície do titânio submetido a um ensaio de cultura de células

This work aims to develop a methodology for analysis of images using overlapping, which assists in identification of microstructural features in areas of titanium, which may be associated with its biological response. That way, surfaces of titanium heat treated for 08 (eight) different ways have been subjected to a test culture of cells. It was a relationship between the grain, texture and shape of grains of surface of titanium (attacked) trying to relate to the process of proliferation and adhesion. We used an open source software for cell counting adhered to the surface of titanium. The juxtaposition of images before and after cell culture was obtained with the aid of micro-hardness of impressions made on the surface of samples. From this image where there is overlap, it is possible to study a possible relationship between cell growth with microstructural characteristics of the surface of titanium. This methodology was efficient to describe a set of procedures that are useful in the analysis of surfaces of titanium subjected to a culture of cells

Estudo e desenvolvimento de um capacitor eletrolítico de nióbio

It seeks to find an alternative to the current tantalum electrolytic capacitors in the market due to its high cost. Niobium is a potential substitute, since both belong to the same group of the periodic table and because of this have many similar physical and chemical properties. Niobium has several technologically important applications, and Brazil has the largest reserves, around 96%. There are including niobium in reserves of tantalite and columbite in Rio Grande do Norte. These electrolytic capacitors have high capacitance specifies, ie they can store high energy in small volumes compared to other types of capacitors. This is the main attraction of this type of capacitor because is growing demand in the production of capacitors with capacitance specifies increasingly high, this because of the miniaturization of various devices such as GPS devices, televisions, computers, phones and many others. The production route of the capacitor was made by powder metallurgy. The initial niobium powder supplied by EEL-USP was first characterized by XRD, SEM, XRF and laser particle size, to then be sieved into three particle size, 200, 400 e 635mesh. The powders were then compacted and sintered at 1350, 1450 and 1550°C using two sintering time 30 and 60min. Sintering is one of the most important parts of the process as it affects properties as porosity and surface cleaning of the samples, which greatly affected the quality of the capacitor. The sintered samples then underwent a process of anodic oxidation, which created a thin film of niobium pentóxido over the whole porous surface of the sample, this film is the dielectric capacitor. The oxidation process variables influence the performance of the film and therefore the capacitor. The samples were characterized by electrical measurements of capacitance, loss factor, ESR, relative density, porosity and surface area. After the characterizations was made an annealing in air ate 260ºC for 60min. After this treatment were made again the electrical measurements. The particle size of powders and sintering affected the porosity and in turn the specific area of the samples. The larger de area of the capacitor, greater is the capacitance. The powder showed the highest capacitance was with the smallest particle size. Higher temperatures and times of sintering caused samples with smaller surface area, but on the other hand the cleaning surface impurities was higher for this cases. So a balance must be made between the gain that is achieved with the cleaning of impurities and the loss with the decreased in specific area. The best results were obtained for the temperature of 1450ºC/60min. The influence of annealing on the loss factor and ESR did not follow a well-defined pattern, because their values increased in some cases and decreased in others. The most interesting results due to heat treatment were with respect to capacitance, which showed an increase for all samples after treatment

Estudo sobre o efeito dos parâmetros de processamento dos pós e sinterização do aço inox 316l reforçado com NbC

This masther dissertation presents a contribution to the study of 316L stainless steel sintering aiming to study their behavior in the milling process and the effect of isotherm temperature on the microstructure and mechanical properties. The 316L stainless steel is a widely used alloy for their high corrosion resistance property. However its application is limited by the low wear resistance consequence of its low hardness. In previous work we analyzed the effect of sintering additives as NbC and TaC. This study aims at deepening the understanding of sintering, analyzing the effect of grinding on particle size and microstructure and the effect of heating rate and soaking time on the sintered microstructure and on their microhardness. Were milled 316L powders with NbC at 1, 5 and 24 hours respectively. Particulates were characterized by SEM and . Cylindrical samples height and diameter of 5.0 mm were compacted at 700 MPa. The sintering conditions were: heating rate 5, 10 and 15◦C/min, temperature 1000, 1100, 1200, 1290 and 1300◦C, and soaking times of 30 and 60min. The cooling rate was maintained at 25◦C/min. All samples were sintered in a vacuum furnace. The sintered microstructure were characterized by optical and electron microscopy as well as density and microhardness. It was observed that the milling process has an influence on sintering, as well as temperature. The major effect was caused by firing temperature, followed by the grinding and heating rate. In this case, the highest rates correspond to higher sintering.

Estudo do efeito da adição da escória de aciaria no processamento e propriedades finais de telhas cerâmicas

This research presents an overview of the addition steelwork dust of ceramic shingles in order to contribute to the utilization use of such residue. The ceramic industry perspective in the Brazilian State of Piauí is quite promising. Unlike other productive sectors, the ceramic industry uses basically natural raw materials. Its final products are, in short, the result of transforming clay compounds. These raw materials are composed primarily of aluminum oxide, silicon, iron, sodium, magnesium, end calcium, among others. It was verified that steelwork dust is composed primarily of these same oxides, so that its incorporation in to structural ceramics is a very reasonable idea. Both clay and steelwork powder were characterized by AG, XRF, XRD, TGA and DTA. In addition, steelwork dust samples containing (0%, 5%, 10%, 15%, 20% and 25%) were extruded and burned at 800°C, 850°C, 900°C and 950°C. Then t echnological tests of linear shrinkage, water uptake, apparent porosity, apparent density and flexural strengthwere carried at. The results showed the possibility of using steelwork powder in ceramic shingles until 15% significant improvement in physical and mechanical properties. This behavior shows the possibility of burning at temperatures lower than 850ºC, thus promoting a product final cost reduction

