Implementations of the superhistory method

Conference · Wed Jul 05 04:00:00 EDT 2000

OSTI ID:759066

Gelbard, E M

The superhistory method is incorporated, in different implementations, into two versions of MONK. In this paper the authors intercompare the efficiencies of these implementations via the Figure Of Merit (FOM), and compare the efficiencies of each with that of conventional Monte Carlo (MC). Finally, they suggest preferred versions of MC for eigenvalue calculations. Here, FOM {approx} 1/N{sigma}{sup 2}, where N is the number of histories, and {sigma} is the variance of a quantity of interest. In the criticality-safety version MONK, fission is simulated as suggested in Ref. 1 (Method-1). Every absorption site is a potential fission site, with weight W = {sigma}{sub f}/kx{sigma}{sub a}, where {sigma}{sub f} and {sigma}{sub a} are fission and absorption cross sections, and k is an estimate of the eigenvalue. If W < 1, W is taken as pf, the fission probability. A Method-1 fission produces, on average, {nu} offspring at each site. The reactor-physics MONK uses the standard MC fission treatment (Method-0), i.e. {nu}xW is the average number of neutrons born in a fission, and pf = 1. For consistency, they take absorption sites as potential fission sites in both methods. For v = 1 and a single generation per supergeneration, conventional and superhistory methods coincide.

• Research Organization: Argonne National Lab., IL (US)
• Sponsoring Organization: US Department of Energy (US)
• DOE Contract Number: W-31109-ENG-38
• OSTI ID: 759066
• Report Number(s): ANL/RA/CP-102252
• Country of Publication: United States
• Language: English

Related Subjects

22 GENERAL STUDIES OF NUCLEAR REACTORS
CALCULATION METHODS
COMPARATIVE EVALUATIONS
CROSS SECTIONS
EIGENVALUES
FISSION
M CODES
PROGENY
REACTOR PHYSICS