Reconhecimento de voz através de unidades menores do que a palavra, utilizando Wavelet Packet e SVM, em uma nova estrutura hierárquica de decisão

The automatic speech recognition by machine has been the target of researchers in the past five decades. In this period have been numerous advances, such as in the field of recognition of isolated words (commands), which has very high rates of recognition, currently. However, we are still far from developing a system that could have a performance similar to the human being (automatic continuous speech recognition). One of the great challenges of searches for continuous speech recognition is the large amount of pattern. The modern languages such as English, French, Spanish and Portuguese have approximately 500,000 words or patterns to be identified. The purpose of this study is to use smaller units than the word such as phonemes, syllables and difones units as the basis for the speech recognition, aiming to recognize any words without necessarily using them. The main goal is to reduce the restriction imposed by the excessive amount of patterns. In order to validate this proposal, the system was tested in the isolated word recognition in dependent-case. The phonemes characteristics of the Brazil s Portuguese language were used to developed the hierarchy decision system. These decisions are made through the use of neural networks SVM (Support Vector Machines). The main speech features used were obtained from the Wavelet Packet Transform. The descriptors MFCC (Mel-Frequency Cepstral Coefficient) are also used in this work. It was concluded that the method proposed in this work, showed good results in the steps of recognition of vowels, consonants (syllables) and words when compared with other existing methods in literature

Uma arquitetura baseada em agentes para configuração dinâmica de aplicações inteligentes em ambiente foundation fieldbus

The industrial automation is directly linked to the development of information tecnology. Better hardware solutions, as well as improvements in software development methodologies make possible the rapid growth of the productive process control. In this thesis, we propose an architecture that will allow the joining of two technologies in hardware (industrial network) and software field (multiagent systems). The objective of this proposal is to join those technologies in a multiagent architecture to allow control strategies implementations in to field devices. With this, we intend develop an agents architecture to detect and solve problems which may occur in the industrial network environment. Our work ally machine learning with industrial context, become proposed multiagent architecture adaptable to unfamiliar or unexpected production environment. We used neural networks and presented an allocation strategies of these networks in industrial network field devices. With this we intend to improve decision support at plant level and allow operations human intervention independent

Avaliação de robustez, desempenho e aplicação do controlador adaptativo por posicionamento de pólos e estrutura variável

In this work, the variable structure adaptive pole placement controller (VS-APPC) robustness and performance are evaluated and this algorithm is applied in a motor control system. The controller robustness evaluation will be done through simulations, where will be introduced in the system the following adversities: time delay, actuator response boundeds, disturbances, parametric variation and unmodeled dynamics. The VS-APPC will be compared with PI control, pole placement control (PPC) and adaptive pole placement controller (APPC). The VS-APPC will be simulated to track a step and a sine reference. It will be applied in a three-phase induction motor control system to track a sine signal in the stator reference frame. Simulation and experimental results will prove the efficiency and robustness of this control strategy

Desempenho, eficiência de utilização dos nutrientes e estrutura do trato digestório de pintos de corte alimentados na fase pré-inicial com rações de diferentes formas físicas

Este estudo foi realizado com o objetivo de avaliar o efeito da forma física da ração sobre o desempenho, a utilização dos nutrientes das rações e o desenvolvimento do trato disgestório de pintos de corte na primeira semana de vida. Foram utilizados 150 pintos machos Cobb distribuídos ao acaso em três tratamentos com cinco repetições de dez aves. Os tratamentos consistiram das formas físicas farelada, triturada e peletizada de uma ração pré-inicial comercial. Com exceção da moela, os órgãos do trato digestório e a estrutura morfométrica do intestino não foram influenciados pela forma física da ração. Maior digestibilidade do N foi obtida com a ração peletizada, enquanto a ração triturada possibilitou maior digestibilidade do EE. A energia metabolizável determinada para a ração farelada foi menor que a obtida nas demais formas físicas da ração. As aves alimentadas com ração farelada tiveram menor consumo de ração e ganho de peso e pior conversão alimentar. Ingeriram ainda menos energia, apresentaram menores retenções de proteína, gordura e energia corporal e utilizaram maior proporção da energia ingerida para produção de calor. As rações pré-iniciais na forma peletizada ou triturada possibilitam maior aproveitamento dos nutrientes da ração pelos pintos e melhor desempenho.

