JFNK方法在SN输运计算中的应用

JFNK(Jacobian-free Newton-Krylov)方法是一种求解非线性方程的高效迭代算法。传统输运计算中的负通量修正与k-特征值迭代使得原本线性的输运计算转变为非线性问题数值求解。为提高非线性输运问题的计算效率,将这两类非线性问题离散成残差形式的非线性方程组,并采用JFNK方法对其进行迭代求解。分析不同约束条件对JFNK方法性能的影响,并将其与NK(Newton-Krylov)方法进行对比。针对JFNK方法的内迭代过程,分析两类子空间方法(GMRES(m)与LGMRES)对整体计算效率的影响。数值结果表明：①相比于传统的幂迭代方法,JFNK方法具有更高的计算效率;②Jacobian矩阵向量积的差分近似对结果没有影响,且基于物理的约束条件比标准的数学约束更加高效;③LGMRES可以充分利用子空间的信息,从而使得JFNK方法整体表现更加高效。     

节块内嵌离散纵标(SN)方法的算法研究及程序开发

介绍节块内嵌离散纵标(S_N)方法求解三维堆芯中子输运/扩散方程的算法框架.在基于扩散理论的三维粗网节块展开方法(NEM)的算法体系中,用基于输运理论的径向二维细网节块离散纵标方法(NDOM)的内迭代过程,替代节块展开方法(NEM)内迭代的径向求解过程.该算法充分考虑了核电厂反应堆堆芯的三维结构特点,另一方面,也充分利用了已经成熟的三维粗网节块展开方法(NEM)和二维离散纵标方法(S_N)的研究成果,同时有效避免了利用离散纵标方法(S_N)求解三维中子输运方程所面临的计算内存和计算时间的瓶颈问题.编制开发二维多群节块离散纵标方法(NDOM)模块程序NS_NM和三维多群节块展开方法(NEM)模块程序MGNEM,并以此为基础编制开发节块内嵌S_N方法的模块程序HANWIND;其中,NS_NM为HANWIND求解两维问题的功能模块.针对OECD/NEA-2D C5G7MOX基准问题以及两环路核电厂三维堆芯的数值验算结果表明,节块内嵌S_N方法的算法开发及程序编制有效、切实可行.     

JFNK在高温堆扩散计算中的应用

研究了JFNK框架下高温堆中子扩散问题的求解方法。研究结果表明,JFNK方法在求解与源迭代相同形式中子扩散方程时,相对残差下降趋势为逐渐加快并趋于稳定,有利于更高求解精度的实现。使用通量归一化附加方程可以获得更好的JFNK非线性迭代特性,但在算例中其部分牛顿修正方程求解时间偏多,总计算时间高于显式有效增殖系数附加方程法,需要研究更高效的JFNK预处理方法对线性求解环节进行改善。     

JFNK方法迭代过程与物理约束

对Jacobian-free Newton-Krylov(JFNK)方法迭代过程进行分析,通过在迭代过程中吸收物理约束信息,对JFNK方法进行改进.改进后的JFNK方法迭代过程中的迭代序列总是满足物理约束,克服了迭代过程中可能出现的非物理现象.采用改进之后的算法求解二维三温能量方程,可以保证在迭代过程中不会出现负温度问题,使JFNK方法的健壮性得到提高.     

求解二维三温辐射扩散方程组的一种代数两层迭代方法

在二维三温辐射扩散方程离散代数方程组的求解中,由于光子、电子和离子温度之间存在耦合关系,而且三个温度在同种介质中有不同的扩散性质,使得经典的代数多重网格(AMG)方法难以直接应用.基于特殊粗化策略,在粗网格层解除了这种耦合关系,得到一种代数两层网格方法,而粗网格方程由经典AMG方法求解.将这一算法具体应用于JFNK(Jacobian自由的Newton-Krylov)框架中预处理方程的求解,并基于该框架求解二维三温辐射扩散方程组.数值结果显示了算法的可扩展性和健壮性.     

