Published 1997 | Version v1
Miscellaneous

Analysis and optimization of gas-centrifugal separation of uranium isotopes by neural networks

  • 1. Instituto de Pesquisas Energeticas e Nucleares (IPEN), Sao Paulo, SP (Brazil)
  • 2. Sao Paulo Univ., SP (Brazil). Escola Politecnica
  • 3. Centro Tecnologico da Marinha (CTMSP), Sao Paulo, SP (Brazil)

Description

Neural networks are an attractive alternative for modeling complex problems that show too many difficulties to be solved by phenomenological model. A feed-forward neural network was used to model a uranium isotopes gas-centrifugal separation. The prediction showed good agreement with the experimental data. An optimization study was performed. The optimal operation condition was tested by a new experiment and a difference of less than 1% was found. (author). 18 refs., 9 figs., 1 tab

Part of:
Proceedings of the 11. ENFIR: Meeting on reactor physics and thermal hydraulics

Additional details

Publishing Information

Imprint Title
Proceedings of the 11. ENFIR: Meeting on reactor physics and thermal hydraulics
Imprint Pagination
838 p.
Journal Page Range
p. 350-356.

Conference

Title
Meeting on reactor physics and thermal hydraulics.
Acronym
11. ENFIR
Dates
18-22 Aug 1997.
Place
Pocos de Caldas, MG (Brazil).

Optional Information

Notes
Imprint:Joint nuclear conference with the 4. ENAN: Brazilian meeting on nuclear applications.