Article ID Journal Published Year Pages File Type
1729526 Annals of Nuclear Energy 2011 9 Pages PDF
Abstract
Reactor noise simulations have been performed with the analog Monte Carlo technique in the past. The applicability of the non-analog Monte Carlo technique, which uses “weighing” for the purpose of variance reduction, to reactor noise simulations has been discussed. The joint probability of a pair of counts and Feynman variance-to-mean ratio in the non-analog Monte Carlo technique are formulated for one-speed neutron random walk in an infinite homogeneous medium. Unlike the analog Monte Carlo technique, the fission-related correlation exists even for the number of fission neutrons ν = 1 because the neutron causing a fission survives and will contribute to subsequent detections. As a results, the joint probability and the variance-to-mean ratio has the same function of time as the analog Monte Carlo technique. The probability of an uncorrelated pair of counts for a coincidence detection within one detector is different from the analog Monte Carlo technique, which introduces an extra unknown parameter in Feynman-α method. In the two-detector system, the extra parameter does not arise and the conventional Feynman-α formula can be applied to non-analog Monte Carlo simulations. The formulations derived in this work are verified by the fact that the theoretical values agree well with the numerical results.
Related Topics
Physical Sciences and Engineering Energy Energy Engineering and Power Technology
Authors
,