非定常粒子输运蒙特卡罗散射源分层抽样方法

定常粒子输运蒙特卡罗并行计算是成功的,因为粒子游动是独立的,可以把模拟的粒子数等分到每个处理器去.然而,对非定常问题,由于每个时间步涉及散射源和几何网格的通讯,它严重的制约了并行规模,导致并行不可扩展.研究了两种算法,采用自适应分配处理器,提高了加速比和处理器的利用率;采用蒙特卡罗分层抽样大大降低了处理器之间散射源的通讯量,并行可扩展性显著改善,取得了理想的加速比.     

Analysis of communication costs for domain decomposed Monte Carlo methods in nuclear reactor analysis

A domain decomposed Monte Carlo communication kernel is used to carry out performance tests to establish the feasibility of using Monte Carlo techniques for practical Light Water Reactor (LWR) core analyses. The results of the prototype code are interpreted in the context of simplified performance models which elucidate key scaling regimes of the parallel algorithm.     

组合几何蒙特卡罗粒子输运支撑软件框架JCOGIN介绍

介绍了JCOGIN支撑软件框架层次式、模块化的体系结构及其核心数据结构。这些使得JCOGIN软件框架能够支撑大规模复杂几何的蒙特卡罗（MC）粒子输运并行计算。同时介绍了在JCOGIN支撑软件框架下研发的JMCT程序,并进行了性能测试,在2万处理器核上模拟20亿粒子,并行效率达到70%。     

Particle filtering with path sampling and an application to a bimodal ocean current model

This paper introduces a recursive particle filtering algorithm designed to filter high dimensional systems with complicated non-linear and non-Gaussian effects. The method incorporates a parallel marginalization (PMMC) step in conjunction with the hybrid Monte Carlo (HMC) scheme to improve samples generated by standard particle filters. Parallel marginalization is an efficient Markov chain Monte Carlo (MCMC) strategy that uses lower dimensional approximate marginal distributions of the target distribution to accelerate equilibration. As a validation the algorithm is tested on a 2516 dimensional, bimodal, stochastic model motivated by the Kuroshio current that runs along the Japanese coast. The results of this test indicate that the method is an attractive alternative for problems that require the generality of a particle filter but have been inaccessible due to the limitations of standard particle filtering strategies.     

蒙特卡罗区域分解并行计算中确保串并行结果一致的伪随机数应用

物理建模的精细化和三维模拟给蒙特卡罗粒子输运计算的规模成千上万倍的增加,甚至超过单核内存的最大规模,仅仅依靠传统的粒子并行蒙特卡罗计算无法实现对模型模拟,区域分解并行是可能的解决方法之一.然而区域分解带来了粒子在各区域间进行迁移,导致现有的伪随机数应用方式无法确保串行计算和并行计算的结果一致.针对这种现象,本文提出赋予粒子随机数属性和动态派生次级粒子随机数的技巧来确保区域分解并行计算的串并行结果一致.     

非定常输运问题适应于消息传递并行编程环境的香农熵计算方法

在多计算步的非定常输运问题的蒙特卡罗模拟中,为自动调整每一步的样本数以获得较高的计算效率,可以有多种准则.一种可选的方法是在每一步每隔若干样本监测一次系统中未死亡粒子属性分布对应的香农熵的收敛情况以决定何时停止追加样本,此种方法需要在每一步频繁计算香农熵值.由于在MPI消息传递并行编程环境下香农熵的经典计算方法必须广播大量的数据,导致每一步的计算时间随香农熵计算频率的提高而快速增大,这显然是不能满足实际需求的.本文提出了一种适应于消息传递并行编程环境的香农熵计算新方法,该方法计算得到的香农熵值并不等价于经典方法,但二者之间的差别会随着样本数的增加而趋于零.新方法的最大优势是高频计算香农熵值的时间代价大为降低,为最终实现基于香农熵收敛判断的每步样本数的自动调整奠定了必要的基础.     

