Sistema Híbrido de Inferência Baseado em Análise de Componentes Principais e Redes Neurais Artificiais Aplicado a Plantas de Processamento de Gás Natural

Nowadays, where the market competition requires products with better quality and a constant search for cost savings and a better use of raw materials, the research for more efficient control strategies becomes vital. In Natural Gas Processin Units (NGPUs), as in the most chemical processes, the quality control is accomplished through their products composition. However, the chemical composition analysis has a long measurement time, even when performed by instruments such as gas chromatographs. This fact hinders the development of control strategies to provide a better process yield. The natural gas processing is one of the most important activities in the petroleum industry. The main economic product of a NGPU is the liquefied petroleum gas (LPG). The LPG is ideally composed by propane and butane, however, in practice, its composition has some contaminants, such as ethane and pentane. In this work is proposed an inferential system using neural networks to estimate the ethane and pentane mole fractions in LPG and the propane mole fraction in residual gas. The goal is to provide the values of these estimated variables in every minute using a single multilayer neural network, making it possibly to apply inferential control techniques in order to monitor the LPG quality and to reduce the propane loss in the process. To develop this work a NGPU was simulated in HYSYS R software, composed by two distillation collumns: deethanizer and debutanizer. The inference is performed through the process variables of the PID controllers present in the instrumentation of these columns. To reduce the complexity of the inferential neural network is used the statistical technique of principal component analysis to decrease the number of network inputs, thus forming a hybrid inferential system. It is also proposed in this work a simple strategy to correct the inferential system in real-time, based on measurements of the chromatographs which may exist in process under study

Contribuição para o estudo do embarque de uma rede neural artificial em field programmable gate array (FPGA)

This study shows the implementation and the embedding of an Artificial Neural Network (ANN) in hardware, or in a programmable device, as a field programmable gate array (FPGA). This work allowed the exploration of different implementations, described in VHDL, of multilayer perceptrons ANN. Due to the parallelism inherent to ANNs, there are disadvantages in software implementations due to the sequential nature of the Von Neumann architectures. As an alternative to this problem, there is a hardware implementation that allows to exploit all the parallelism implicit in this model. Currently, there is an increase in use of FPGAs as a platform to implement neural networks in hardware, exploiting the high processing power, low cost, ease of programming and ability to reconfigure the circuit, allowing the network to adapt to different applications. Given this context, the aim is to develop arrays of neural networks in hardware, a flexible architecture, in which it is possible to add or remove neurons, and mainly, modify the network topology, in order to enable a modular network of fixed-point arithmetic in a FPGA. Five synthesis of VHDL descriptions were produced: two for the neuron with one or two entrances, and three different architectures of ANN. The descriptions of the used architectures became very modular, easily allowing the increase or decrease of the number of neurons. As a result, some complete neural networks were implemented in FPGA, in fixed-point arithmetic, with a high-capacity parallel processing

Desenvolvimento de uma arquitetura em hardware prototipada em FPGA para aplicações genéricas utilizando redes neurais artificiais embarcadas

This work proposes hardware architecture, VHDL described, developed to embedded Artificial Neural Network (ANN), Multilayer Perceptron (MLP). The present work idealizes that, in this architecture, ANN applications could easily embed several different topologies of MLP network industrial field. The MLP topology in which the architecture can be configured is defined by a simple and specifically data input (instructions) that determines the layers and Perceptron quantity of the network. In order to set several MLP topologies, many components (datapath) and a controller were developed to execute these instructions. Thus, an user defines a group of previously known instructions which determine ANN characteristics. The system will guarantee the MLP execution through the neural processors (Perceptrons), the components of datapath and the controller that were developed. In other way, the biases and the weights must be static, the ANN that will be embedded must had been trained previously, in off-line way. The knowledge of system internal characteristics and the VHDL language by the user are not needed. The reconfigurable FPGA device was used to implement, simulate and test all the system, allowing application in several real daily problems

Efeito da inclinação do plano de terra para antenas de microfita multicamadas com elementos parasitas fractais

This work presents the analysis of an antenna of fractal microstrip of Koch with dielectric multilayers and inclinations in the ground plane, whose values of the angles are zero degree (without inclinations), three, seven and twelve degrees. This antenna consists of three dielectric layers arranged vertically on each other, using feeding microstrip line in patch 1, of the first layer, which will feed the remaining patches of the upper layers by electromagnetic coupling. The objective of this work is to analyze the effects caused by increase of the angle of inclination of the ground plane in some antenna parameters such as return loss, resonant frequency, bandwidth and radiation pattern. The presented results demonstrate that with the increase of the inclination angle it is possible to get antennas with characteristics multiband, with bigger bandwidth, and improving the impedance matching for each case analyzed, especially the larger angle

Análise espectral de reflectarrays com substratos de duas camadas dielétricas anisotrópicas uniaxiais

