You need JavaScript to view this

Variance reduction techniques for 14 MeV neutron streaming problem in rectangular annular bent duct

Abstract

Monte Carlo method is the powerful technique for solving wide range of radiation transport problems. Its features are that it can solve the Boltzmann`s transport equation almost without approximation, and that the complexity of the systems to be treated rarely becomes a problem. However, the Monte Carlo calculation is always accompanied by statistical errors called variance. In shielding calculation, standard deviation or fractional standard deviation (FSD) is used frequently. The expression of the FSD is shown. Radiation shielding problems are roughly divided into transmission through deep layer and streaming problem. In the streaming problem, the large difference in the weight depending on the history of particles makes the FSD of Monte Carlo calculation worse. The streaming experiment in the 14 MeV neutron rectangular annular bent duct, which is the typical streaming bench mark experiment carried out of the OKTAVIAN of Osaka University, was analyzed by MCNP 4B, and the reduction of variance or FSD was attempted. The experimental system is shown. The analysis model by MCNP 4B, the input data and the results of analysis are reported, and the comparison with the experimental results was examined. (K.I.)
Authors:
Ueki, Kotaro [1] 
  1. Ship Research Inst., Mitaka, Tokyo (Japan)
Publication Date:
Mar 01, 1998
Product Type:
Technical Report
Report Number:
JAERI-Review-98-010
Reference Number:
SCA: 663610; PA: JPN-98:006315; EDB-98:080864; SN: 98001981586
Resource Relation:
Other Information: PBD: Mar 1998; Related Information: Is Part Of Use experiences of MCNP in nuclear energy study. 2. Review of variance reduction techniques; Sakurai, Kiyoshi; Yamamoto, Toshihiro [Japan Atomic Energy Research Inst., Tokai, Ibaraki (Japan). Tokai Research Establishment] [eds.]; PB: 313 p.
Subject:
66 PHYSICS; FAST NEUTRONS; DUCTS; RADIATION STREAMING; MONTE CARLO METHOD; COMPUTER CALCULATIONS; ACCURACY; M CODES; RECTANGULAR CONFIGURATION; ANNULAR SPACE; POLYETHYLENES
OSTI ID:
630032
Research Organizations:
Japan Atomic Energy Research Inst., Tokyo (Japan)
Country of Origin:
Japan
Language:
Japanese
Other Identifying Numbers:
Other: ON: DE98755716; TRN: JP9806315
Availability:
OSTI as DE98755716
Submitting Site:
JPN
Size:
pp. 193-209
Announcement Date:
Aug 18, 1998

Citation Formats

Ueki, Kotaro. Variance reduction techniques for 14 MeV neutron streaming problem in rectangular annular bent duct. Japan: N. p., 1998. Web.
Ueki, Kotaro. Variance reduction techniques for 14 MeV neutron streaming problem in rectangular annular bent duct. Japan.
Ueki, Kotaro. 1998. "Variance reduction techniques for 14 MeV neutron streaming problem in rectangular annular bent duct." Japan.
@misc{etde_630032,
title = {Variance reduction techniques for 14 MeV neutron streaming problem in rectangular annular bent duct}
author = {Ueki, Kotaro}
abstractNote = {Monte Carlo method is the powerful technique for solving wide range of radiation transport problems. Its features are that it can solve the Boltzmann`s transport equation almost without approximation, and that the complexity of the systems to be treated rarely becomes a problem. However, the Monte Carlo calculation is always accompanied by statistical errors called variance. In shielding calculation, standard deviation or fractional standard deviation (FSD) is used frequently. The expression of the FSD is shown. Radiation shielding problems are roughly divided into transmission through deep layer and streaming problem. In the streaming problem, the large difference in the weight depending on the history of particles makes the FSD of Monte Carlo calculation worse. The streaming experiment in the 14 MeV neutron rectangular annular bent duct, which is the typical streaming bench mark experiment carried out of the OKTAVIAN of Osaka University, was analyzed by MCNP 4B, and the reduction of variance or FSD was attempted. The experimental system is shown. The analysis model by MCNP 4B, the input data and the results of analysis are reported, and the comparison with the experimental results was examined. (K.I.)}
place = {Japan}
year = {1998}
month = {Mar}
}