Loading...

Analysis of latent variance reduction methods in phase space Monte Carlo calculations for 6, 10 and 18MV photons by using MCNP code

Ezzati, A. O ; Sharif University of Technology | 2012

468 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.nima.2012.10.089
  3. Publisher: 2012
  4. Abstract:
  5. In this study, azimuthal particle redistribution (APR), and azimuthal particle rotational splitting (APRS) methods are implemented in MCNPX2.4 source code. First of all, the efficiency of these methods was compared to two tallying methods. The APRS is more efficient than the APR method in track length estimator tallies. However in the energy deposition tally, both methods have nearly the same efficiency. Latent variance reduction factors were obtained for 6, 10 and 18. MV photons as well. The APRS relative efficiency contours were obtained. These obtained contours reveal that by increasing the photon energies, the contours depth and the surrounding areas were further increased. The relative efficiency contours indicated that the variance reduction factor is position and energy dependent. The out of field voxels relative efficiency contours showed that latent variance reduction methods increased the Monte Carlo (MC) simulation efficiency in the out of field voxels. The APR and APRS average variance reduction factors had differences less than 0.6% for splitting number of 1000
  6. Keywords:
  7. Latent variance ; Linear accelerator ; Monte Carlo ; Phase space file
  8. Source: Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment ; Volume 701 , 2012 , Pages 93-98 ; 01689002 (ISSN)
  9. URL: http://www.sciencedirect.com/science/article/pii/S0168900212012570