蒙特卡罗计算中颗粒型燃料的随机分布模型比较

高温气冷堆是新一代反应堆系统的热门候选堆型，已经受到国际上越来越多的关注。为设计和分析这种堆型，因其特有的包覆颗粒燃料引入了双重非均匀性，需要应用随机分布模型。对粗网格模型、细网格随机(FLS)模型、随机顺序添加(RSA)模型、子网格随机(Sub-FLS)模型和Metropolis模型等进行了研究，通过计算分析比较给出了各种模型的优缺点。结果表明:子网格随机模型和连续的RSA模型非常接近参考值，但是连续RSA模型的建模时间随着燃料体积份额的增加连续快速上升。 Key words: coated particle fuels; stochastic transport model; Monte Carlo; random distribution

Comparison of stochastic models in Monte Carlo simulation of coated particle fuels

There is growing interest worldwide in very high temperature gas cooled reactors as candidates for next generation reactor systems. For design and analysis of such reactors with double heterogeneity introduced by the coated particle fuels that are randomly distributed in graphite pebbles, stochastic transport models are becoming essential. Several models were reported in the literature, such as coarse lattice models, fine lattice stochastic(FLS) models, random sequential addition (RSA) models, metropolis models. The principles and performance of these stochastic models are described and compared in this paper. Compared with the usual fixed lattice methods, sub-FLS modeling allows more realistic stochastic distribution of fuel particles and thus results in more accurate criticality calculation. Compared with the basic RSA method, sub-FLS modeling requires simpler and more efficient overlapping checking procedure.