蒙特卡罗临界计算能量偏倚的最佳源偏倚方法

使用堆用蒙特卡罗程序RMC进行临界计算时采用了传统的裂变源迭代法，即每代源中子按照真实物理过程产生、存库和再抽样。传统裂变源迭代法的计算代之间存在较大的相关性，导致了方差低估计现象，同时总体方差未实现最优化。为实现源中子在空间和能量上的最佳分布，并消除方差低估计现象，提出了能量偏倚的最佳源偏倚方法。该方法基于最佳分层抽样法，结合香农熵诊断和组统计方法，对源中子的空间和能量进行偏倚，实现了全局减方差的计算效果。在RMC程序中开发了能量偏倚的最佳源偏倚方法，并对典型压水堆组件进行测试，计算结果证明了该方法的正确性和有效性。

Energy Biased Optimal Source Bias Method for Monte Carlo Criticality Calculation

The traditional fission source iterative method is adopted when the Monte Carlo code RMC is used to do the criticality calculation, that is, each source neutron is generated, stored and resampled according to the real physical process. There is a large correlation between the calculation cycles from the traditional fission source iterative method, which leads to the phenomenon of the underestimate of variance and the overall variance is not optimized. To achieve the optimal distribution in space and energy of each source neutron, and eliminate the underestimate of variance, the energy biased optimal source bias method was proposed. Based on the optimal stratified sampling method, combined with the Shannon entropy diagnosis and batch method, the method achieves the effect of global variance reduction. The method was tested by a standard PWR assembly, and the results show that the method is correct and effective.