Recently, an amazing development has been observed in telecommunication systems. Two good examples of this development are observed in mobile communication and aerospace systems. This impressive development is related to the increasing need for receiving and transmitting communication signals. Particularly, this development has required the study of new antennas and filters. This work presents a fullwave analysis of reflectarrays. The considered structures are composed by arrays of rectangular conducting patches printed on multilayer dieletric substrates, that are mounted on a ground plane. The analysis is developed in the spectral domain, using an equivalent transmission line method in combination with Galerkin method. Results for the reflection coefficient of these structures are presented and compared to those available in the literature. A good agreement was observed. Particularly, the developed analysis uses the transmission lines theory in combination with the incident potentials and the field continuity equations, at the structures interfaces, for obtaining the scattered field components expressions as function of the patch surface currents and of the incident field. Galerkin method is used to determine the unknown coefficients in the boundary value problem. Curves for the reflection coefficient of several reflectarray geometries are presented as function of frequency and of the structural parameters

Análise do espalhamento espectral em superfícies de estruturas complexas para comunicações móveis

In this work, the transmission line method is explored on the study of the propagation phenomenon in nonhomogeneous walls with finite thickness. It is evaluated the efficiency and applicability of the method, considering materials like gypsum, wood and brick, found in the composition of the structures of walls in question. The results obtained in this work are compared to those available in the literature, for several particular cases. A good agreement is observed, showing that the performed analysis is accurate and efficient in modeling, for instance, the wave propagation through building walls and integrated circuit layers in mobile communication and radar system applications. Later, simulations of resistive sheets devices such as Salisbury screens and Jaumann absorbers and of transmission lines made of metal-insulator-semiconductor (MIS) are made. Thereafter, it is described a study on frequency surface selective structures (FSS). It is proposed the development of devices and microwave integrated circuits (MIC) of such structures, for the accomplishment of experiments. Finally, future works are suggested, for instance, on the development of reflectarrays, frequency selective surfaces with dissimilar elements, and coupled frequency selective surfaces with elements located on different layers

Aplicação do método LTT às estruturas retangulares e triangulares em multicamadas e empilhadas em substratos PBG para comunicações móveis

Recently the planar antennas have been studied due to their characteristics as well as the advantages that they offers when compared with another types of antennas. In the mobile communications area, the need for this kind of antennas have became each time bigger due to the intense increase of the mobile communications that needs of antennas which operate in multifrequency and wide bandwidth. The microstrip antennas presents narrow bandwidth due the loss in the dielectric generated by radiation. Another limitation is the radiation pattern degradation due the generation of surface waves in the substrate. In this work some used techniques to minimize the disadvantages (previously mentioned) of the use of microstrip antennas are presented, those are: substrates with PBG material - Photonic Bandgap, multilayer antennas and with stacked patches. The developed analysis in this work used the TTL - Transverse Transmission Line method in the domain of Fourier transform, that uses a component of propagation in the y direction (transverse to the direction real of propagation z), treating the general equations of electric and magnetic field as functions of y and y . This work has as objective the application of the TTL method to microstrip structures with single and multilayers of rectangular and triangular patches, to obtaining the resonance frequency and radiation pattern of each structure. This method is applied for the treatment of the fields in stacked structures. The Homogenization theory will be applied to obtaining the effective permittivity for s and p polarizations of the substrate composed of PBG material. Numerical results for the triangular and rectangular antennas with single layer, multilayers resonators with triangular and rectangular patches are presented (in photonic and isotropic substrates). Conclusions and suggestions for continuity of this work are presented

Análise espectral de reflectarrays com substrato de duas camadas dielétricas anisotrópicas uniaxiais

Recently, an amazing development has been observed in telecommunication systems. Two good examples of this development are observed in mobile communication and aerospace systems. This impressive development is related to the increasing need for receiving and transmitting communication signals. Particularly, this development has required the study of new antennas and filters. This work presents a fullwave analysis of reflectarrays. The considered structures are composed by arrays of rectangular conducting patches printed on multilayer dieletric substrates, that are mounted on a ground plane. The analysis is developed in the spectral domain, using an equivalent transmission line method in combination with Galerkin method. Results for the reflection coefficient of these structures are presented and compared to those available in the literature. A good agreement was observed. Particularly, the developed analysis uses the transmission lines theory in combination with the incident potentials and the field continuity equations, at the structures interfaces, for obtaining the scattered field components expressions as function of the patch surface currents and of the incident field. Galerkin method is used to determine the unknown coefficients in the boundary value problem. Curves for the reflection coefficient of several reflectarray geometries are presented as function of frequency and of the structural parameters

Sistema inteligente para detecção de manchas de óleo na superfície marinha através de imagens de SAR

