Metodologia para mapeamento 3D de dose em um depósito de rejeitos radioativos utilizando MCNP e redes neurais

• Ano de defesa: 2018
• Autor(a) principal: Botelho, Mario Hugo Dias da Silva
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Rio de Janeiro; Brasil; Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia; Programa de Pós-Graduação em Engenharia Nuclear; UFRJ
• Palavras-chave em Português: Resíduos radioativos; MCNP; Redes neurais; CNPQ::ENGENHARIAS::ENGENHARIA NUCLEAR::APLICACOES DE RADIOISOTOPOS::INSTRUMENTACAO PARA MEDIDA E CONTROLE DE RADIACAO
• Link de acesso: http://hdl.handle.net/11422/19311

Among the many applications of nuclear energy a usual problem and one of the most critical is the management of radioactive waste. For the waste management be considered adequate it is essential that the principles of radiation protection, including the optimization principle, be followed. In this research was elaborated a methodology for mapping the radiation dose in a radioactive waste disposal, depending on the threedimensional coordinate in space, with the purpose of optimize the radiation dose received by the disposal operator. The developed methodology join the Monte Carlo NParticle (MCNP) code with the technique of Artificial Neural Networks (RNAs), both widely diffused in nuclear engineering. The methodology was applied in a case study and the results obtained were accurate. Therefore, the combination of MCNP and RNAs proved to be adequate to perform the three-dimensional dose mapping on a radioactive waste disposal.