A pure dynamic Monte Carlo code for the neutronic analysis of nuclear reactors

Analysis of Population Control Techniques for Time-Dependent and Eigenvalue Monte Carlo Neutron Transport Calculations

“…(2) Duplicate-Discard (DD) [4], (3) Splitting-Roulette (SR) [5], (4) Particle Combing (CO) [15], and (5) Modified Particle Combing (COX) [1]. Additionally, there are three different sampling bases with which each of the PCTs can be implemented: uniform, weight-based, and importance-based sampling.…”

“…A modification of CO is recently proposed by Ajami et al [1]. Different from the weight-based CO shown in Fig.…”

“…Nevertheless, thanks to the advancement of high-performance parallel computing, the inherently parallel features of MC can be effectively exploited to a very large extentwhich can significantly reduce run time to solution, particularly for the computationally expensive time-dependent neutron transport simulations [2,3,4,5,6,7,8,9,10,1].…”

“…When all particles already hit the time boundary, the time census is completed, and PCT can be performed on the censused particles. More recent applications of PCT include the use of random particle duplication or deletion [4] in Serpent 2 [13], splitting and Russian-Roulette technique [5] in MCATK [14], particle combing technique [15] in TRIPOLI-4 [16,17] and GUARDYAN [9], and a modified combing technique which is most recently introduced in [1].…”

“…Despite the multiple distinct PCTs proposed in the literature [3,4,5,10,1], documented studies in characterizing and assessing relative performances of all the identified PCTs are still very limited. More recent efforts found in [5,19] specifically compare the splitting and Russian-Roulette technique [5] to the particle combing technique [15]-hereafter referred to as Splitting-Roulette (SR) and Particle Combing (CO), respectively.…”