Published June 2013 | Version v1
Journal article

Cross section expression based on neural network algorithm

  • 1. National Key Discipline Laboratory of Nuclear Safety and Simulation Technology, Harbin Engineering University, Harbin (China)
  • 2. Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing (China)

Description

In order to obtain the cross sections in different operating conditions affected by the thermal parameters and depletion, the neural network algorithm was used for function approximation. Bayesian normalization function and the early termination method were used to train sample cross section. Further more, this method was tested by the two examples. By comparing with FITLINK, the results show that cross section expression based on neural network algorithm is feasible. (authors)

Additional details

Publishing Information

Journal Title
Atomic Energy Science and Technology
Journal Volume
47
Journal Issue
suppl
Journal Page Range
p. 207-210
ISSN
1000-6931

INIS

Country of Publication
China
Country of Input or Organization
China
INIS RN
48068987
Subject category
S22: GENERAL STUDIES OF NUCLEAR REACTORS;
Descriptors DEI
ALGORITHMS; COMPARATIVE EVALUATIONS; CROSS SECTIONS; F CODES; FUEL ASSEMBLIES; NEURAL NETWORKS; REACTOR CORES
Descriptors DEC
COMPUTER CODES; EVALUATION; MATHEMATICAL LOGIC; REACTOR COMPONENTS

Optional Information

Notes
2 figs., 3 tabs., 5 refs.