Development of efficient Krylov preconditioning techniques for multi-dimensional method of characteristics

Two techniques are proposed in the preconditioning for the Krylov sub-space method called the Generalized Minimal RESidual (GMRES) method to accelerate inner iterations based on the method of characteristics (MOC). The GMRES method is an iterative method to solve a linear algebraic system byminimizing the norm of the residual vector. The proposed preconditioning technique is based on the first flight collision probability which is efficiently made by the multi-dimensional MOC code. To simplify the preconditioner, slight couplings among regions are ignored by considering the mean free path. And another proposed technique makes simplified preconditioner by the scaling matrix which can homogenize and de-homogenize the fuel region and the cladding region. The scaling technique reduces the size of the matrix and also reduces the calculation time of inverse matrix. Numerical results show that the preconditioner simplified by the mean free path efficiently reduces the number of iterations for the GMRES algorithm. And the scaling technique keeps the efficiency of preconditioner even in the multi-dimensional geometry. The total calculation time is found to be reduced when these techniques are employed.     

一维特征线法中子输运程序开发及验证

为探讨两维/一维综合法堆芯分析方法,本文基于特征线法研制了一维中子输运程序--PEACH-1D.不同于通常的平源近似特征线方法,PEACH-1D可对子区的中子源项作线性近似;程序运用指数函数插值表和渐近源外推技术来加速计算过程.相关数值结果表明,PEACH-1D具有很高的计算精度和效率,线性源近似的特征线法具备处理较粗网格的能力,值得推广.     

Development and verification of an MOC code employing assembly modular ray tracing and efficient acceleration techniques

In this paper, we report the development and verification of a method of characteristics (MOC) code, PEACH, at Shanghai Jiao Tong University. Both the usual flat-source step characteristics (SC) scheme and the linear source (LS) approximation scheme are adopted for the tracking calculation along the neutron trajectory. The assembly-based modular ray tracing (AMRT) technique that possesses a good geometric flexibility and high efficiency is employed, which makes PEACH able to deal with practical LWR assembly and core problems. Moreover, in order to reduce the computational time of the MOC iteration process, both the multi/few-group two-level cell-based coarse mesh finite difference (CMFD) acceleration and the exponential function interpolation technique are used. This results in a significant acceleration. Numerical results for the OECD NEA C5G7 MOX benchmark problem and a 69-group BWR mini-core problem demonstrate that PEACH is accurate and efficient. Numerical results also demonstrate that the LS scheme is more efficient than the SC scheme, taking less time and system memory to generate results of comparable accuracy. In addition, we find that MOC with CMFD acceleration always converges with almost the same number of outer iterations regardless of the physical problem size and the discretization parameters used. This shows an ideal linear relationship between the run time and the geometric size of the problem.     

Evaluation of Dancoff Factors in Complicated Geometry using the Method of Characteristics

A simple and efficient method to estimate the Dancoff factor in a complicated geometry, named “the Neutron current method,” is presented in this paper. In this method, Dancoff factors are evaluated from the flux values obtained by the method of characteristics (MOC). By setting appropriate neutron sources in the non-fuel regions of target geometry and then executing fixed source calculation by MOC, the neutron current method can evaluate Dancoff factors for complicated geometry. It was demonstrated that the neutron current method can easily be adopted for complicated geometries, such as a PWR fuel assembly or large-scale geometry that is difficult to handle by the traditional collision probability method. By utilizing the neutron current method instead of a traditional collision probability method, the calculation time of Dancoff factors in complicated large geometry is drastically reduced.     

Resonance Calculation for Large and Complicated Geometry Using Tone's Method by Incorporating the Method of Characteristics

A new efficient approach for evaluating the background cross section, which is based on Tone's method, is presented. Though the collision probability method is used in the conventional Tone's method, the method of characteristics (MOC) is used in the present method. Since the computation time of MOC is shorter than that of the collision probability method in a large and complicated geometry, the present method will be useful not only for lattice physics calculation, but also for analyses of advanced reactors with complicated geometry. Verification calculations are carried out in two configurations, i.e., a PWR fuel assembly geometry and a multiassembly geometry adjacent to the baffle-reflector region. The validity of the present method has been confirmed through the results of verification calculations.     

