...
首页> 外文期刊>Applied numerical mathematics >A posteriori sub-cell finite volume limiting of staggered semi-implicit discontinuous Galerkin schemes for the shallow water equations
【24h】

A posteriori sub-cell finite volume limiting of staggered semi-implicit discontinuous Galerkin schemes for the shallow water equations

机译:浅水方程组的交错半隐式不连续Galerkin格式的后验子单元有限体积限制

获取原文
获取原文并翻译 | 示例
           

摘要

We propose a novel family of a posteriori sub-cell finite volume limiters for spatially high order accurate semi-implicit discontinuous Galerkin (DG) schemes on staggered Cartesian grids for the solution of the shallow water equations expressed in conservative form in one and two space dimensions.We start from the unlimited arbitrary high order accurate staggered semi-implicit DG scheme proposed by Dumbser and Casulli (2013). In this method, the continuity equation and the momentum equations are integrated using a discontinuous finite element strategy on staggered control volumes, where the discrete free surface elevation is defined on the main grid and the discrete momentum is defined on edge-based staggered dual control volumes. In the semi-implicit approach, pressure terms are discretized implicitly, while the nonlinear convective terms are discretized explicitly. Inserting the momentum equations into the discrete continuity equation leads to a well conditioned block penta-diagonal linear system for the free surface elevation which can be efficiently solved with modern iterative methods. However, according to Godunov's theorem, any unlimited high order scheme inevitably produces spurious oscillations in the vicinity of discontinuities and strong gradients.In the present paper, we therefore propose to extend the successful family of a posteriori subcell finite volume limiters recently introduced by Dumbser et al. (2014) for explicit DG schemes also to semi-implicit time discretizations. At time t(n) the unlimited DG scheme is run in order to produce a so-called candidate solution for time t(n+)(1). Then, the cells characterized by a non-admissible candidate solution are found by using physical and numerical detection criteria based on the positivity of the solution, the absence of floating point errors and the use of a relaxed discrete maximum principle (DMP) according to the MOOD strategy of Clain, Loubere and Diot (2013). In all the cells that are flagged as troubled control volumes a more robust semi-implicit finite volume (FV) method is then applied on a sub-grid composed of 2P + 1 cells, where P denotes the polynomial degree used for approximating the discrete solution within the DG scheme. Then, after having identified the troubled cells, the linear system for the new free surface elevation is assembled and solved again, where unlimited cells use the high order semi-implicit DG scheme and limited cells are evolved via the more robust finite volume method. Finally, from the subcell finite volume averages a higher order DG polynomial is reconstructed and the scheme proceeds with the next time step.We apply the new semi-implicit staggered DG method with a posteriori subcell FV limiter to classical benchmarks such as Riemann problems in 1D and circular dam-break problems in 2D with shock waves, showing that the new subcell finite volume limiter is able to resolve shocks accurately without producing spurious oscillations. Moreover, if the solution is smooth, the detector does not find any troubled cells, as expected; consequently, the limiter is not activated and the method corresponds to the unlimited staggered semi-implicit DG scheme. In addition, we carry out numerical tests which show that the new scheme is well-balanced and able to deal with wet and dry fronts. (C) 2018 IMACS. Published by Elsevier B.V. All rights reserved.
机译:我们提出了一种新颖的后验子单元有限体积限制器族,用于交错的笛卡尔网格上的空间高阶精确半隐式不连续伽勒金(DG)方案,用于解决在一维和二维空间中以保守形式表示的浅水方程我们从Dumbser和Casulli(2013)提出的无限的任意高阶精确交错半隐式DG方案开始。在这种方法中,连续性方程和动量方程在交错的控制体积上使用不连续有限元策略进行积分,其中离散自由表面高程定义在主网格上,离散动量在基于边的交错双重控制体积上定义。在半隐式方法中,压力项被隐式离散,而非线性对流项被显式离散。将动量方程式插入离散的连续性方程式中,可以得到条件良好的块状五角形对角线性系统,用于自由表面标高,可以使用现代迭代方法有效地对其进行求解。但是,根据Godunov定理,任何无限的高阶方案都不可避免地在不连续和强梯度附近产生寄生振荡。因此,本文提出了扩展Dumbser等人最近成功的后验子单元有限体积限制器族的方法。等(2014年)的显式DG方案也将半隐式时间离散化。在时间t(n)运行无限DG方案,以便为时间t(n +)(1)生成所谓的候选解。然后,通过物理和数字检测准则,基于解决方案的积极性,不存在浮点误差以及根据规则使用松弛离散最大原理(DMP),找到以不允许的候选解决方案为特征的单元格。 Clain,Loubere和Diot的情绪策略(2013)。然后,在所有标记为有问题的控制量的像元中,将更鲁棒的半隐式有限量(FV)方法应用于由2P +1个像元组成的子网格,其中P表示用于逼近离散解的多项式度在总计划中。然后,在确定有问题的单元之后,重新组装并求解新的自由表面高程的线性系统,其中无限单元使用高阶半隐式DG方案,而有限单元则通过更鲁棒的有限体积方法进行演化。最后,从子单元有限体积平均值中重构出更高阶的DG多项式,并在下一步进行下一步。我们将带有后验子单元FV限制器的新的半隐式交错DG方法应用于经典基准,例如一维Riemann问题以及带有冲击波的2D圆形坝溃决问题,表明新的子电池有限体积限制器能够准确地解决冲击,而不会产生伪振荡。此外,如果解决方案是平滑的,则检测器将不会发现任何有问题的单元,这与预期的一样。因此,限制器未激活,该方法对应于无限交错半隐式DG方案。此外,我们进行的数值测试表明,新方案平衡良好,能够应对干湿两面。 (C)2018年IMACS。由Elsevier B.V.发布。保留所有权利。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号