Analysis of error in Monte Carlo transport calculations
The Monte Carlo method for neutron transport calculations suffers, in part, because of the inherent statistical errors associated with the method. Without an estimate of these errors in advance of the calculation, it is difficult to decide what estimator and biasing scheme to use. Recently, integral equations have been derived that, when solved, predicted errors in Monte Carlo calculations in nonmultiplying media. The present work allows error prediction in nonanalog Monte Carlo calculations of multiplying systems, even when supercritical. Nonanalog techniques such as biased kernels, particle splitting, and Russian Roulette are incorporated. Equations derived here allow prediction of how much a specific variance reduction technique reduces the number of histories required, to be weighed against the change in time required for calculation of each history. 1 figure, 1 table. (RWR)
- Research Organization:
- Los Alamos Scientific Lab., NM (USA)
- DOE Contract Number:
- W-7405-ENG-36
- OSTI ID:
- 6286976
- Report Number(s):
- LA-7636-T
- Country of Publication:
- United States
- Language:
- English
Similar Records
Analysis of error in Monte Carlo transport calculations
Minimizing the cost of splitting in Monte Carlo radiation transport simulation
Monte Carlo techniques for analyzing deep-penetration problems
Journal Article
·
Wed Aug 01 00:00:00 EDT 1979
· Nucl. Sci. Eng.; (United States)
·
OSTI ID:5865186
Minimizing the cost of splitting in Monte Carlo radiation transport simulation
Technical Report
·
Wed Oct 01 00:00:00 EDT 1980
·
OSTI ID:6753224
Monte Carlo techniques for analyzing deep-penetration problems
Journal Article
·
Fri Jan 31 23:00:00 EST 1986
· Nucl. Sci. Eng.; (United States)
·
OSTI ID:5903948