Related Experiment Videos
Performance of the improved version of monte Carlo code A3MCNP for large-scale shielding problems
M Omura1, Y Miyake, T Hasegawa
1Mitsubishi Heavy Industries, Yokohama 220-8401, Japan. masaki_omura@mhi.co.jp
Radiation Protection Dosimetry
|April 11, 2006
Summary
Automatic Adjoint Accelerated MCNP (A3MCNP) software was enhanced to A3MCNPV, improving computational efficiency for radiation shielding simulations. The new version efficiently handles volumetric sources, reducing memory requirements and calculation time.
Area of Science:
- Nuclear Engineering
- Computational Physics
- Radiation Shielding
Background:
- Monte Carlo methods are crucial for radiation transport simulations.
- Variance reduction techniques are essential for efficient Monte Carlo calculations.
- Existing methods like CADIS require accurate importance functions.
Purpose of the Study:
- To introduce A3MCNPV, an improved version of A3MCNP with volumetric source capabilities.
- To demonstrate the efficiency and accuracy of A3MCNPV for complex shielding problems.
- To reduce computational resources needed for Monte Carlo simulations.
Main Methods:
- Utilized the Consistent Adjoint Driven Importance Sampling (CADIS) methodology.
- Employed the 3-D Sn transport TORT code to generate importance functions.
- Implemented a volumetric source option in A3MCNP, creating A3MCNPV.
- Applied A3MCNPV to concrete cask shielding and PWR dosimetry problems.
Main Results:
- A3MCNPV significantly reduces memory requirements compared to point source configurations.
- The enhanced code provides precise calculation results with reduced computation time.
- Demonstrated successful application to realistic shielding and dosimetry scenarios.
Conclusions:
- A3MCNPV offers a more efficient and memory-sparing approach for Monte Carlo radiation transport simulations.
- The volumetric source option is a key improvement for large-scale shielding applications.
- A3MCNPV enhances the practical applicability of automated variance reduction techniques.