Em um reator nuclear PWR a temperatura do refrigerante do circuito primário e a da água de realimentação são medidas usando RTD (Resistance Temperature Detectors), ou termômetros de resistência. Estes RTDs alimentam os sistemas de controle e segurança da usina e devem, portanto, ser muito precisos e ter bom desempenho dinâmico. O tempo de resposta dos RTDs é caracterizado por um parâmetro denominado de Constante de Tempo, definido como sendo o tempo que o sensor leva para atingir 63,2% do seu valor final após sofrer uma variação de temperatura em forma de degrau. Este valor é determinado em laboratório, porém as condições de operação de reatores nucleares são difíceis de ser reproduzidas. O método LCSR (Loop Current Step Response), ou teste de resposta a um degrau de corrente, foi desenvolvido para medir remotamente o tempo de resposta dos RTDs. A partir desse teste, a constante de tempo do sensor é calculada através de uma transformação LCSR que envolve a determinação das constantes modais do modelo de transferência de calor. Este cálculo não é simples e requer pessoal especializado. Por este motivo, utilizou-se a metodologia de Redes Neurais Artificiais para estimar a constante de tempo do RTD a partir do LCSR. Os testes LCSR foram usados como dados de entrada da RNA; os testes de Imersão Rápida foram usados para determinar a constante de tempo dos sensores, sendo estes os valores desejados de saída da rede. Esta metodologia foi aplicada inicialmente a dados teóricos, simulando dez sensores com diferentes valores de constante de tempo, resultando em um erro médio de aproximadamente 0,74 %. Dados experimentais de 3 diferentes RTDs foram usados para estimar a constante de tempo, resultando em um erro máximo de 3,34 %. Os valores de constante de tempo estimados pelas RNAs foram comparados com aqueles obtidos pelo método tradicional, obtendo-se um erro médio de 18 % o que mostra que as RNAs são capazes de estimar a constante de tempo de uma forma precisa. / In a PWR nuclear power plant, the primary coolant temperature and feedwater temperature are measured using RTDs (Resistance Temperature Detectors). These RTDs typically feed the plants control and safety systems and must, therefore, be very accurate and have good dynamic performance. The response time of RTDs is characterized by a single parameter called the Plunge Time Constant defined as the time it takes the sensor output to achieve 63.2 percent of its final value after a step change in temperature. Nuclear reactor service conditions are difficult to reproduce in the laboratory, and an in-situ test method called LCSR (Loop Current Step Response) test was developed to measure remotely the response time of RTDs. From this test, the time constant of the sensor is identified by means of the LCSR transformation that involves the dynamic response modal time constants determination using a nodal heat-transfer model. This calculation is not simple and requires specialized personnel. For this reason an Artificial Neural Network has been developed to predict the time constant of RTD from LCSR test transient. It eliminates the transformations involved in the LCSR application. A series of LCSR tests on RTDs generates the response transients of the sensors, the input data of the networks. Plunge tests are used to determine the time constants of the RTDs, the desired output of the ANN, trained using these sets of input/output data. This methodology was firstly applied to theoretical data simulating 10 RTDs with different time constant values, resulting in an average error of about 0.74 %. Experimental data from three different RTDs was used to predict time constant resulting in a maximum error of 3,34 %. The time constants values predicted from ANN were compared with those obtained from traditional way resulting in an average error of about 18 % and that shows the network is able to predict accurately the sensor time constant.

