首页> 外文会议>AIAA aerospace sciences meeting;AIAA SciTech forum >Fast Time Integration of Navier-Stokes Equations with an Exponential-Integrator Scheme
【24h】

Fast Time Integration of Navier-Stokes Equations with an Exponential-Integrator Scheme

机译:Navier-Stokes方程与指数积分器方案的快速时间积分

获取原文

摘要

In this paper, the second-order exponential time-integrator scheme, named the Predictor-Corrector Exponential scheme (PCEXP), originally developed by S.-J. Li, etc, in AIAA-2017-0753 is successfully extended to solve the compressible Navier-Stokes equations over three-dimensional geometries with hybrid curved elements. The stiffness of viscous flows that leads to a severe time-step size restriction in the explicit methods is addressed in the PCEXP content using a nearly exact evaluation of the residual Jacobians. The second method of Bassi-Rebay (BR2) is employed to discretize the viscous terms with a discontinuous Galerkin method. A hybrid approach for accurately computing the analytical high-order viscous Jacobian matrices is presented and an efficient way to simplify the computation is discussed, resulting in a robust while fast exponential time-integration method. We evaluate the PCEXP scheme using several benchmark test cases to demonstrate its substantial speedup compared to the classical third-order TVD Runge-Kutta approach and comparable performance to the implicit backward Euler (BE) method with large time steps.
机译:在本文中,二阶指数时间积分器方案称为Predictor-Corrector指数方案(PCEXP),最初由S.-J开发。 Li等人在AIAA-2017-0753中成功扩展了求解具有混合弯曲元素的三维几何上的可压缩Navier-Stokes方程。在PCEXP内容中,使用残差雅可比定律的几乎精确评估,解决了显式方法中导致严格的时间步长限制的粘性流的刚度。 Bassi-Rebay的第二种方法(BR2)用于通过不连续的Galerkin方法离散化粘性项。提出了一种用于精确计算解析高阶粘性雅可比矩阵的混合方法,并讨论了一种简化计算的有效方法,从而产生了一种健壮而快速的指数时间积分方法。我们使用几个基准测试案例评估PCEXP方案,以证明与传统的三阶TVD Runge-Kutta方法相比,它的实质性提速以及与具有较大时间步长的隐式后向Euler(BE)方法的性能相当。

著录项

  • 来源
  • 会议地点 Kissimmee(US)
  • 作者单位

    Beijing Computational Science Research Center Beijing 100193 China;

    University of Kansas Lawrence. KS 66045 USA;

    University of South Carolina Columbia SC 29208 USA;

    Beijing Computational Science Research Center Beijing 100193 China Old Dominion University Norfolk VA 23529 USA;

  • 会议组织
  • 原文格式 PDF
  • 正文语种
  • 中图分类
  • 关键词

  • 入库时间 2022-08-26 14:34:59

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