Modelagem e controle de um microveículo aéreo: uma aplicação de estabilidade robusta com a técnica backstepping em uma estrutura hexarrotor

In this Thesis, the development of the dynamic model of multirotor unmanned aerial vehicle with vertical takeoff and landing characteristics, considering input nonlinearities and a full state robust backstepping controller are presented. The dynamic model is expressed using the Newton-Euler laws, aiming to obtain a better mathematical representation of the mechanical system for system analysis and control design, not only when it is hovering, but also when it is taking-off, or landing, or flying to perform a task. The input nonlinearities are the deadzone and saturation, where the gravitational effect and the inherent physical constrains of the rotors are related and addressed. The experimental multirotor aerial vehicle is equipped with an inertial measurement unit and a sonar sensor, which appropriately provides measurements of attitude and altitude. A real-time attitude estimation scheme based on the extended Kalman filter using quaternions was developed. Then, for robustness analysis, sensors were modeled as the ideal value with addition of an unknown bias and unknown white noise. The bounded robust attitude/altitude controller were derived based on globally uniformly practically asymptotically stable for real systems, that remains globally uniformly asymptotically stable if and only if their solutions are globally uniformly bounded, dealing with convergence and stability into a ball of the state space with non-null radius, under some assumptions. The Lyapunov analysis technique was used to prove the stability of the closed-loop system, compute bounds on control gains and guaranteeing desired bounds on attitude dynamics tracking errors in the presence of measurement disturbances. The controller laws were tested in numerical simulations and in an experimental hexarotor, developed at the UFRN Robotics Laboratory

Estrutura ANFIS modificada para identificação e controle de plantas com ampla faixa de operação e não linearidade acentuada

In this work a modification on ANFIS (Adaptive Network Based Fuzzy Inference System) structure is proposed to find a systematic method for nonlinear plants, with large operational range, identification and control, using linear local systems: models and controllers. This method is based on multiple model approach. This way, linear local models are obtained and then those models are combined by the proposed neurofuzzy structure. A metric that allows a satisfactory combination of those models is obtained after the structure training. It results on plant s global identification. A controller is projected for each local model. The global control is obtained by mixing local controllers signals. This is done by the modified ANFIS. The modification on ANFIS architecture allows the two neurofuzzy structures knowledge sharing. So the same metric obtained to combine models can be used to combine controllers. Two cases study are used to validate the new ANFIS structure. The knowledge sharing is evaluated in the second case study. It shows that just one modified ANFIS structure is necessary to combine linear models to identify, a nonlinear plant, and combine linear controllers to control this plant. The proposed method allows the usage of any identification and control techniques for local models and local controllers obtaining. It also reduces the complexity of ANFIS usage for identification and control. This work has prioritized simpler techniques for the identification and control systems to simplify the use of the method

Agentes polinizadores e produção de grãos em cultura de café arábica cv. “Catuaí Vermelho”

Coffee (Coffea arabica L.) plants had the pollination process of their flowers studied in three coffee producing regions in Brazil: the first in Ribeirao Preto, state of São Paulo, in September of 2004, the second in Carmo do Paraiba, in the state of Minas Gerais, in October of 2006, and, the third, in Altinopolis, state of São Paulo, in September of 2007. In all these locations the coffee variety used was the 'Catuai Vermelho'. The pollinator species visiting the flowers, the frequency with which the flowers were visited by the insects, and the type of material collected by them ( nectar and/or pollen) were determined by counts taking place at the first 10 minutes of each hour from 7:00 AM till 6:00 PM during the months mentioned above. To estimate the fructification percentage, 25 branches were identified with color threads and protected with wire frames covered with a nylon fabric to prevent them from being visited by insects. Another group of 25 branches was similarly identified but remained without any covering so that their flowers were thus freely visited by insects. on the average, the most frequent pollinator was the Africanized bee Apis mellifera (73.7%) followed by the bee Trigona spinipes (14.5%), and Tetragonisca angustula (9.5%). The Africanized bee collected nectar and pollen to almost the same extent: 58.5% of the first and 41.5% of the second. To collect nectar, that species increased its visiting frequency up to 1:00 PM. To collect pollen, A. mellifera visited the flowers up to 1:00 PM with frequency peaks between 8:00 and 9:00 AM. T. spinipes collected pollen (58,5%) and nectar (41.5%). T. angustula collected only pollen. The branches freely visited by insects produced 181 grains on the average whereas those covered with the nylon fabric produced only 81 grains, this meaning a reduction of 55.25% in grain production. Coffee grains produced on the freely visited branches had a mean weight (1.13 g) significantly superior to that of those produced on the covered branches (0.94 g).

