The apparent uncertainty of the results accumulated in the active cycle is automatically estimated by MC simulation codes with the stochastic theory. After the inactive cycle, responses are accumulated during the active cycle. The inactive cycles insure avoidance of biased results caused by non-converged fission source distribution. The MC power iteration options need to be selected to accurately obtain the results, including the number of inactive cycles, batch size per iteration cycle, the number of active cycles and the others. For the MC eigenvalue calculations, the power iteration method is generally used to estimate reactor characteristics by obtaining unknown fission source distribution. The Monte Carlo (MC) particle transport method has received great attention as an enhancer of computational performance. The proposed method is expected to contribute to improving the reliability of the Monte Carlo criticality calculations. The results show that the batch sizes determined by the proposed method allow the evaluation of local responses without large biases. To determine the criterion for the diagnostics of stability, real-to-apparent standard deviations are analyzed with the stability of fission source distribution. This study proposes a method for selecting a reference criterion of batch size avoiding significant biases in the local responses, and mainly emphasizing the maintenance of the stability of fission source distribution. To solve these issues, previous studies have recommended the selection of large-size batches, which nonetheless, would require large calculation cost. This dependency of the sampled fission source can cause underestimation of response uncertainty. In the process, the fission sources at each cycle are not independently sampled by the iteration scheme. The power iteration method is generally used for Monte Carlo eigenvalue calculations.

Law of ueki download batch eng sub