A new three-dimensional method of characteristics for the neutron transport calculation

The method of characteristics (MOC) is a very flexible and effective method for the neutron transport calculation in a complex geometry. It has been well developed in two-dimensional geometries but meets serious difficulty in three-dimensional geometries because of the requirements of large computer memory and long computational time. Due to the demand related to the advanced reactor design for complex geometries, an efficient and flexible three-dimensional MOC is needed. This paper presents a modular ray tracing technique to reduce the amount of the ray tracing data and consequently reduce the memory. In this method, the object geometry is dissected into many cuboid cells by a background mesh. Typical geometric cells are picked out and ray traced, and only the ray tracing data in these typical cells is stored. Furthermore, the Coarse Mesh Finite Difference (CMFD) acceleration method is employed to save computing time. Numerical results demonstrate that the modular ray tracing technique can significantly reduce the amount of ray tracing data, and the CMFD acceleration is effective in shorting the computing time.     

二维MOC的CMFD加速计算模块开发与验证

DRAGON程序中的二维特征线法(MOC)计算模块包含多种加速方法,但加速效率很高的粗网有限差分(CMFD)加速算法并未得到应用。为提高DRAGON程序中现有的二维MOC计算模块的计算效率,开发了CMFD加速计算模块并探究其收敛稳定性。运用C5G7-2D基准题验证所开发的CMFD加速计算模块,验证结果表明,所开发的CMFD加速计算模块与DRAGON程序中原有的加速模块相比具有很高的计算效率,较同类计算程序OpenMOC有更好的收敛稳定性。     

数字化反应堆高保真中子学程序SHARK研发

SHARK程序是由中国核动力研究设计院新近研发的基于全堆芯确定论非均匀输运理论体系的数字化反应堆软件。该软件从多群数据库的截面与共振数据出发,采用改进子群方法刻画有效共振截面的复杂非均匀效应,采用二维/一维或准三维特征线方法开展堆芯层面非均匀输运计算。目前该程序的定态微观问题计算能力已建立完毕。数值结果显示,SHARK程序对于商用压水堆相关基准问题具有良好的计算精度和效率。     

针对球床的三维特征线方法几何建模研究

特征线方法通过在计算区域密置特征线来计算角通量,对于计算区域的材料分布和几何结构没有要求,因此特征线方法的几何处理能力受制于几何描述模块对于各种几何区域的描述能力。基于体素构造（CSG）方法,开发了三维特征线程序MOCP的几何描述模块。该几何描述模块可描述随机分布的球床。针对球形燃料的网格划分方式进行了研究,临界球的计算结果表明,当径向网格超过30层时,k_eff的相对误差小于0.1%。通过对几何描述方式的改进大幅提高了三维特征线追踪的效率,并且实现了在各种形状边界上的特征线布置。     

基于最小二乘的线性源特征线方法研究

特征线方法一般采用平源近似来求解,平源近似需要大量较小的网格才能获得精确的结果,从而会消耗较多内存,影响计算效率。相比平源近似,线性源可在网格较少的情况下得到较精确的数值结果。本文提出一种基于最小二乘的线性源特征线方法,并通过计算压水堆栅元及C5G7基准题等算例,对该方法进行了分析。数值结果表明,基于最小二乘的线性源特征线方法能在网格较少的情况下提供稳定、精确的数值结果。     

Application of wavelets scaling function expansion method in resonance self-shielding calculation

The wavelets expansion method is widely used in various fields due to its powerful ability to simulate the oscillating functions. This method is applied to discretize the energy variable of neutron angular flux within the resonant energy range. Meanwhile, the conventional multi-group method is applied in fast and thermal energy ranges. This coupled method can obtain the problem-dependent continuous-energy neutron flux spectrum within the resonant energy range. The method of characteristics (MOC) is employed as a space-variable solver in this paper to keep the powerful capability of dealing with the complex geometry problems. A pressurized water reactor (PWR) fuel cell problem with UO2 fuel (UOX) and mixed oxide fuel (MOX), and a cylindrical cluster fuel problem are calculated by utilizing this coupled method. Results of these problems are all in good agreement with the results of the Monte Carlo statistical transport code MCNP. It is concluded that this is a valuable method to solve the resonance self-shielding calculation problems in a complex geometry, and it is promising to be applicable for realistic reactor problems.     

