Automated variance reduction for MCNP using deterministic methods

In order to reduce the user's time and the computer time needed to solve deep penetration problems, an automated variance reduction capability has been developed for the MCNP Monte Carlo transport code. This new variance reduction capability developed for MCNP5 employs the PARTISN multigroup di...

Full description

Saved in:
Bibliographic Details
Published in:Radiation protection dosimetry 2005-12, Vol.116 (1-4), p.508-512
Main Authors: Sweezy, J., Brown, F., Booth, T., Chiaramonte, J., Preeg, B.
Format: Article
Language:English
Subjects:
Algorithms
Analysis of Variance
Computer Simulation
Models, Statistical
Monte Carlo Method
Radiation Dosage
Radiation Protection - methods
Radiometry - methods
Software
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In order to reduce the user's time and the computer time needed to solve deep penetration problems, an automated variance reduction capability has been developed for the MCNP Monte Carlo transport code. This new variance reduction capability developed for MCNP5 employs the PARTISN multigroup discrete ordinates code to generate mesh-based weight windows. The technique of using deterministic methods to generate importance maps has been widely used to increase the efficiency of deep penetration Monte Carlo calculations. The application of this method in MCNP uses the existing mesh-based weight window feature to translate the MCNP geometry into geometry suitable for PARTISN. The adjoint flux, which is calculated with PARTISN, is used to generate mesh-based weight windows for MCNP. Additionally, the MCNP source energy spectrum can be biased based on the adjoint energy spectrum at the source location. This method can also use angle-dependent weight windows.
ISSN:0144-8420
1742-3406
DOI:10.1093/rpd/nci257