Algoritmo SOM com estrutura hierárquica e dinâmica aplicado a compressão de imagens

ln this work the implementation of the SOM (Self Organizing Maps) algorithm or Kohonen neural network is presented in the form of hierarchical structures, applied to the compression of images. The main objective of this approach is to develop an Hierarchical SOM algorithm with static structure and another one with dynamic structure to generate codebooks (books of codes) in the process of the image Vector Quantization (VQ), reducing the time of processing and obtaining a good rate of compression of images with a minimum degradation of the quality in relation to the original image. Both self-organizing neural networks developed here, were denominated HSOM, for static case, and DHSOM, for the dynamic case. ln the first form, the hierarchical structure is previously defined and in the later this structure grows in an automatic way in agreement with heuristic rules that explore the data of the training group without use of external parameters. For the network, the heuristic mIes determine the dynamics of growth, the pruning of ramifications criteria, the flexibility and the size of children maps. The LBO (Linde-Buzo-Oray) algorithm or K-means, one ofthe more used algorithms to develop codebook for Vector Quantization, was used together with the algorithm of Kohonen in its basic form, that is, not hierarchical, as a reference to compare the performance of the algorithms here proposed. A performance analysis between the two hierarchical structures is also accomplished in this work. The efficiency of the proposed processing is verified by the reduction in the complexity computational compared to the traditional algorithms, as well as, through the quantitative analysis of the images reconstructed in function of the parameters: (PSNR) peak signal-to-noise ratio and (MSE) medium squared error

Mudanças na ultra-estrutura da parede celular de mangas 'Tommy Atkins' tratadas com cloreto de cálcio na pré-colheita

Mangas 'Tommy Atkins' produzidas na região de Ibirá, São Paulo, foram pulverizadas na pré-colheita com cloreto de cálcio, nas concentrações de 0,0%, 2,5% e 5,0%, em três épocas de seu desenvolvimento (40; 60 e 90 dias após a floração) a fim de verificar a influência do cálcio na estrutura da parede celular destes frutos através de microscopia eletrônica de transmissão, imediatamente após a colheita e depois de 35 dias de armazenamento. Para fixar o material da polpa, utilizou-se metodologia descrita por Jacob e Gowanlock (1995). Nas condições experimentais, verificou-se que os frutos do tratamento-controle (sem cloreto de cálcio), no dia da colheita, já apresentavam desestruturação da parede celular e dissolução da lamela média (LM). A degradação da parede celular ocorre inicialmente na LM, levando à formação de espaços vazios bastante distintos, apresentando uma dissolução ainda maior, com o armazenamento prolongado (35 dias). Os frutos tratados com cloreto de cálcio a 5,0% apresentaram uma LM bem definida e ausência de espaços vazios, mesmo após o armazenamento, mostrando ser uma concentração efetiva na preservação da lamela média.

Uma proposta de um controlador adaptativo por posicionamento de pólos e estrutura variável

There are two main approaches for using in adaptive controllers. One is the so-called model reference adaptive control (MRAC), and the other is the so-called adaptive pole placement control (APPC). In MRAC, a reference model is chosen to generate the desired trajectory that the plant output has to follow, and it can require cancellation of the plant zeros. Due to its flexibility in choosing the controller design methodology (state feedback, compensator design, linear quadratic, etc.) and the adaptive law (least squares, gradient, etc.), the APPC is the most general type of adaptive control. Traditionally, it has been developed in an indirect approach and, as an advantage, it may be applied to non-minimum phase plants, because do not involve plant zero-pole cancellations. The integration to variable structure systems allows to aggregate fast transient and robustness to parametric uncertainties and disturbances, as well. In this work, a variable structure adaptive pole placement control (VS-APPC) is proposed. Therefore, new switching laws are proposed, instead of using the traditional integral adaptive laws. Additionally, simulation results for an unstable first order system and simulation and practical results for a three-phase induction motor are shown

