Real variance analysis of Monte Carlo eigenvalue calculation by McCARD for BEAVRS benchmark

The real variances of local tallies, such as the pin-wise and assembly-wise fission powers, were estimated for the Benchmark for Evaluation and Validation of Reactor Simulations (BEAVRS) fresh core problem using real variance estimation methods implemented in the Seoul National University Monte Carlo (MC) code, McCARD. This code employs Gelbard's batch method, Ueki's method, the fission-source distribution inter-cycle correlation method, and the history-based batch method. Results show that the estimated apparent variances of the local tallies tend to be smaller than the real one, whereas the apparent variance of a global MC tally such as the effective multiplication factor is similar to the real one. Moreover, it was observed that the real-to-apparent standard deviation (SD) ratio of the assembly-wise fission power is larger than that of the pin-wise fission power. The large real-to-apparent SD ratio of the former is explained by considering the correlation coefficients between the local tallies.