Análise de desempenho da rede neural artificial do tipo multilayer perceptron na era multicore

Artificial neural networks are usually applied to solve complex problems. In problems with more complexity, by increasing the number of layers and neurons, it is possible to achieve greater functional efficiency. Nevertheless, this leads to a greater computational effort. The response time is an important factor in the decision to use neural networks in some systems. Many argue that the computational cost is higher in the training period. However, this phase is held only once. Once the network trained, it is necessary to use the existing computational resources efficiently. In the multicore era, the problem boils down to efficient use of all available processing cores. However, it is necessary to consider the overhead of parallel computing. In this sense, this paper proposes a modular structure that proved to be more suitable for parallel implementations. It is proposed to parallelize the feedforward process of an RNA-type MLP, implemented with OpenMP on a shared memory computer architecture. The research consistes on testing and analizing execution times. Speedup, efficiency and parallel scalability are analyzed. In the proposed approach, by reducing the number of connections between remote neurons, the response time of the network decreases and, consequently, so does the total execution time. The time required for communication and synchronization is directly linked to the number of remote neurons in the network, and so it is necessary to investigate which one is the best distribution of remote connections

Sistema de teleoftalmologia para auxílio ao pré-diagnóstico de disfunções oculomotoras

The vision is one of the five senses of the human body and, in children is responsible for up to 80% of the perception of world around. Studies show that 50% of children with multiple disabilities have some visual impairment, and 4% of all children are diagnosed with strabismus. The strabismus is an eye disability associated with handling capacity of the eye, defined as any deviation from perfect ocular alignment. Besides of aesthetic aspect, the child may report blurred or double vision . Ophthalmological cases not diagnosed correctly are reasons for many school abandonments. The Ministry of Education of Brazil points to the visually impaired as a challenge to the educators of children, particularly in literacy process. The traditional eye examination for diagnosis of strabismus can be accomplished by inducing the eye movements through the doctor s instructions to the patient. This procedure can be played through the computer aided analysis of images captured on video. This paper presents a proposal for distributed system to assist health professionals in remote diagnosis of visual impairment associated with motor abilities of the eye, such as strabismus. It is hoped through this proposal to contribute improving the rates of school learning for children, allowing better diagnosis and, consequently, the student accompaniment

Classificação de padrões através de um comitê de máquinas aprimorado por aprendizagem por reforço

Reinforcement learning is a machine learning technique that, although finding a large number of applications, maybe is yet to reach its full potential. One of the inadequately tested possibilities is the use of reinforcement learning in combination with other methods for the solution of pattern classification problems. It is well documented in the literature the problems that support vector machine ensembles face in terms of generalization capacity. Algorithms such as Adaboost do not deal appropriately with the imbalances that arise in those situations. Several alternatives have been proposed, with varying degrees of success. This dissertation presents a new approach to building committees of support vector machines. The presented algorithm combines Adaboost algorithm with a layer of reinforcement learning to adjust committee parameters in order to avoid that imbalances on the committee components affect the generalization performance of the final hypothesis. Comparisons were made with ensembles using and not using the reinforcement learning layer, testing benchmark data sets widely known in area of pattern classification

Desenvolvimento de software para simulação de motores com dispositivos de partida baseada na integração do atp com o toprede

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Roteamento em Redes de Sensores Sem Fios Com Base Em Aprendizagem Por Reforço

The use of wireless sensor and actuator networks in industry has been increasing past few years, bringing multiple benefits compared to wired systems, like network flexibility and manageability. Such networks consists of a possibly large number of small and autonomous sensor and actuator devices with wireless communication capabilities. The data collected by sensors are sent directly or through intermediary nodes along the network to a base station called sink node. The data routing in this environment is an essential matter since it is strictly bounded to the energy efficiency, thus the network lifetime. This work investigates the application of a routing technique based on Reinforcement Learning s Q-Learning algorithm to a wireless sensor network by using an NS-2 simulated environment. Several metrics like energy consumption, data packet delivery rates and delays are used to validate de proposal comparing it with another solutions existing in the literature

Classificação Automática de Modulação Digital com uso de Correntropia para Ambientes de Rádio Cognitivo