Processamento e caracterização de compósitos Poli(METACRILATO DE METILA)/Sílica (PMMA/SiO2)

Currently the search for new materials with properties suitable for specific applications has increased the number of researches that aim to address market needs. The poly (methyl methacrylate) (PMMA) is one of the most important polymers of the family of polyacrylates and polymethacrylates, especially for its unique optical properties and weathering resistance, and exceptional hardness and gloss. The development of polymer composites by the addition of inorganic fillers to the PMMA matrix increases the potential use of this polymer in various fields of application. The most commonly used inorganic fillers are particles of silica (SiO2), modified clays, graphite and carbon nanotubes. The main objective of this work is the development of PMMA/SiO2 composites at different concentrations of SiO2, for new applications as engineering plastics. The composites were produced by extrusion of tubular film, and obtained via solution for application to commercial PMMA plates, and also by injection molding, for improved the abrasion and scratch resistance of PMMA without compromising transparency. The effects of the addition of silica particles in the polymer matrix properties were evaluated by the maximum tensile strength, hardness, abrasion and scratch resistance, in addition to preliminary characterization by torque rheometry and melt flow rate. The results indicated that it is possible to use silica particles in a PMMA matrix, and a higher silica concentration produced an increase of the abrasion and scratch resistance, hardness, and reduced tensile strength

Processamento e caracterização da blenda poli (metacrilato de metila) (PMMA) elastomerico e poli (tereftalato de etileno) (PET) pós-consumo

This work studied the immiscible blend of elastomeric poly(methyl methacrylate) (PMMA) with poly(ethylene terephthalate) (PET) bottle grade with and without the use of compatibilizer agent, poly(methyl methacrylate-co-glycidyl methacrylate - co-ethyl acrylate) (MGE). The characterizations of torque rheometry, melt flow index measurement (MFI), measuring the density and the degree of cristallinity by pycnometry, tensile testing, method of work essential fracture (EWF), scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were performed in pure polymer and blends PMMA/PET. The rheological results showed evidence of signs of chemical reaction between the epoxy group MGE with the end groups of the PET chains and also to the elastomeric phase of PMMA. The increase in the concentration of PET reduced torque and adding MGE increased the torque of the blend of PMMA/PET. The results of the MFI also show that elastomeric PMMA showed lower flow and thus higher viscosity than PET. In the results of picnometry observed that increasing the percentage of PET resulted in an increase in density and degree crystallinity of the blends PMMA/PET. The tensile test showed that increasing the percentage of PET resulted in an increase in ultimate strength and elastic modulus and decrease in elongation at break. However, in the phase inversion, where the blend showed evidence of a co-continuous morphology and also, with 30% PET dispersed phase and compatibilized with 5% MGE, there were significant results elongation at break compared to elastomeric PMMA. The applicability of the method of essential work of fracture was shown to be possible for most formulations. And it was observed that with increasing elastomeric PMMA in the formulations of the blends there was an improvement in specific amounts of essential work of fracture (We) and a decrease in the values of specific non-essential work of fracture (βWp)

Estudo de adsorvente obtido de terras diatomáceas para remoção de metais pesados em águas produzidas: processamento, caracterização e dinâmica em leito fixo

The acceleration of industrial growth in recent decades on all continents aroused the interest of the companies to counter the impacts produced on the environment, spurred primarily by major disasters in the petroleum industry. In this context, the water produced is responsible for the largest volume of effluent from the production and extraction of oil and natural gas. This effluent has in its composition some critical components such as inorganic salts, heavy metals (Fe, Cu, Zn, Pb, Cd, ), presence of oil and chemicals added in the various production processes. In response to impact, have been triggered by research alternative adsorbent materials for water treatment and water produced, in order to removing oils and acids and heavy metals. Many surveys of diatomaceous earth (diatomite) in Brazil involve studies on the physico-chemical, mineral deposits, extraction, processing and applications. The official estimated Jazi are around 2.5 million tonnes, the main located in the states of Bahia (44%) and Rio Grande do Norte (37,4%). Moreover, these two states appear as large offshore producers, earning a prominent role in research of adsorbents such as diatomite for treatment of water produced. Its main applications are as an agent of filtration, adsorption of oils and greases, industrial load and thermal insulator. The objective of this work was the processing and characterization of diatomite diatomaceous earth obtained from the municipality of Macaíba-RN (known locally as tabatinga) as a low cost regenerative adsorbent for removal of heavy metals in the application of water produced treatment. In this work we adopted a methodology for batch processing, practiced by small businesses located in producing regions of Brazil. The characterization was made by X-ray diffraction (XRD), scanning electron microscopy (SEM) and specific surface area (BET). Research conducted showed that the improvement process used was effective for small volume production of diatomite concentrated. The diatomite obtained was treated by calcination at temperature of 900 oC for 2 hours, with and without fluxing Na2CO3 (4%), according to optimal results in the literature. Column adsorption experiments were conducted to percolation of the in nature, calcined and calcined fluxing diatomites. Effluent was used as a saline solution containing ions of Cu, Zn, Na, Ca and Mg simulating the composition of produced waters in the state of Rio Grande do Norte, Brazil. The breakthrough curves for simultaneous removal of copper ions and zinc as a result, 84.3% for calcined diatomite and diatomite with 97.3 % for fluxing. The calcined fluxing diatomite was more efficient permeability through the bed and removal of copper and zinc ions. The fresh diatomite had trouble with the permeability through the bed under the conditions tested, compared with the other obtained diatomite. The results are presented as promising for application in the petroleum industry

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