2383 research outputs found
Sort by
Genetic and neural approaches to nuclear transient identification
No full textThis work presents two approaches for pattern recognition to the same set of reactor signals. The first one
describes a possibilistic approach optimized by genetic algorithm. The use of a possibilistic classification
provides a natural and consistent classification rules, leading naturally to a good heuristic to handle the “don’t
know” response, in case of unrecognized transient, which is fairly desirable in transient classification systems
where safety is critical, since wrong or not reliable classifications can be catastrophic. Application of the
proposed approach to a nuclear transient identification problem reveals good capability of the genetic algorithm
in learning optimized possibilistic classification rules for efficient diagnosis including “don’t know” response.
The second one uses two multilayer neural networks (NN). The first NN is responsible for the dynamic
identification. This NN uses, as input, a short set (in a moving time window) of recent measurements of each
variable avoiding the necessity of using starting events. The second NN is used to validate the instantaneous
identification (from the first net) through the validation of each variable. This net is responsible for allowing the
system to provide a “don’t know” response. In order to validate both methods, a Nuclear Power Plant (NPP) transient identification problem comprising postulated accidents, simulated for a pressurized water reactor, was proposed in the validation process it has been considered noisy data in order to evaluate the method robustness. Obtained results reveal the ability of the methods in dealing with both dynamic identification of transients and correct “don’t know” response
Um modelo axissimétrico para a equação de transferência radiativa bidimensional
No full textA equação de transferência radiativa num meio que se caracteriza por ser axialmente simétrico é aproximada angularmente por um método de ordenadas discretas. A formulação variacional para o sistema de equações diferenciais de 1ª ordem com condições de contorno no fluxo incidente é aproximada por uma malha de polígonos (pixeis) que são obtidos com a utilização da base natural. O sistema de equações diferenciais parciais gerado dessa aproximação é resolvido numericamente pelo método dos elementos finitos descontínuos. Este trabalho investiga a sensibilidade ao fluxo emergente dadas as propriedades de absorção e espalhamento do meio. Algumas simulações envolvendo a influência das propriedades do meio no mapeamento de fluxo incidente–emergente (albedo) serão apresentadas
Produção do radiotraçador gasoso CH3Br para monitoração de gasodutos
No full textO crescente desenvolvimento da indústria de petróleo tem motivado a aplicação de novas tecnologias e metodologias para o estudo da avaliação de unidades industriais. Atualmente, a disponibilidade de uma grande diversidade de métodos de produção de traçadores radioativos proporcionou um avanço no emprego de técnicas nucleares para o monitoramento e a avaliação dessas unidades. O objetivo do trabalho foi estudar a produção de um gás orgânico radioativo para monitorar sistemas com fluxo de material gasoso empregado em diversas indústrias. O gás orgânico brometo de metila, CH3Br, foi escolhido como radiotraçador e sua produção se realizou através da irradiação do sal de brometo de potássio, através do isótopo 82Br, metanol e ácido sulfúrico 9M (catalisador). Um sistema de vidraria foi montado para se produzir o gás CH3Br. A identificação do gás halogenado foi feita, através de uma mistura contendo acetona/NaI, colocada no frasco de coleta. O transporte do brometo de metila foi estudado por dois detectores cintiladores de NaI 2” x 2”, que foram posicionados no frasco de reação e no final do sistema de produção do gás. Para a aquisição de dados, cada detector utilizou um conjunto de módulos eletrônicos tradicionalmente empregados para a aquisição e o processamento do sinal gerado em detectores cintiladores
A hybrid niched-island genetic algorithm applied to a nuclear core optimization problem
No full textDiversity maintenance is a key-feature in most genetic-based optimization processes. The quest for such
characteristic, has been motivating improvements in the original genetic algorithm (GA). The use of multiple
populations (called islands) has demonstrating to increase diversity, delaying the genetic drift. Island Genetic
Algorithms (IGA) lead to better results, however, the drift is only delayed, but not avoided. An important
advantage of this approach is the simplicity and efficiency for parallel processing. Diversity can also be
improved by the use of niching techniques. Niched Genetic Algorithms (NGA) are able to avoid the genetic
drift, by containing evolution in niches of a single-population GA, however computational cost is increased. In
this work it is investigated the use of a hybrid Niched-Island Genetic Algorithm (NIGA) in a nuclear core
optimization problem found in literature. Computational experiments demonstrate that it is possible to take
advantage of both, performance enhancement due to the parallelism and drift avoidance due to the use of niches. Comparative results shown that the proposed NIGA demonstrated to be more efficient and robust than an IGA and a NGA for solving the proposed optimization problem
Preparação e caracterização de membranas poliméricas sulfonadas e avaliação do potencial de utilização como membranas ion-seletivas