Modern wireless systems employ adaptive techniques to provide high throughput while observing desired coverage, Quality of Service (QoS) and capacity. An alternative to further enhance data rate is to apply cognitive radio concepts, where a system is able to exploit unused spectrum on existing licensed bands by sensing the spectrum and opportunistically access unused portions. Techniques like Automatic Modulation Classification (AMC) could help or be vital for such scenarios. Usually, AMC implementations rely on some form of signal pre-processing, which may introduce a high computational cost or make assumptions about the received signal which may not hold (e.g. Gaussianity of noise). This work proposes a new method to perform AMC which uses a similarity measure from the Information Theoretic Learning (ITL) framework, known as correntropy coefficient. It is capable of extracting similarity measurements over a pair of random processes using higher order statistics, yielding in better similarity estimations than by using e.g. correlation coefficient. Experiments carried out by means of computer simulation show that the technique proposed in this paper presents a high rate success in classification of digital modulation, even in the presence of additive white gaussian noise (AWGN)

Projeto de um Transformador utilizado em uma Planta de Plasma

This work discusses the design of a transformer used in a plant plasma. This plant, which is being developed in UFRN, will be used in the treatment of waste. It consists basically of a radio frequency power supply and a inductive plasma torch. The transformer operates at the nominal frequency of 400 kHz, with 50 kW, allowing the adaptation of impedance between the power supply and torch. To develop the project, a study was done on the fabrication technologies and physical effects on the frequency of operation. This was followed by the modeling of this transformer. Finally, simulations and tests were conducted to validate the design

Extração e Representação de Conhecimento de Séries Temporais de Demanda de Energia Elétrica Usando TSKR

The opening of the Brazilian market of electricity and competitiveness between companies in the energy sector make the search for useful information and tools that will assist in decision making activities, increase by the concessionaires. An important source of knowledge for these utilities is the time series of energy demand. The identification of behavior patterns and description of events become important for the planning execution, seeking improvements in service quality and financial benefits. This dissertation presents a methodology based on mining and representation tools of time series, in order to extract knowledge that relate series of electricity demand in various substations connected of a electric utility. The method exploits the relationship of duration, coincidence and partial order of events in multi-dimensionals time series. To represent the knowledge is used the language proposed by Mörchen (2005) called Time Series Knowledge Representation (TSKR). We conducted a case study using time series of energy demand of 8 substations interconnected by a ring system, which feeds the metropolitan area of Goiânia-GO, provided by CELG (Companhia Energética de Goiás), responsible for the service of power distribution in the state of Goiás (Brazil). Using the proposed methodology were extracted three levels of knowledge that describe the behavior of the system studied, representing clearly the system dynamics, becoming a tool to assist planning activities

Identificação Visual de Caixas de Medicamentos Usando Features Correspondentes

This work uses computer vision algorithms related to features in the identification of medicine boxes for the visually impaired. The system is for people who have a disease that compromises his vision, hindering the identification of the correct medicine to be ingested. We use the camera, available in several popular devices such as computers, televisions and phones, to identify the box of the correct medicine and audio through the image, showing the poor information about the medication, such: as the dosage, indication and contraindications of the medication. We utilize a model of object detection using algorithms to identify the features in the boxes of drugs and playing the audio at the time of detection of feauteres in those boxes. Experiments carried out with 15 people show that where 93 % think that the system is useful and very helpful in identifying drugs for boxes. So, it is necessary to make use of this technology to help several people with visual impairments to take the right medicine, at the time indicated in advance by the physician

Viabilidade da Utilização de Máquinas de Indução Convencionais como Motores sem Mancais Mecânicos

Electrical Motors transform electrical energy into mechanic energy in a relatively easy way. In some specific applications, there is a need for electrical motors to function with noncontaminated fluids, in high speed systems, under inhospitable conditions, or yet, in local of difficult access and considerable depth. In these cases, the motors with mechanical bearings are not adequate as their wear give rise to maintenance. A possible solution for these problems stems from two different alternatives: motors with magnetic bearings, that increase the length of the machine (not convenient), and the bearingless motors that aggregate compactness. Induction motors have been used more and more in research, as they confer more robustness to bearingless motors compared to other types of machines building with others motors. The research that has already been carried out with bearingless induction motors utilized prototypes that had their structures of stator/rotor modified, that differ most of the times from the conventional induction motors. The goal of this work is to study the viability of the use of conventional induction Motors for the beringless motors applications, pointing out the types of Motors of this category that can be more useful. The study uses the Finite Elements Method (FEM). As a means of validation, a conventional induction motor with squirrel-cage rotor was successfully used for the beringless motor application of the divided winding type, confirming the proposed thesis. The controlling system was implemented in a Digital Signal Processor (DSP)