线性源近似的中子输运方程特征线解法

基于非结构网格的中子输运方程特征线解法已成为堆芯设计程序中组件输运计算的标准方法之一。但现有的大部分特征线法程序均是基于平源近似的步特征线法模型开发的,平源近似是特征线法中除角度变量直接离散外又一基本假定。本工作提出一种基于线性源近似的中子输运方程特征线解法,并提出相关负中子源分布的修正方法。采用自行研制的数值计算软件PEACH,对OECD/NEAC5G7-MOX2D基准问题和自定义沸水堆小堆芯问题进行检验。数值计算结果表明,本工作提出的线性源近似特征线法模型在相同计算精度的前提下,占用更少的系统内存和运行时间。     

NECP软件包中NECL子群数据与任意几何共振模块研制

本文基于子群方法对NECP软件包开发了多群数据库模块和子群共振计算模块。采用本实验室开发的二维任意几何输运程序矩阵MOC作为子群共振模块的求解器。使用MCNP与子群程序计算一系列的共振基准题,并比较了两者所计算的无限增殖因数k_inf和²³⁵U与²³⁸U的微观截面。结果表明,子群程序对任意几何有良好的适应性和精度,可适用于二维任意几何的共振计算。     

适用于任意几何的特征线边界条件处理方法

边界条件处理是特征线方法（MOC）向任意三维几何拓展时遇到的难点之一。本文提出一种边界条件处理方法,既保留循环特征线中首尾相连的特性,又能像插值方法一样适用于任意几何。首先推导了平源近似下的特征线方程,提出了一种将源项和边界角通量分离处理的内迭代解法。然后证明了该解法具有唯一解,并类似于循环特征线方法给出解的构造方法。最后借助数值积分和权重插值给出迭代计算流程。采用Takeda算例、单铀球水腔模型和C5G7算例进行验证计算,k_eff的最大计算误差分别为21、319和138.8 pcm,表明方法可靠。该方法可应用于任意几何,且不需存储边界通量和进行边界迭代。     

Boundary Condition Processing Method for MOC Calculation in Arbitrary Geometry

Boundary condition processing is one of the difficulties encountered in the application of method of characteristics (MOC) to arbitrary three-dimensional geometry. In this paper, a boundary condition processing method was proposed, which not only preserved the track continuity as cyclic track method, but also could be applied to arbitrary geometry as the interpolation method. The MOC equation was derived under the flat source approximation and an internal iterative method was proposed in which the source term and the boundary angular flux were processed separately. It was proved that the equation had a unique solution which could be constructed similarly to the cyclic track method. The iterative calculation flow was given by numerical integration and weight interpolation. Takeda benchmark, single uranium sphere model with water cavity and C5G7 benchmark were calculated to test the accuracy. The maximum error of k_eff is 21, 319 and 138.8 pcm respectively, which shows that the method is reliable. This method can be applied to arbitrary geometry without storing boundary fluxes and performing boundary iteration.     

基于GPU并行的二维时空中子动力学MOC程序开发及验证

目前特征线方法（MOC）被广泛应用于反应堆精细中子输运计算。为提高基于MOC方法的时空中子动力学输运计算效率,本文开发了ALPHA程序的动力学计算模块,实现了基于GPU并行的二维精细动力学输运计算。同时,实现了基于GPU并行的CMFD加速计算,并对TWIGL基准题和MINI-CORE基准题进行验证。数值结果显示,基于GPU并行的中子动力学计算方法能保证良好的计算精度,且具有明显的加速效果。     

Application of the multigrid amplitude function method for time-dependent MOC based on the linear source approximation

ABSTRACT

An efficient numerical scheme for time-dependent MOC calculations is proposed. In the present scheme, one of the most successful factorization method, the multigrid amplitude function (MAF) method, is employed to achieve faster computation with the minimum degradation for the temporal integration of the scalar flux. In addition, the MAF method is re-derived based on the linear source approximation, which is not applied for time-dependent MOC calculations in the past studies as far as the authors’ knowledge, to reduce the spatial discretization error with the coarser flux region separation. The accuracy and computational time of the present scheme are evaluated through the calculation of the TWIGL and the C5G7-TD 2D benchmark problems. The present calculation results show that the present scheme is 6.2 times faster than the conventional method while achieving the same accuracy in the C5G7-TD benchmark problem.     

一步法输运计算程序KuaFu开发与验证

为了拓展一步法输运计算方法在结构复杂先进反应堆中的应用,基于构建实体几何理论及二维/一维耦合方法,采用C++、Python混合编程开发了一步法输运计算程序KuaFu,并应用粗网有限差分方法(CMFD)、大规模并行技术对二维/一维耦合方法进行加速。通过C5G7基准题对几何建模的可视化功能、并行功能及计算精度进行评估,获得计算结果与蒙特卡罗程序(MCNP)的相对误差。计算结果表明,程序具有较好的可视化功能和用户友好性;KuaFu程序与MCNP参考解符合较好,计算精度良好。