热辐射输运问题的高效蒙特卡罗模拟方法

惯性约束聚变研究中,热辐射光子在介质中的输运以及热辐射光子与介质的相互作用是重要研究课题,蒙特卡罗方法是该类问题的重要研究手段之一.隐式蒙特卡罗方法虽然能正确地模拟热辐射在介质中的输运过程,但当模拟重介质(材料的吸收系数大)问题时,该方法花费的计算时间将变得很长,导致模拟效率很低.本文以离散扩散蒙特卡罗方法为基础,开发了"离散扩散蒙特卡罗方法辐射输运模拟程序",可以较好地解决重介质区的计算效率问题,但是离散扩散蒙卡罗方法在模拟轻介质区时精度不够高.辐射输运问题中通常既有轻介质也有重介质,为了能同时解决蒙特卡罗方法模拟的效率和精度问题,本文研究了离散扩散蒙特卡罗方法与隐式蒙特卡罗方法相结合的模拟方法,并提出了新的扩散区与输运区界面处理方法,研制了混合蒙特卡罗方法的辐射输运模拟程序.典型辐射输运问题模拟显示:在模拟重介质问题时,该程序能大幅缩短模拟时间,且能取得与隐式蒙特卡罗方法一致的结果;在模拟轻重介质均存在的问题时,与隐式蒙特卡罗方法相比,混合蒙特卡罗方法的模拟精度与其相当且计算效率同样能够得到显著提升.     

Numerical modeling of thermal force in a plasma for test-ion transport simulation based on Monte Carlo Binary Collision Model

We present a new numerical model of the thermal force in a plasma, based on the Monte Carlo Binary Collision Model (BCM) [T. Takizuka, H. Abe, J. Comput.Phys. 25 (1977) 205]. This model can be applied for the transport simulation of test ions. The model consists of two steps: (i) choosing a background plasma ion velocity from a distorted Maxwell distribution under the temperature gradient, and (ii) calculating a Coulomb collision between a test particle and the above chosen ion by using the BCM. For the step (i), we developed a velocity sampling method from a distorted Maxwellian, which enables the BCM to bring the thermal force on a test particle in the step (ii).A systematic series of simulations has been performed under various conditions to examine the model. The results of these simulations have been compared with the theoretical values, and it is shown that our model simulates the thermal force correctly for important characteristic features; dependences on the temperature gradient, the test particle velocity, and the background plasma density.     

粒子输运蒙特卡罗模拟现状概述

蒙特卡罗(MC)方法发展已有60多年历史,广泛应用于核科学及相关领域.MC方法超强的几何处理能力,使用精密的连续点截面参数,能够模拟各种复杂几何系统内的中子、光子、电子、α粒子、质子及其耦合输运问题.随着计算机的快速发展,通过大规模并行计算,MC方法及程序已成为模拟各种粒子输运问题的首选工具.     

蒙特卡罗临界计算全局计数效率新算法研究

为提高蒙特卡罗临界计算时全局计数的整体效率, 对比分析了新提出的均匀计数密度算法、均匀径迹数密度算法和原有的均匀裂变点算法. 以大亚湾核反应堆pin-by-pin模型的全局体平均通量计数和中子沉积能计数为例, 前两种算法较均匀裂变点算法都获得了整体效率的提高. 上述算法已经在自主开发的并行蒙特卡罗输运程序JMCT上予以实现.     