Controlador adaptativo por posicionamento de pólos e estrutura variável aplicado ao controle de geradores eólicos baseados em máquina de indução duplamente alimentada

In the last decade, the renewable energy sources have present a major propulsion in the world due to several factors: political, environmental, financial and others. Within this context, we have in particular the energy obtained through wind, wind energy - that has highlighted with rapid growth in recent years, including in Brazil, mostly in the Northeast, due to it s benefit-cost between the clean energies. In this context, we propose to compare the variable structure adaptive pole placement control (VS-APPC) with a traditional control technique proportional integral controller (PI), applied to set the control of machine side in a conversion system using a wind generator based on Double-Fed Induction Generator (DFIG). Robustness and performance tests were carried out to the uncertainties of the internal parameters of the machine and variations of speed reference.

Identificação de modelos polinomiais narx utilizando algoritmos combinados de detecção de estrutura e estimação de parâmetros com aplicações práticas

A modelagem de processos industriais tem auxiliado na produção e minimização de custos, permitindo a previsão dos comportamentos futuros do sistema, supervisão de processos e projeto de controladores. Ao observar os benefícios proporcionados pela modelagem, objetiva-se primeiramente, nesta dissertação, apresentar uma metodologia de identificação de modelos não-lineares com estrutura NARX, a partir da implementação de algoritmos combinados de detecção de estrutura e estimação de parâmetros. Inicialmente, será ressaltada a importância da identificação de sistemas na otimização de processos industriais, especificamente a escolha do modelo para representar adequadamente as dinâmicas do sistema. Em seguida, será apresentada uma breve revisão das etapas que compõem a identificação de sistemas. Na sequência, serão apresentados os métodos fundamentais para detecção de estrutura (Modificado Gram- Schmidt) e estimação de parâmetros (Método dos Mínimos Quadrados e Método dos Mínimos Quadrados Estendido) de modelos. No trabalho será também realizada, através dos algoritmos implementados, a identificação de dois processos industriais distintos representados por uma planta de nível didática, que possibilita o controle de nível e vazão, e uma planta de processamento primário de petróleo simulada, que tem como objetivo representar um tratamento primário do petróleo que ocorre em plataformas petrolíferas. A dissertação é finalizada com uma avaliação dos desempenhos dos modelos obtidos, quando comparados com o sistema. A partir desta avaliação, será possível observar se os modelos identificados são capazes de representar as características estáticas e dinâmicas dos sistemas apresentados nesta dissertação

Desenvolvimento de um sistema para análise de integridade estrutural de equipamentos estáticos baseado em cad vps micro cracks e ultrassom

In general, the designs of equipment takes into account the effects and processes of deterioration it will undergo and arrives at an approximate useful life. However, changes in operational processes and parameters, the action of external agents, the kind of maintenance conducted, the means of monitoring, and natural and accidental occurrences completely modify the desired performance of the equipment. The discontinuities that occur in anisotropic materials often and due to different factors evolve from being subcritical to critical acquiring the status of defect and compromising the physical integrity of the equipment. Increasingly sophisticated technological means of detection, monitoring and assessment of these discontinuities are required to respond ever more rapidly to the requirements of industry. This paper therefore presents a VPS (Virtual Pipe System) computational tool which uses the results of ultrasonic tests on equipment, plotting the discontinuities found in models created in the CAD and CAE systems, and then simulates the behavior of these defects in the structure to give an instantaneous view of the final behavior. This paper also presents an alternative method of conventional ultrasonic testing which correlates the integrity of an overlay (carbon steel and stainless steel attached by welding) and the reflection of ultrasonic waves coming from the interface between the two metals, thus making it possible to identify cracks in the casing and a shift of the overlay

Simulação computacional da interação fluido-estrutura em bombas de cavidades progressivas

