Método experimental para determinação das capacitancias parasitas do motor de indução trifásico acionado por inversor MLP

A proposta deste trabalho é apresentar uma nova metodologia para determinação experimental das capacitancias parasitas do motor de indução trifásico de rotor em gaiola. As capacitancias parasitas fazem parte do circuito equivalente do motor para estudos de interferência eletromagnética causada no motor de indução em modo comum quando ele for acionado por inversor controlado por modulação por largura de pulsos (MLP). Os procedimentos propostos para o desenvolvimento deste novo método consistem em: a) determinação dos parâmetros do circuito equivalente do motor de indução trifásico, em regime permanente, através de ensaio em laboratório; b) estabelecer configurações de ligações entre o inversor MLP e o motor para medições das grandezas de interesse que são as seguintes: tensões de modo comum e de eixo, correntes de fuga e de eixo, através de circuito de medição desenvolvido para este fim; c) calcular os valores das capacitancias parasitas entre estator e carcaça do motor; estator e rotor; rotor e carcaça e de rolamento utilizando a expressão matemática da definição de capacitancia; d) utilizar o software Pspice para simular o sistema motor de indução trifásico, alimentado por inversor MLP, com os circuitos equivalentes em baixas e altas frequências; e) obter as formas de onda características do fenômeno de modo comum.

Avaliação estatística de métodos de poda aplicados em neurônios intermediários da rede neural MLP

There are several papers on pruning methods in the artificial neural networks area. However, with rare exceptions, none of them presents an appropriate statistical evaluation of such methods. In this article, we proved statistically the ability of some methods to reduce the number of neurons of the hidden layer of a multilayer perceptron neural network (MLP), and to maintain the same landing of classification error of the initial net. They are evaluated seven pruning methods. The experimental investigation was accomplished on five groups of generated data and in two groups of real data. Three variables were accompanied in the study: apparent classification error rate in the test group (REA); number of hidden neurons, obtained after the application of the pruning method; and number of training/retraining epochs, to evaluate the computational effort. The non-parametric Friedman's test was used to do the statistical analysis.

Inversor CS Boost monofásico em aplicações com fontes renováveis

This paper presents a new methodology for the operation and control of a single-phase current-source (CS) Boost Inverter, considering that the conventional current-source inverter (CSI) has a right-half-plane (RHP) zero in its control-to-output transfer function, and this RHP zero causes the known non-minimum-phase effects. In this context, a special design with low boost inductance and a multi-loop control is developed in order to assure stable and very fast dynamics. Furthermore, the Inverter presents output voltage with very low total harmonic distortion (THD), reduced components and high power density. Therefore, this paper presents the inverter operation, the proposed control technique, and main simulation and experimental results in order to demonstrate the feasibility of the proposal. © 2010 IEEE.

Application of MLP and Kohonen networks for recognition of wear patterns of single-point dressers

Grinding is a parts finishing process for advanced products and surfaces. However, continuous friction between the workpiece and the grinding wheel causes the latter to lose its sharpness, thus impairing the grinding results. This is when the dressing process is required, which consists of sharpening the worn grains of the grinding wheel. The dressing conditions strongly affect the performance of the grinding operation; hence, monitoring them throughout the process can increase its efficiency. The objective of this study was to estimate the wear of a single-point dresser using intelligent systems whose inputs were obtained by the digital processing of acoustic emission signals. Two intelligent systems, the multilayer perceptron and the Kohonen neural network, were compared in terms of their classifying ability. The harmonic content of the acoustic emission signal was found to be influenced by the condition of dresser, and when used to feed the neural networks it is possible to classify the condition of the tool under study.

Desenvolvimento e implementação de um sistema de controle de posição e velocidade de uma esteira transportadora usando inversor de frequência e microcontrolador

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Uma nova metodologia de projeto e controle para o inversor Boost (CSI) monofásico, para o aproveitamento de fontes alternativas e renováveis de energia elétrica

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Inversor Buck-Boost integrado para aplicações com micro-geradores eólicos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

O uso de rede neural artificial MLP na predição de estruturas secundárias de proteínas

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Aplicação do inversor de frequência para acionamento de sistemas de bombeamento em irrigação por aspersão

Pós-graduação em Agronomia (Irrigação e Drenagem) - FCA

Controle da injeção de potências ativa e reativa em inversor de Geração Distribuída conectado à rede de distribuição em corrente alternada em baixa tensão, empregando LMIs com realimentação de estados e critérios de D-estabilidade

Pós-graduação em Engenharia Elétrica - FEIS

Aplicação da rede neural MLP (Multilayer Perceptron) em indústria de pisos e revestimentos do Pólo Cerâmico de Santa Gertrudes - SP

Pós-graduação em Geociências e Meio Ambiente - IGCE

Uso de inversor de frequência em sistema de irrigação do tipo pivô central e seu efeito na lâmina e uniformidade de distribuição de água

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Aplicação de inversor de freqüência para economia de energia elétrica, em sistema de irrigação por aspersão

Pós-graduação em Agronomia (Irrigação e Drenagem) - FCA

Pruning methods to MLP neural networks considering proportional apparent error rate for classification problems with unbalanced data

This article deals with classification problems involving unequal probabilities in each class and discusses metrics to systems that use multilayer perceptrons neural networks (MLP) for the task of classifying new patterns. In addition we propose three new pruning methods that were compared to other seven existing methods in the literature for MLP networks. All pruning algorithms presented in this paper have been modified by the authors to do pruning of neurons, in order to produce fully connected MLP networks but being small in its intermediary layer. Experiments were carried out involving the E. coli unbalanced classification problem and ten pruning methods. The proposed methods had obtained good results, actually, better results than another pruning methods previously defined at the MLP neural network area. (C) 2014 Elsevier Ltd. All rights reserved.

Órtese de tornozelo: efeito sobre a razão de pico torque muscular eversor excêntrico/inversor concêntrico do tornozelo durante simulação do basquetebol

The ankle sprains represent the most common injuries in sports and basketball. In this sense, the use of ankle bracing and strength capacity analysis of the ankle evertor and invertor muscles, have been suggested as preventive measures and important tools for identifying risk factors associated with ankle sprains. However, questions still persist as to effect of the use ankle bracing on biomechanical variables related to the stability of the ankle. For this reason, this study aims to analyze the effect of the use of ankle bracing on peak torque (PT) of ankle evertor and invertor muscles and on eccentric evertor/concentric invertor torque ratio (EVEECC/INVCON), during the basketball match-play simulation. Ten healthy college basketball players, without mechanics or functional ankle instability performed a laboratory-based protocol representative of work rates observed during basketball match-play, in two different situations, with and without use of ankle bracing. The test was composed of a succession of intermittent physical effort equally distributed in four periods of 10 minutes each, considering the mechanical and physiological demands of a basketball match-play. Prior to the start of the trial (Evaluation 1) and after 2° (Evaluation 2) and 4° (Evaluation 3) periods, the subjects performed five maximal isokinetic concentric and eccentric contractions of ankle invertor and evertor muscles, separated by two minutes rest, at 60 °/s and 120 °/s. After testing for normality of data distribution with the Shapiro-Wilk test, was used the ANOVA repeated measures for two factors and post-hoc Bonferroni test for comparison of variables between assessments. Was adopted p < 0.05. There was no significant difference for PT and EVEECC/INVCON torque ratio between assessments. There was a decrease in PT EVEECC at 60º/s and 120º/s for the ...(Complete abstract click electronic access below)