Oil spill on the sea, accidental or not, generates enormous negative consequences for the affected area. The damages are ambient and economic, mainly with the proximity of these spots of preservation areas and/or coastal zones. The development of automatic techniques for identification of oil spots on the sea surface, captured through Radar images, assist in a complete monitoring of the oceans and seas. However spots of different origins can be visualized in this type of imaging, which is a very difficult task. The system proposed in this work, based on techniques of digital image processing and artificial neural network, has the objective to identify the analyzed spot and to discern between oil and other generating phenomena of spot. Tests in functional blocks that compose the proposed system allow the implementation of different algorithms, as well as its detailed and prompt analysis. The algorithms of digital image processing (speckle filtering and gradient), as well as classifier algorithms (Multilayer Perceptron, Radial Basis Function, Support Vector Machine and Committe Machine) are presented and commented.The final performance of the system, with different kind of classifiers, is presented by ROC curve. The true positive rates are considered agreed with the literature about oil slick detection through SAR images presents

Modelagem de Superfícies Seletivas de Freqüência e Antenas de Microfita utilizando Redes Neurais Artificiais

This work has as main objective the application of Artificial Neural Networks, ANN, in the resolution of problems of RF /microwaves devices, as for example the prediction of the frequency response of some structures in an interest region. Artificial Neural Networks, are presently a alternative to the current methods of analysis of microwaves structures. Therefore they are capable to learn, and the more important to generalize the acquired knowledge, from any type of available data, keeping the precision of the original technique and adding the low computational cost of the neural models. For this reason, artificial neural networks are being increasily used for modeling microwaves devices. Multilayer Perceptron and Radial Base Functions models are used in this work. The advantages/disadvantages of these models and the referring algorithms of training of each one are described. Microwave planar devices, as Frequency Selective Surfaces and microstrip antennas, are in evidence due the increasing necessities of filtering and separation of eletromagnetic waves and the miniaturization of RF devices. Therefore, it is of fundamental importance the study of the structural parameters of these devices in a fast and accurate way. The presented results, show to the capacities of the neural techniques for modeling both Frequency Selective Surfaces and antennas

Controle inteligente de sistemas eletroidráulicos utilizando redes neurais artificiais

This work describes the development of a nonlinear control strategy for an electro-hydraulic actuated system. The system to be controlled is represented by a third order ordinary differential equation subject to a dead-zone input. The control strategy is based on a nonlinear control scheme, combined with an artificial intelligence algorithm, namely, the method of feedback linearization and an artificial neural network. It is shown that, when such a hard nonlinearity and modeling inaccuracies are considered, the nonlinear technique alone is not enough to ensure a good performance of the controller. Therefore, a compensation strategy based on artificial neural networks, which have been notoriously used in systems that require the simulation of the process of human inference, is used. The multilayer perceptron network and the radial basis functions network as well are adopted and mathematically implemented within the control law. On this basis, the compensation ability considering both networks is compared. Furthermore, the application of new intelligent control strategies for nonlinear and uncertain mechanical systems are proposed, showing that the combination of a nonlinear control methodology and artificial neural networks improves the overall control system performance. Numerical results are presented to demonstrate the efficacy of the proposed control system

Filtros acústicos em cristais fonônicos

In this work, we have studied the acoustic phonon wave propagation within the periodic and quasiperiodic superlattices of Fibonacci type. These structures are formed by phononic crystals, whose periodicity allows the raise of regions known as stop bands, which prevent the phonon propagation throughout the structure for specific frequency values. This phenomenon allows the construction of acoustic filters with great technological potential. Our theoretical model were based on the method of the transfer matrix, thery acoustics phonons which describes the propagation of the transverse and longitudinal modes within a unit cell, linking them with the precedent cell in the multilayer structure. The transfer matrix is built taking into account the elastic and electromagnetic boundary conditions in the superllatice interfaces, and it is related to the coupled differential equation solutions (elastic and electromagnetic) that describe each model under consideration. We investigated the piezoelectric properties of GaN and AlN the nitride semiconductors, whose properties are important to applications in the semiconductor device industry. The calculations that characterize the piezoelectric system, depend strongly on the cubic (zinc-bend) and hexagonal (wurtzite) crystal symmetries, that are described the elastic and piezoelectric tensors. The investigation of the liquid Hg (mercury), Ga (gallium) and Ar (argon) systems in static conditions also using the classical theory of elasticity. Together with the Euler s equation of fluid mechanics they one solved to the solid/liquid and the liquid/liquid interfaces to obtain and discuss several interesting physical results. In particular, the acoustical filters obtained from these structures are again presented and their features discussed

