Use este identificador para citar ou linkar para este item: http://carpedien.ien.gov.br:8080/handle/ien/2502
Tipo: conferenceObject
Título: Genetic and neural approaches to nuclear transient identification
Autor(es): Almeida, Jose Carlos Soares de
Mol, Antonio Carlos de Abreu
Pereira, Claudio Marcio do Nascimento Abreu
Lapa, Celso Marcelo Franklin
Resumo: This 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.
Palavras-chave: Redes neurais
Algoritmos genéticos
Transientes nucleares
Idioma: por
País: Brasil
Editor: Instituto de Engenharia Nuclear
Sigla da Instituição: IEN
Tipo de Acesso: openAccess
URI: http://carpedien.ien.gov.br:8080/handle/ien/2502
Data do documento: Ago-2005
Aparece nas coleções:Desenvolvimento de Tecnologia para Sistemas Complexos - Trabalhos de Congresso

Arquivos associados a este item:
Arquivo Descrição TamanhoFormato 
GENETIC AND NEURAL APPROACHES TO NUCLEAR TRANSIENT IDENTIFICATION.pdf124,88 kBAdobe PDFVisualizar/Abrir


Os itens no repositório estão protegidos por copyright, com todos os direitos reservados, salvo quando é indicado o contrário.