国立情報学研究所で電子化したコンテンツを使用している。

Em um reator nuclear PWR a temperatura do refrigerante do circuito primário e a da água de realimentação são medidas usando RTD (Resistance Temperature Detectors), ou termômetros de resistência. Estes RTDs alimentam os sistemas de controle e segurança da usina e devem, portanto, ser muito precisos e ter bom desempenho dinâmico. O tempo de resposta dos RTDs é caracterizado por um parâmetro denominado de Constante de Tempo, definido como sendo o tempo que o sensor leva para atingir 63,2% do seu valor final após sofrer uma variação de temperatura em forma de degrau. Este valor é determinado em laboratório, porém as condições de operação de reatores nucleares são difíceis de ser reproduzidas. O método LCSR (Loop Current Step Response), ou teste de resposta a um degrau de corrente, foi desenvolvido para medir remotamente o tempo de resposta dos RTDs. A partir desse teste, a constante de tempo do sensor é calculada através de uma transformação LCSR que envolve a determinação das constantes modais do modelo de transferência de calor. Este cálculo não é simples e requer pessoal especializado. Por este motivo, utilizou-se a metodologia de Redes Neurais Artificiais para estimar a constante de tempo do RTD a partir do LCSR. Os testes LCSR foram usados como dados de entrada da RNA; os testes de Imersão Rápida foram usados para determinar a constante de tempo dos sensores, sendo estes os valores desejados de saída da rede. Esta metodologia foi aplicada inicialmente a dados teóricos, simulando dez sensores com diferentes valores de constante de tempo, resultando em um erro médio de aproximadamente 0,74 %. Dados experimentais de 3 diferentes RTDs foram usados para estimar a constante de tempo, resultando em um erro máximo de 3,34 %. Os valores de constante de tempo estimados pelas RNAs foram comparados com aqueles obtidos pelo método tradicional, obtendo-se um erro médio de 18 % o que mostra que as RNAs são capazes de estimar a constante de tempo de uma forma precisa. / In a PWR nuclear power plant, the primary coolant temperature and feedwater temperature are measured using RTDs (Resistance Temperature Detectors). These RTDs typically feed the plants control and safety systems and must, therefore, be very accurate and have good dynamic performance. The response time of RTDs is characterized by a single parameter called the Plunge Time Constant defined as the time it takes the sensor output to achieve 63.2 percent of its final value after a step change in temperature. Nuclear reactor service conditions are difficult to reproduce in the laboratory, and an in-situ test method called LCSR (Loop Current Step Response) test was developed to measure remotely the response time of RTDs. From this test, the time constant of the sensor is identified by means of the LCSR transformation that involves the dynamic response modal time constants determination using a nodal heat-transfer model. This calculation is not simple and requires specialized personnel. For this reason an Artificial Neural Network has been developed to predict the time constant of RTD from LCSR test transient. It eliminates the transformations involved in the LCSR application. A series of LCSR tests on RTDs generates the response transients of the sensors, the input data of the networks. Plunge tests are used to determine the time constants of the RTDs, the desired output of the ANN, trained using these sets of input/output data. This methodology was firstly applied to theoretical data simulating 10 RTDs with different time constant values, resulting in an average error of about 0.74 %. Experimental data from three different RTDs was used to predict time constant resulting in a maximum error of 3,34 %. The time constants values predicted from ANN were compared with those obtained from traditional way resulting in an average error of about 18 % and that shows the network is able to predict accurately the sensor time constant.

This diploma thesis deals with development of a new generation of magnetic circuits of fast magnetoreological (MR) dampers that will not use problematic ferrite materials. To ensure a short response time, eddy currents must be avoided. Using the Finite Element Method, 2 solutions were proposed - the use of soft magnetic composites with high electrical resistance and the cutting of grooves in metallic material. Subsequently, 2 magnetic circuits were produced - one from Sintex SMC prototyping material composite and one from 11SMn30 steel with 48 deep grooves 0,35 mm wide. Both were implemented into the MR damper and tested on a hydraulic pulsator. The results of the experiments were compared with 2 different MR dampers: one with 11SMn30 magnetic circuit without grooves and another with the Epcos N87 ferrite magnetic core. It has been found that both new circuits have ensured an equally short response of the damper force to the change of electric current like ferrite. At the same time, a much larger dynamic range was ensured. Compared to 11SMn30 without grooves, the new circuits provided approximately 7x faster response.