一种求解三维热辐射输运方程的整体预处理迭代方法及并行计算

为三维灰体热辐射输运方程的隐式离散纵标方法发展一个整体预处理迭代方法并研制并行程序。该方法采用组装线性代数方程组策略,同时求出所有离散方向上的辐射强度。借助预处理的Krylov子空间迭代法,避免复杂网格上扫描算法可能遇到的死锁问题,能够提高健壮性和计算效率。空间离散上采用一阶迎风有限体积格式。数值实验测试变形六面体网格上的收敛率、评估预处理迭代方法的性能并计算辐射和物质的耦合问题,给出三维弯管和黑腔问题的模拟结果,验证程序的正确性和方法的适应性。     

蒙卡中子输运程序JMCT和子通道热工水力程序COBRA-EN耦合计算

介绍了中子输运蒙特卡罗方法与热工水力耦合计算的流程。开发了一套蒙卡中子输运程序JMCT和子通道分析程序COBRA-EN耦合接口。通过33棒束模型的计算展示了考虑耦合计算和不考虑耦合计算的差异,论证了耦合计算在反应堆分析中的重要性。通过对反应堆组件的模拟计算,测试了耦合计算的正确性。最后分析了蒙卡计算的统计涨落和迭代计算过程中收敛标准的关系,讨论了蒙卡中子输运和热工水力耦合过程中收敛标准设置的方案和可行性。     

新型缩减矩阵构造加快特征基函数法迭代求解

针对特征基函数法在分析电大目标电磁散射特性时存在缩减矩阵方程迭代求解收敛慢的问题,提出一种新型缩减矩阵构造方法提高特征基函数法的迭代求解效率.首先,应用奇异值分解技术压缩激励源,求解出新激励源下各子域的特征基函数;其次,将新激励源和特征基函数作为构造缩减矩阵的检验函数和基函数新方法构造的缩减矩阵的对角子矩阵均为单位矩阵,缩减矩阵条件数得到了优化.与传统方法相比,新方法构造的缩减矩阵方程迭代求解效率得到了显著提高;另外,由于矩阵方程求解次数减少,特征基函数的构造效率也得到了提高,数值结果证明了新方法的精确性和有效性.     

改进的节块展开方法求解圆柱几何对流扩散方程

为高效求解球床高温气冷堆物理-热工耦合问题,发展改进节块展开法求解圆柱几何下的对流扩散方程.针对圆柱几何和对流扩散方程的特殊性,采用三阶多项式和指数函数作为r向横向积分方程的展开函数,在节块展开法的框架下高效求解对流扩散方程.数值验证表明,改进的节块展开方法具有固有的迎风特性,在使用粗网节块时依然能保持稳定性和较高的计算精度.     

求解非定常不可压N-S方程的预处理方法

应用预处理技术,对不可压非定常N-S方程使用双时间推进法求解.当沿物理时间层推进时,连续性方程和动量方程沿伪时间方向使用隐式线Gauss-Seidel迭代法求解,对流项采用三阶迎风差分法离散.通过对不同Reynolds数、不同深宽比下非定常驱动腔内流动的模拟,数值研究了预处理法计算非定常不可压粘性流动的收敛特性,分析了沿伪时间层的迭代收敛速度对流场Reynolds数的依赖特征.     

Parallel Jacobian-free Newton Krylov solution of the discrete ordinates method with flux limiters for 3D radiative transfer

The present study introduces a parallel Jacobian-free Newton Krylov (JFNK) general minimal residual (GMRES) solution for the discretized radiative transfer equation (RTE) in 3D, absorbing, emitting and scattering media. For the angular and spatial discretization of the RTE, the discrete ordinates method (DOM) and the finite volume method (FVM) including flux limiters are employed, respectively. Instead of forming and storing a large Jacobian matrix, JFNK methods allow for large memory savings as the required Jacobian-vector products are rather approximated by semiexact and numerical formulations, for which convergence and computational times are presented. Parallelization of the GMRES solution is introduced in a combined memory-shared/memory-distributed formulation that takes advantage of the fact that only large vector arrays remain in the JFNK process. Results are presented for 3D test cases including a simple homogeneous, isotropic medium and a more complex non-homogeneous, non-isothermal, absorbing–emitting and anisotropic scattering medium with collimated intensities. Additionally, convergence and stability of Gram–Schmidt and Householder orthogonalizations for the Arnoldi process in the parallel GMRES algorithms are discussed and analyzed. Overall, the introduction of JFNK methods results in a parallel, yet scalable to the tested 2048 processors, and memory affordable solution to 3D radiative transfer problems without compromising the accuracy and convergence of a Newton-like solution.     