基于蒙特卡罗正算输运的全局减方差方法

蒙特卡罗方法是当前形势下辐射屏蔽计算的首选分析工具。小概率深穿透问题则是屏蔽计算的关键与亟待解决的核心问题,需要使用有效的减方差技巧。针对全局问题,利用蒙特卡罗正算输运得到的粒子通量或探测响应来构建权重窗参数,将现有的粒子位置偏移拓展到位置和能量偏倚。利用国际屏蔽基准题进行测试验证,通过使用该方法,粒子被引导到模型的所有位置。平均相对误差降低到10%以下,几乎所有网格区域都有粒子统计。结果表明,基于蒙特卡罗正算输运的输运偏倚参数构建方法能够实现全局减方差。     

Particle in Cell/Monte Carlo Collision Method for Simulation of RF Glow Discharges: Effect of Super Particle Weighting

A parallel Particle in Cell/Monte Carlo Collision (PIC/MCC) numerical code for glow discharge plasma simulations is developed and verified. This method is based on simultaneous solution of the Lorentz equations of motion of super particles, coupled with the Poisson's equation for electric field. Collisions between the particles are modelled by the Monte Carlo method. Proper choice of particle weighting is critically important in order to perform adequate and efficient PIC simulations of plasma. Herein, effects of particle weighting on the simulations of capacitive radio‐frequency argon plasma discharges are studied in details. (© 2014 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Monte Carlo algorithm for drift and diffusion of ions in anisotropic,non-homogeneous media

A new Monte Carlo algorithm for ion transport in two-dimensional anisotropic media is reported. It is based on physical considerations of drift and diffusion in anisotropic media with or without an impermeable boundary. Inhomogeneities in the medium and electric field can be taken into account by averaging along the ion trajectory. The algorithm has been applied to the calculation of ion transport in liquid crystal displays and has been successfully compared with a finite difference program on a one-dimensional liquid crystal structure.     

A benchmark comparison of Monte Carlo particle transport algorithms for binary stochastic mixtures

We numerically investigate the accuracy of two Monte Carlo algorithms originally proposed by Zimmerman [1] and Zimmerman and Adams [2] for particle transport through binary stochastic mixtures. We assess the accuracy of these algorithms using a standard suite of planar geometry incident angular flux benchmark problems and a new suite of interior source benchmark problems. In addition to comparisons of the ensemble-averaged leakage values, we compare the ensemble-averaged material scalar flux distributions. Both Monte Carlo transport algorithms robustly produce physically realistic scalar flux distributions for the benchmark transport problems examined. The base Monte Carlo algorithm reproduces the standard Levermore-Pomraning model [3] and [4] results. The improved Monte Carlo algorithm generally produces significantly more accurate leakage values and also significantly more accurate material scalar flux distributions. We also present deterministic atomic mix solutions of the benchmark problems for comparison with the benchmark and the Monte Carlo solutions. Both Monte Carlo algorithms are generally significantly more accurate than the atomic mix approximation for the benchmark suites examined.     

Particle Size Distribution of Nanospheres by Monte Carlo Fitting of Small Angle X‐Ray Scattering Curves

A method for recovering the size distribution of spherical particles from small angle scattering data by using a Monte Carlo interference function fitting algorithm is presented. The method is based on the direct simulation of the small angle scattering data upon the assumption of non‐interacting hard sphere ensembles (“dilute” solution approximation). The algorithm for retrieving the particle size distribution does not require any additional parameters apart from the input of the scattering data. The fitting strategy necessarily implies positive particle size distributions, while preserving the advantage of the indirect transformation method for data desmearing. Furthermore, the present approach does not use any regularisation procedures of the best fit solution and favours smooth particle size distributions. The Monte Carlo procedure has been tested against several simulated cases with various types of mono‐ and bi‐modal size distributions and different noise levels. In the special case of non‐interacting spheres, the Monte Carlo fitting algorithm had the same retrieving ability as the well assessed indirect transformation, structure interference and maximum entropy methods. Finally, the algorithm was applied to retrieve the distribution of spherical nanopowders produced by gas‐to‐particle conversion both as free powder and as reinforcing second‐phase agent in polymer nanocomposites.     