Multicamadas Ferromagnéticas Nanométricas Periódicas e Quasiperiódicas

We present a study of nanostructured magnetic multilayer systems in order to syn- thesize and analyze the properties of periodic and quasiperiodic structures. This work evolved from the deployment and improvement of the sputtering technique in our labora- tories, through development of a methodology to synthesize single crystal ultrathin Fe (100) films, to the final goal of growing periodic and quasiperiodic Fe/Cr multilayers and investi- gating bilinear and biquadratic exchange coupling between ferromagnetic layer dependence for each generation. Initially we systematically studied the related effects between deposition parameters and the magnetic properties of ultrathin Fe films, grown by DC magnetron sput- tering on MgO(100) substrates. We modified deposition temperature and film thickness, in order to improve production and reproduction of nanostructured monocrystalline Fe films. For this set of samples we measured MOKE, FMR, AFM and XPS, with the aim of investi- gating their magnocrystalline and structural properties. From the magnetic viewpoint, the MOKE and FMR results showed an increase in magnetocrystalline anisotropy due to in- creased temperature. AFM measurements provided information about thickness and surface roughness, whereas XPS results were used to analyze film purity. The best set of parame- ters was used in the next stage: investigation of the structural effect on magnetic multilayer properties. In this stage multilayers composed of interspersed Fe and Cr films are deposited, following the Fibonacci periodic and quasiperiodic growth sequence on MgO (100) substrates. The behavior of MOKE and FMR curves exhibit bilinear and biquadratic exchange coupling between the ferromagnetic layers. By computationally adjusting magnetization curves, it was possible to determine the nature and intensity of the interaction between adjacent Fe layers. After finding the global minimum of magnetic energy, we used the equilibrium an- gles to obtain magnetization and magnetoresistance curves. The results observed over the course of this study demonstrate the efficiency and versatility of the sputtering technique in the synthesis of ultrathin films and high-quality multilayers. This allows the deposition of magnetic nanostructures with well-defined magnetization and magnetoresistance parameters and possible technological applications

Ondas de spin em nanofilmes nagnéticos acoplados quasiperiodicamente

We studied the spin waves modes that can propagate in magnetic multilayers composed of ferromagnetic metallic films in the nanometer scale. The ferromagnetic films (iron) are separated and coupled through the nonmagnetic spacer films (chromium). The films that make up the multilayer are stacked in a quasiperiodic pattern, following the Fibonacci and double period sequences. We used a phenomenological theory taking into account: the Zeeman energy (between the ferromagnetic films and the external magnetic field), the energy of the magneto-crystalline anisotropy (present in the ferromagnetic films), the energy of the bilinear and biquadratic couplings (between the ferromagnetic films) and the energy of the dipole-dipole interaction (between the ferromagnetic films), to describe the system. The total magnetic energy of the system is numerically minimized and the equilibrium angles of the magnetization of each ferromagnetic film are determined. We solved the equation of motion of the multilayer to find the dispersion relation for the system and, as a consequence, the spin waves modes frequencies. Our theoretical results show that, in the case of trilayers (Fe/Cr/Fe), our model reproduces with excellent agreement experimental results of Brillouin light scattering, known from the literature, by adjusting the physical parameters of the nanofilms. Furthermore, we generalize the model to N ferromagnetic layers which allowed us to determine how complex these systems become when we increase the number of components. It is worth noting that our theoretical calculations generalize all the results known from the literature

Excitações em cristais fotônicos unidimensionais

In this work, we present a theoretical study of the propagation of electromagnetic waves in multilayer structures called Photonic Crystals. For this purpose, we investigate the phonon-polariton band gaps in periodic and quasi-periodic (Fibonacci-type) multilayers made up of both positive and negative refractive index materials in the terahertz (THz) region. The behavior of the polaritonic band gaps as a function of the multilayer period is investigated systematically. We use a theoretical model based on the formalism of transfer matrix in order to simplify the algebra involved in obtaining the dispersion relation of phonon-polaritons (bulk and surface modes). We also present a quantitative analysis of the results, pointing out the distribution of the allowed polaritonic bandwidths for high Fibonacci generations, which gives good insight about their localization and power laws. We calculate the emittance spectrum of the electromagnetic radiation, in THZ frequency, normally and obliquely incident (s and p polarized modes) on a one-dimensional multilayer structure composed of positive and negative refractive index materials organized periodically and quasi-periodically. We model the negative refractive index material by a effective medium whose electric permittivity is characterized by a phonon-polariton frequency dependent dielectric function, while for the magnetic permeability we have a Drude like frequency-dependent function. Similarity to the one-dimensional photonic crystal, this layered effective medium, called polaritonic Crystals, allow us the control of the electromagnetic propagation, generating regions named polaritonic bandgap. The emittance spectra are determined by means of a well known theoretical model based on Kirchoff s second law, together with a transfer matrix formalism. Our results shows that the omnidirectional band gaps will appear in the THz regime, in a well defined interval, that are independent of polarization in periodic case as well as in quasiperiodic case