No full textEste trabalho visa a preparação e caracterização de membranas planas poliméricas de ultrafiltração com alta permeabilidade e seletividade a cátions para serem aplicadas no tratamento de rejeitos radioativos. Para obtenção de membranas de ultrafiltração é necessário obter características morfológicas específicas, tais como estrutura assimétrica e distribuição estreita de tamanho de poros na superfície. Estas características são determinadas pelas condições de preparação das membranas e para isto foi utilizado o método de inversão de fase por imersão/precipitação que permite obter membranas com características adequadas para cada aplicação. Para este estudo foram preparadas membranas de ultrafiltração a partir de soluções poliméricas contendo poli (éter-sulfona) (PES), polisulfona sulfonada (PSS) e polivinilpirrolidona (PVP) solubilizadas em N,N-dimetilacetamida. A polisulfona sulfonada foi adicionada à composição das membranas com o objetivo de conferir o caráter protônico às membranas. Neste trabalho os seguintes parâmetros foram investigados: presença ou não da PSS, tempo de exposição e composição do banho de imersão (água pura e soluções de N,N-dimetilacetamida/água). As membranas foram caracterizadas por Microscopia Eletrônica de Varredura (MEV) e experimentos de permeação. Para comparação foi utilizada membrana comercial de ultrafiltração à base de poli(éter-sulfona), 30 K, da Milipore
Nuclear power plant transient identification using a neuro-fuzzy inference system
No full textIn this work, a approach for the identification of transients in presented, aiming at helping the operator to make a decision relative to the procedure to be followed in situations of accidents/transients at nuclear power plants. In this way, a diagnostic strategy based on hierarchical use artificial neural networks (ANN) for a first level transient diagnose. After the ANN has done a preliminary transient type identification, a fuzzy-logic system analyzes the results emiting reliability degree of it. In order to validate the method, a Nuclear Power Plant transient identification problem, comprising postulated accidents, in proposed. Noisy data was used to evaluate the method robustness. The results obtained reveal the ability of the method in dealing with dynamic identification of transients and its reliability degree
Implementation of the immersive virtual reality laboratory in nuclear engineering institute
No full textThe Immersive Virtual Reality Laboratory (IVRL) under development in Human System Interface Laboratory
(HSIL) [1] constitute a powerful general-purpose facility for experimental and computational work on human
perception and perceptually guided action. Virtual reality (VR) or virtual environment(VE) [2] are computergenerated environments with and within people can interact. The advantage of VR is that people can be immersed by the simulated environment, which would sometimes be unavailable due to cost, safety, or
perceptual restrictions in the real environment. There are many applications of virtual reality on the nuclear area [3][4]. Training is one of the most common of them. A significant advantage of a virtual training environment over a real one is it’s enormous flexibility. A virtual environment can be used as the basis for training in any number of different scenarios, so that trainees can learn to cope with many different situations, some of which may be impossible to prepare for any other way. Another advantage of using virtual environments for training purposes is that trainees learn by actively performing actions. This has a significant effect on their ability to retain what they learn, and is clearly superior to passive training techniques, such as videos and books, for training where spatial understanding is important. This kind of Laboratory is the first in Brazilian nuclear area. A safe virtual environment can be used to simulate a real environment that is either too dangerous, complex, or expensive to training. Virtual environments can therefore be used to increase safety standards, improve efficiency, and reduce overall training costs
A methodology for evaluation and licensing of nuclear power plant control rooms
No full textIn the safe operation of nuclear power plants the performance of the control room crews plays an important role. In this respect, a well-designed control room and human-system interfaces (HSI) are crucial for safe and
efficient operation of the plant. It is therefore essential that the design, development and evaluation of nuclear
control rooms are conducted in a well-structured way, applying human factors principles, ergonomics
requirements and guidelines in all phases. The aim of this paper is to propose a new methodology to evaluate
nuclear power plant control rooms that includes the ergonomic approach, based on the operator activity
analysis, together with human factors standards and guidelines already used. We describe a case study in which this methodology has been applied in a nuclear power plant control room. We show that the operator activity analysis uncovers a series of important safety related features in control room operation that are not detected by using the traditional methodology, based only on the human factors standards and human factors guidelines. The information gathered through this methodology make possible the generation of a series of
recommendations for the adequacy of the control room to the legal requirements of the regulating agency,
assisting the licensing process