The pumping through progressing cavities system has been more and more employed in the petroleum industry. This occurs because of its capacity of elevation of highly viscous oils or fluids with great concentration of sand or other solid particles. A Progressing Cavity Pump (PCP) consists, basically, of a rotor - a metallic device similar to an eccentric screw, and a stator - a steel tube internally covered by a double helix, which may be rigid or deformable/elastomeric. In general, it is submitted to a combination of well pressure with the pressure generated by the pumping process itself. In elastomeric PCPs, this combined effort compresses the stator and generates, or enlarges, the clearance existing between the rotor and the stator, thus reducing the closing effect between their cavities. Such opening of the sealing region produces what is known as fluid slip or slippage, reducing the efficiency of the PCP pumping system. Therefore, this research aims to develop a transient three-dimensional computational model that, based on single-lobe PCP kinematics, is able to simulate the fluid-structure interaction that occurs in the interior of metallic and elastomeric PCPs. The main goal is to evaluate the dynamic characteristics of PCP s efficiency based on detailed and instantaneous information of velocity, pressure and deformation fields in their interior. To reach these goals (development and use of the model), it was also necessary the development of a methodology for generation of dynamic, mobile and deformable, computational meshes representing fluid and structural regions of a PCP. This additional intermediary step has been characterized as the biggest challenge for the elaboration and running of the computational model due to the complex kinematic and critical geometry of this type of pump (different helix angles between rotor and stator as well as large length scale aspect ratios). The processes of dynamic generation of meshes and of simultaneous evaluation of the deformations suffered by the elastomer are fulfilled through subroutines written in Fortan 90 language that dynamically interact with the CFX/ANSYS fluid dynamic software. Since a structural elastic linear model is employed to evaluate elastomer deformations, it is not necessary to use any CAE package for structural analysis. However, an initial proposal for dynamic simulation using hyperelastic models through ANSYS software is also presented in this research. Validation of the results produced with the present methodology (mesh generation, flow simulation in metallic PCPs and simulation of fluid-structure interaction in elastomeric PCPs) is obtained through comparison with experimental results reported by the literature. It is expected that the development and application of such a computational model may provide better details of the dynamics of the flow within metallic and elastomeric PCPs, so that better control systems may be implemented in the artificial elevation area by PCP

Caracterização da Estrutura Espacial do Psilídeo-de-concha Glycaspis brimblecombei (Hemiptera: Psyllidae) e de seu Parasitóide Psyllaephagus bliteus (Hymenoptera: Encyrtidae) em Floresta de Eucalyptus camaldulensis

Os psilídeos são um grupo de pragas amplamente distribuídas, causando perdas consideráveis em sistemas agrícolas e florestais. A recente descoberta da ocorrência do psilídeo-de-concha Glycaspis brimblecombei Moore (Hemiptera: Psyllidae), no Brasil tornou-se um problema de grande importância, por ser específico de Eucalyptus. O parasitóide Psyllaephagus bliteus Riek (Hymenoptera: Encyrtidae) é indicado como principal agente no controle biológico desta praga. Portanto, para o monitoramento adequado de ambas espécies é necessário um método eficiente de amostragem. Devido a falta de informações referentes ao seu comportamento em plantações extensivas de eucalipto, realizou-se este trabalho, para verificar a distribuição espacial de G. brimblecombei e de seu parasitóide P. bliteus, através de armadilhas adesivas em floresta de E. camaldulensis. Para determinação da distribuição espacial foram utilizadas 53 armadilhas amarelas distribuídas uniformemente no talhão, formando uma grade amostral espaçadas de 50 x 36 m, à altura aproximada de 1,80 m do solo. Cada armadilha foi presa com auxílio de um arame plastificado em um fio estendido entre duas árvores em um talhão de 19 ha com E. camaldulensis em Luiz Antônio, SP. Foram realizadas dez avaliações, entre janeiro e junho de 2005. Após a coleta as armadilhas foram levadas ao laboratório para ser feita à triagem e contagem do número total de adultos de G. brimblecombei e de P. bliteus, nas duas faces das armadilhas. Concluiu-se que as populações de ambas espécies apresentaram distribuição espacial agregada, no período avaliado.