The diploma thesis deals with the development of the magnetic circuit of Magnetoreological (MR) dampers with a short time response. To achieve a short response time, a shape approach was chosen whereby the geometry of the magnetic circuit was chosen to significantly eliminate the occurrence of eddy currents. The influence of structures on magnetic properties was first examined on a simpler toroidal core and then the optimization was subjected to the magnetic circuit of the MR damper itself. Geometry optimization was done using FEM simulations. The resulting geometry was made of pure iron using Selective Laser Melting technology (SLM). In addition, a MR damper was completed and its properties on air and with MR fluid were measured, which were then compared with previously developed rapid MR dampers. It has been found that the newly designed magnetic circuit achieves similar time responses as all other compared fast MR dampers and reaches a higher dynamic range than most of the compared variants. The proposed magnetic circuit also has a significantly reduced weight.

This diploma thesis deals with an engineering design of a fail-safe magnetorheological (MR) damper capable of semi-active control. The first part of the thesis is devoted to the current state of knowledge of fail-safe MR dampers and permanent magnets contained in these dampers. The next part contains an engineering design of a fail-safe MR damper, made using FEM simulations, and its subsequent testing in terms of magnetic and hydraulic properties. Finally, a design of a fail-safe MR damper with fast response time was made and simulated using verified FEM analysis. Fast response time is ensured by limiting the generation of eddy currents in the piston core by grooving.

The diploma thesis deals with testing of sports car suspension. The aim is the testing of fast magnetorheological damper in semiactive suspension that corresponds to quarter car model. The fast magnetorheological damper has a response time 2 ms. If the response time is short enough (order of units miliseconds) it is possible (based on earlier simulations) to achieve improvement of driving safety (better stability of force of tyre on roadway) and comfort (reduction of vibrations). The thesis proves this statement with series of experiments.

This thesis explored fundamental concepts of linear optics focusing on the modal interaction within waveguide/microresonator systems. In addition, it investigated a nonlinear process of stimulated degenerate four-wave mixing in a channel waveguide exploiting the analogy between photons and cold boson atoms. The backscattering phenomenon due to the surface wall roughness of a microresonator is addressed by adding to the usual conservative (Hermitian) coupling coefficient, a dissipative (non-Hermitian) term. This allows explaining the experimental measurements of a multimodal microresonator, which exhibits an asymmetrical resonance splitting characterized by a difference in the peak depths of the transmission spectra. It is shown theoretically, numerically and experimentally that the stochastic nature of the roughness along with the inter-modal dissipative coupling could give rise to a different exchange of energy between the co-propagating and the counter-propagating mode. The unbalanced exchange of energy between the two modes with opposite angular momenta can generate a different reflection by swapping the injection of the light between the input and the output ports. This effect lies at the heart of the realization of an unidirectional reflection device and it finds an explanation in the physics of the exceptional points. The realization of an optical setup based on a Mach-Zehnder interferometer, which exploits some particular techniques of data acquisition, allows obtaining a full knowledge of the complex electric field of a propagating mode. In this way, the spectrum of a wedge microresonator vertically coupled to a bus waveguide is explained using analysis methods based on parametric phasors and inverse complex representations. In addition, the energy exchange between the co-propagating and counter-propagating modes is studied from a temporal point of view by extrapolating a simple model based on the Green function. In particular, it is discussed the analytical temporal response of a microring resonator excited through a bus waveguide by an optical rectangular pulse. Here, it is shown theoretically and experimentally, how the temporal response leads to the characterization of the coupling regime simply from the knowledge of the electric field intensity. In this thesis, the isomorphism between the Schroedinger’s equation and the Helmholtz wave equation is analyzed in the nonlinear case. Considering a bulk nonlinear medium of the Kerr type, the complex amplitude of the optical field is a slowly varying function of space and time, which satisfies a nonlinear Schroedinger equation. The well-known nonlinear optical phenomenon of stimulated degenerate four wave mixing is reformulated in the language of the Bogoliubov theory. This parallelism between photons and cold atoms allows showing that the phase of the signal assumes a peculiar sound-like dispersion under proper assumptions.