Improving the numerical convergence of viscous-plastic sea ice models with the Jacobian-free Newton–Krylov method

We have implemented the Jacobian-free Newton–Krylov (JFNK) method to solve the sea ice momentum equation with a viscous-plastic (VP) formulation. The JFNK method has many advantages: the system matrix (the Jacobian) does not need to be formed and stored, the method is parallelizable and the convergence can be nearly quadratic in the vicinity of the solution. The convergence rate of our JFNK implementation is characterized by two phases: an initial phase with slow convergence and a fast phase for which the residual norm decreases significantly from one Newton iteration to the next. Because of this fast phase, the computational gain of the JFNK method over the standard solver used in existing VP models increases with the required drop in the residual norm (termination criterion). The JFNK method is between 3 and 6.6 times faster (depending on the spatial resolution and termination criterion) than the standard solver using a preconditioned generalized minimum residual method. Resolutions tested in this study are 80, 40, 20 and 10 km. For a large required drop in the residual norm, both JFNK and standard solvers sometimes do not converge. The failure rate for both solvers increases as the grid is refined but stays relatively small (less than 2.3% of failures). With increasing spatial resolution, the velocity gradients (sea ice deformations) get more and more important. Nonlinear solvers such as the JFNK method tend to have difficulties when there are such sharp structures in the solution. This lack of robustness of both solvers is however a debatable problem as it mostly occurs for large required drops in the residual norm. Furthermore, when it occurs, it usually affects only a few grid cells, i.e., the residual is small for all the velocity components except in very localized regions. Globalization approaches for the JFNK solver, such as the line search method, have not yet proven to be successful. Further investigation is needed.     

预处理JFNK方法求解非平衡辐射扩散方程组

分别采用四种半隐式离散方法构造预处理.针对一维辐射扩散方程组,采用预处理的Jacobian-free Newton-Krylov(PJFNK)求解.数值结果表明预处理方法能够很好地改进JFNK方法的收敛行为.     

A Jacobian-free Newton–Krylov algorithm for compressible turbulent fluid flows

Despite becoming increasingly popular in many branches of computational physics, Jacobian-free Newton–Krylov (JFNK) methods have not become the approach of choice in the solution of the compressible Navier–Stokes equations for turbulent aerodynamic flows. To a degree, this is related to some subtle aspects of JFNK methods that are not well understood, and, if poorly handled, can lead to inefficient and unreliable performance. These are described here, along with strategies for addressing them, leading to an efficient JFNK algorithm for turbulent aerodynamic flows applicable to multi-block structured grids and a one-equation turbulence model. Development of globalization strategies for field-equation turbulence models represents one of the key contributions of the paper. Numerous examples of subsonic and transonic flows over single and multi-element airfoils are presented in order to demonstrate the efficiency and reliability of the algorithm. In addition, a number of guidelines are presented to aid in diagnosing problems with JFNK algorithms.     

Event-based parareal: A data-flow based implementation of parareal

Parareal is an iterative algorithm that, in effect, achieves temporal decomposition for a time-dependent system of differential or partial differential equations. A solution is obtained in a shorter wall-clock time, but at the expense of increased compute cycles. The algorithm combines a fine solver that solves the system to acceptable accuracy with an approximate coarse solver. The critical task for the successful implementation of parareal on any system is the development of a coarse solver that leads to convergence in a small number of iterations compared to the number of time slices in the full time interval, and is, at the same time, much faster than the fine solver. Very fast coarse solvers may not lead to sufficiently rapid convergence, and slow coarse solvers may not lead to significant gains even if the number of iterations to convergence is satisfactory. We find that the difficulty of meeting these conflicting demands can be substantially eased by using a data-driven, event-based implementation of parareal. As a result, tasks for one iteration do not wait for the previous iteration to complete, but are started when the needed data are available. For given convergence properties, the event-based approach relaxes the speed requirements on the coarse solver by a factor of ～K, where K is the number of iterations required for a converged solution. This may, for many problems, lead to an efficient parareal implementation that would otherwise not be possible or would require substantial coarse solver development. In addition, the framework used for this implementation executes a task when the data dependencies are satisfied and computational resources are available. This leads to improved computational efficiency over previous approaches that pipeline or schedule groups of tasks to a particular processor or group of processors.