Relaxation dynamics in small clusters: A modified Monte Carlo approach

Relaxation dynamics in two-dimensional atomic clusters consisting of mono-atomic particles interacting through Lennard-Jones (L-J) potential has been investigated using Monte Carlo simulation. A modification of the conventional Metropolis algorithm is proposed to introduce realistic thermal motion of the particles moving in the interacting L-J potential field. The proposed algorithm leads to a quick equilibration from the nonequilibrium cluster configuration in a certain temperature regime, where the relaxation time (τ), measured in terms of Monte Carlo Steps (MCS) per particle, vary inversely with the square root of system temperature (√T) and pressure (P); τ ∝ (P√T)^?1. From this a realistic correlation between MCS and time has been predicted.     

Hierarchical fractional-step approximations and parallel kinetic Monte Carlo algorithms

We present a mathematical framework for constructing and analyzing parallel algorithms for lattice kinetic Monte Carlo (KMC) simulations. The resulting algorithms have the capacity to simulate a wide range of spatio-temporal scales in spatially distributed, non-equilibrium physiochemical processes with complex chemistry and transport micro-mechanisms. Rather than focusing on constructing exactly the stochastic trajectories, our approach relies on approximating the evolution of observables, such as density, coverage, correlations and so on. More specifically, we develop a spatial domain decomposition of the Markov operator (generator) that describes the evolution of all observables according to the kinetic Monte Carlo algorithm. This domain decomposition corresponds to a decomposition of the Markov generator into a hierarchy of operators and can be tailored to specific hierarchical parallel architectures such as multi-core processors or clusters of Graphical Processing Units (GPUs). Based on this operator decomposition, we formulate parallel Fractional step kinetic Monte Carlo algorithms by employing the Trotter Theorem and its randomized variants; these schemes, (a) are partially asynchronous on each fractional step time-window, and (b) are characterized by their communication schedule between processors.The proposed mathematical framework allows us to rigorously justify the numerical and statistical consistency of the proposed algorithms, showing the convergence of our approximating schemes to the original serial KMC. The approach also provides a systematic evaluation of different processor communicating schedules. We carry out a detailed benchmarking of the parallel KMC schemes using available exact solutions, for example, in Ising-type systems and we demonstrate the capabilities of the method to simulate complex spatially distributed reactions at very large scales on GPUs. Finally, we discuss work load balancing between processors and propose a re-balancing scheme based on probabilistic mass transport methods.     

The penetration, diffusion and energy deposition of high—energy photon＊

This paper presents a new theory for calculating the transport of high-energy photons and their secondary charged particles. We call this new algorithm characteristic line method, which is completely analytic. Using this new method we cannot only accurately calculate the transport behaviour of energetic photons, but also precisely describes the transport behaviour and energy deposition of secondary electrons, photoelectrons, Compton recoil electrons and positron－electron pairs. Its calculation efficiency is much higher than that of the Monte Carlo method. The theory can be directly applied to layered media situation and obtain a pencil-beam-modelled solution. Therefore, it may be applied to clinical applications for radiation therapy.     

The robustness in dynamics of out of equilibrium bidirectional transport systems with constrained entrances

Macroscopic and microscopic long-distance bidirectional transfer depends on connections between entrances and exits of various transport mediums. Persuaded by the associations, we introduce a small system module of Totally Asymmetric Simple Exclusion Process including oppositely directed species of particles moving on two parallel channels with constrained entrances. The dynamical rules which characterize the system obey symmetry between the two species and are identical for both the channels. The model displays a rich steady-state behavior, including symmetry breaking phenomenon. The phase diagram is analyzed theoretically within the mean-field approximation and substantiated with Monte Carlo simulations. Relevant mean-field calculations are also presented. We further compared the phase segregation with those observed in previous works, and it is examined that the structure of phase separation in proposed model is distinguished from earlier ones. Interestingly, for phases with broken symmetry, symmetry with respect to channels has been observed as the distinct particles behave differently while the similar type of particles exhibits the same conduct in the system. For symmetric phases, significant properties including currents and densities in the channels are identical for both types of particles. The effect of symmetry breaking occurrence on the Monte Carlo simulation results has also been examined based on particle density histograms. Finally, phase properties of the system having strong size dependency have been explored based on simulations findings.