A proposal on multi-response CADIS method to optimize reduction of regional variances in hybrid Monte Carlo simulation

Kim, Do Hyun; Kim, Song Hyun; Shin, Chang Ho; Kim, Gee Suck; Sun, Gwang Min; Seo, Bo Kyun

doi:10.1016/j.anucene.2016.12.027

Cited by 3 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The biased source distribution ˆ( , ) q r E  [11], which minimizes the variance of a user-desired response R, can be found by using the Lagrange multiplier λ method [17],…”

Section: Hybrid Shielding Methodology With Advantg/mcnpmentioning

confidence: 99%

PCA Benchmark Analysis with ADVANTG3.0.1. and MCNP6.1.1b Codes

Matijević¹,

Pevec²,

Petrović³

2022

joe

View full text Add to dashboard Cite

The Pool Critical Assembly Pressure Vessel (PCA) benchmark is a well known benchmark in the reactor shielding community which is described in the Shielding Integral Benchmark Archive and Database (SINBAD). It is based on the experiments performed at the PCA facility in the Oak Ridge National Laboratory (ORNL) and it can be used for the qualification of the pressure vessel fluence calculational methodology. The measured quantities to be compared against the calculated values are the equivalent fission fluxes at several experimental access tubes (A1 to A8) in front, behind, and inside the pressure-vessel wall simulator. This benchmark is particularly suitable to test the capabilities of the shielding calculational methodology and cross-section libraries to predict invessel flux gradients because only a few approximations are necessary in the overall analysis. This benchmark was analyzed using a modern hybrid stochastic-deterministic shielding methodology with ADVANTG3.0.1 and MCNP6.1.1b codes. ADVANTG3.0.1 is an automated tool for generating variance reduction (VR) parameters for Monte Carlo (MC) calculations with MCNP5v1.60 code (and higher versions). It is based on the multigroup, discrete ordinates solver Denovo, used for approximating the forward-adjoint transport fluxes to construct VR parameters for the final MC simulation. The VR parameters in form of the weight windows and the source biasing cards can be directly used with unmodified MCNP input. The underlining CADIS methodology in Denovo code was initially developed for biasing local MC results, such as point detector or a limited region detector. The FW-CADIS extension was developed for biasing MC results globally over a mesh tallies or multiple point/region detectors. Both CADIS and FW-CADIS are based on the concept of the neutron importance function, which is a solution of the adjoint Boltzmann transport equation. The equivalent fission fluxes calculated with MCNP are based on several highenergy threshold reactions from international dosimetry libraries IRDF-2002 and IRDFF-2014, distributed by the IAEA Nuclear Data Section. The obtained results show a good agreement with referenced PCA measurements. Visualization of the deterministic solution in 3D was done using the VisIt code from the Lawrence Livermore National Laboratory (LLNL).

show abstract

“…The biased source distribution ˆ( , ) q r E  [11], which minimizes the variance of a user-desired response R, can be found by using the Lagrange multiplier λ method [17],…”

Section: Hybrid Shielding Methodology With Advantg/mcnpmentioning

confidence: 99%