This work deals with the usage of magneto-rheological (MR) damper in the semi-active car suspension. Semi-active suspension can improve ride comfort or tyre grip to the level, which cannot be achieved with the common passive setting of the damper. MR damper has however features, like time response of the controller with MR damper and the control range of the MR damper, which limit area of application. It was found out that especially the time response of the damper significantly influences the efficiency of semi-active algorithms. Current MR dampers with controllers are not capable of efficient control of the semi-active suspension. For proper design of semi-active suspension with MR dampers, the time response must be decreased. Therefore, a new PWM current controller was designed and verified. Also changes in MR damper design which eliminate eddy-currents in the magnetic circuit were proposed. The results of this work should contribute to the better design of semi-active suspension systems with MR damper.

Submitted by Emanuel Varela Cardoso (emanuel.varela@ufcg.edu.br) on 2018-05-04T19:00:50Z No. of bitstreams: 1 ALANA KELLY XAVIER SANTOS CAMPOS – TESE (PPGEE) 2016.pdf: 5517354 bytes, checksum: 24306600f8b9a28a2d230c48b0811a3a (MD5) / Made available in DSpace on 2018-05-04T19:00:50Z (GMT). No. of bitstreams: 1 ALANA KELLY XAVIER SANTOS CAMPOS – TESE (PPGEE) 2016.pdf: 5517354 bytes, checksum: 24306600f8b9a28a2d230c48b0811a3a (MD5) Previous issue date: 2016-10-21 / Capes / Um método de estimação fasorial é proposto para ser utilizado em relés digitais de proteção. Este método tem o objetivo de melhorar a característica de rejeição de componentes indesejáveis do sinal, que são introduzidas durante faltas e manobras nos sistemas elétricos de potência e podem comprometer o desempenho dos sistemas de proteção. São realizadas comparações entre o método proposto e outros quatro métodos de estimação fasorial, em relação às respostas no tempo e frequência. Para as comparações de resposta no tempo são utilizados sinais analíticos provenientes do MATrix LABoratory (MATLAB®) e sinais simulados provenientes do Alternative Transients Program (ATP). Um relé de distância foi implementado utilizando a linguagem MODELS do ATP para avaliar o desempenho do relé, utilizando o método proposto e demais métodos avaliados. A partir de alguns parâmetros usados para avaliar quantitativamente os métodos de estimação, observou-se que o método proposto apresenta bons resultados em termos de precisão e velocidade de convergência, com resposta no tempo compatível com os demais métodos comumente utilizados. A resposta em frequência do método proposto se apresenta mais seletiva em relação aos demais, por apresentar menores lóbulos entre as harmônicas e uma banda de passagem mais estreita em torno da frequência fundamental. Outra vantagem do método proposto é que ele é de fácil programação e permite que os usuários alterem parâmetros de projeto de acordo com suas necessidades. / A phasor estimation method is proposed to be used in digital relays for power system protection. It aims to improve the characteristic of filtering out undesirable frequency components of signals, which are introduced during faults and switching events in electric power systems that may compromise the performance of the protection systems. Comparisons are made in relation to time response and frequency response for the proposed method and four phasor estimation methods. In relation to response time are used analytical signals from MATrix LABoratory (MATLAB®) and simulated signals from Alternative Transients Program (ATP). A distance relay was implemented by using the MODELS language in ATP to evaluate the performance of the relay. From some parameters used to evaluate the estimation methods, it was observed that the proposed method shows good results in terms of accuracy and convergence speed, with response time compatible with other methods. The frequency response of the proposed method appears to be more selective than those of the others methods because it presents lower lobes between the harmonics and a narrower pass band around the fundamental frequency. Another advantage of the proposed method is that it can be easily programmed, allowing users to change design parameters according to their needs.