ADVANCED MONTE CARLO METHOD

Challenges in Monte Carlo Nuclear Analysis

The "Monte Carlo, MC" method can be used as a counterpart to "Deterministic" transport nuclear design code. However, there are many challenges and issues in replacing the MC method with nuclear design analysis code to fully leverage its advantages.

• Computation Time and Memory

The following indicates areas for advancing Monte Carlo methods.

• Monte Carlo based Two-Step Procedure

• Monte Carlo Depletion Analysis

• Large Scale Calculations (Whole Core)

  • Slow Convergence

  • Bias in Local Tallies

• Uncertainty Quantification (UQ) Analysis

  • Adjoint S/U using Perturbation

  • Stochastics Sampling

Monte Carlo Few Group Constants Generation (몬테칼로 군정수 생산체계)

The two-step procedure is currently the most common basis of neutronics design computations for nuclear reactors. In this procedure, a deterministic assembly or lattice physics code is normally utilized to determine a multigroup flux distribution needed for generation of homogenized few-group diffusion theory constants (FGC).

The fewgroup constants are then inputted into diffusion theory codes to calculate the neutronics design parameters such as the effective multiplication factor, the power density distribution, control rod worth, reactivity coefficients, etc., of the reactor of interest. However, generation of the few-group constants in various nuclear applications by deterministic lattice physics codes has posed some difficulties, which may stem from inherent drawbacks of deterministic methods such as multigroup approximation and inability to implement the detailed geometrical information of the lattices as designed. 

Because of the ability of the MC method to use continuous-energy nuclear data and detailed geometric information, the MC method is free from the inherent drawbacks of, and is supposedly superior to, the deterministic methods. It can calculate not only the groupwise neutron flux distribution known as the infinite medium spectrum but also the groupwise reaction rate distribution with high accuracy, which in turn enables one to produce high-accuracy fewgroup reaction cross sections. 

The MC method as the few-group constants generator has several inherently superior features, compared with its deterministic counterpart. To examine the qualification of the few-group constants generated by the MC method, we combined MASTER, a three-dimensional 3-D core neutronics analysis code, with the few-group constants generation module of McCARD to form the McCARD/MASTER code system for two-step core neutronics calculations.

The following indicates areas for MC FGC generations.

• Diffusion Coefficient Generation

• Application to Various Applications

• Uncertainty Esitmation in FGC

• Uncertainty Propagation Analysis in Burnup Analyses

The following papers show our research papers in this topics.

• Ho Jin Park, Hyung Jin Shim, and Chang Hyo Kim, “Generation of Few-Group Diffusion Theory Constants by Monte Carlo Code McCARD,” Nuclear Science and Engineering, Vol. 172,pp 66-77 (2012).

• Ho Jin Park, Hyung Jin Shim, Han Gyu Joo, and Chang Hyo Kim, “Uncertainty quantification of few-group Diffusion Theory Constants Generated by the B1 Theory-Augmented Monte Carlo Method,” Nuclear Science and Engineering, Vol. 175,pp 28-43 (2013).

• Ho Jin Park, Dong Hyuk Lee, Hyung Jin Shim, and Chang Hyo Kim, “Uncertainty Propagation Analysis for Yonggwang Nuclear Unit 4 by McCARD/MASTER Core Analysis System,” Nuclear Engineering and Technology, Vol. 46,pp 291-298 (2014).

• Ho Jin Park and Jin Young Cho, “Critical Buckling Generation of TCA Benchmark by the B1 Theory-Augmented Monte Carlo Calculation and Estimation of Uncertainties,” Energies, Vol. 14, 2578, April (2021).