A Sub-element Adaptive Shock Capturing Approach for Discontinuous Galerkin Methods

In this paper,a new strategy for a sub-element-based shock capturing for discontinuous Galerkin(DG)approximations is presented.The idea is to interpret a DG element as a col-lection of data and construct a hierarchy of low-to-high-order discretizations on this set of data,including a first-order finite volume scheme up to the full-order DG scheme.The dif-ferent DG discretizations are then blended according to sub-element troubled cell indicators,resulting in a final discretization that adaptively blends from low to high order within a single DG element.The goal is to retain as much high-order accuracy as possible,even in simula-tions with very strong shocks,as,e.g.,presented in the Sedov test.The framework retains the locality of the standard DG scheme and is hence well suited for a combination with adaptive mesh refinement and parallel computing.The numerical tests demonstrate the sub-element adaptive behavior of the new shock capturing approach and its high accuracy.

NUMERICAL SIMULATION OF UNSTEADY-STATE UNDEREXPANDED JET USING DISCONTINUOUS GALERKIN FINITE ELEMENT METHOD

A discontinuous Galerkin finite element method （DG-FEM） is developed for solving the axisymmetric Euler equations based on two-dimensional conservation laws. The method is used to simulate the unsteady-state underexpanded axisymmetric jet. Several flow property distributions along the jet axis, including density, pres- sure and Mach number are obtained and the qualitative flowfield structures of interest are well captured using the proposed method, including shock waves, slipstreams, traveling vortex ring and multiple Mach disks. Two Mach disk locations agree well with computational and experimental measurement results. It indicates that the method is robust and efficient for solving the unsteady-state underexpanded axisymmetric jet.

NUMERICAL INVESTIGATION OF TOROIDAL SHOCK WAVES FOCUSING USING DISCONTINUOUS GALERKIN FINITE ELEMENT METHOD

A numerical simulation of the toroidal shock wave focusing in a co-axial cylindrical shock tube is inves- tigated by using discontinuous Galerkin （DG） finite element method to solve the axisymmetric Euler equations. For validating the numerical method, the shock-tube problem with exact solution is computed, and the computed results agree well with the exact cases. Then, several cases with higher incident Mach numbers varying from 2.0 to 5.0 are simulated. Simulation results show that complicated flow-field structures of toroidal shock wave diffraction, reflection, and focusing in a co-axial cylindrical shock tube can be obtained at different incident Mach numbers and the numerical solutions appear steep gradients near the focusing point, which illustrates the DG method has higher accuracy and better resolution near the discontinuous point. Moreover, the focusing peak pres- sure with different grid scales is compared.

Mixed time discontinuous space-time finite element method for convection diffusion equations

A mixed time discontinuous space-time finite element scheme for secondorder convection diffusion problems is constructed and analyzed. Order of the equation is lowered by the mixed finite element method. The low order equation is discretized with a space-time finite element method, continuous in space but discontinuous in time. Stability, existence, uniqueness and convergence of the approximate solutions are proved. Numerical results are presented to illustrate efficiency of the proposed method.

A multithreaded parallel upwind sweep algorithm for the S_(N) transport equations discretized with discontinuous finite elements

The complex structure and strong heterogeneity of advanced nuclear reactor systems pose challenges for high-fidelity neutron-shielding calculations. Unstructured meshes exhibit strong geometric adaptability and can overcome the deficiencies of conventionally structured meshes in complex geometry modeling. A multithreaded parallel upwind sweep algorithm for S_(N) transport was proposed to achieve a more accurate geometric description and improve the computational efficiency. The spatial variables were discretized using the standard discontinuous Galerkin finite-element method. The angular flux transmission between neighboring meshes was handled using an upwind scheme. In addition, a combination of a mesh transport sweep and angular iterations was realized using a multithreaded parallel technique. The algorithm was implemented in the 2D/3D S_(N) transport code ThorSNIPE, and numerical evaluations were conducted using three typical benchmark problems:IAEA, Kobayashi-3i, and VENUS-3. These numerical results indicate that the multithreaded parallel upwind sweep algorithm can achieve high computational efficiency. ThorSNIPE, with a multithreaded parallel upwind sweep algorithm, has good reliability, stability, and high efficiency, making it suitable for complex shielding calculations.

A Posteriori Error Estimates for Finite Element Methods for Systems of Nonlinear,Dispersive Equations

The present study regards the numerical approximation of solutions of systems of Korteweg-de Vries type,coupled through their nonlinear terms.In our previous work[9],we constructed conservative and dissipative finite element methods for these systems and presented a priori error estimates for the semidiscrete schemes.In this sequel,we present a posteriori error estimates for the semidiscrete and fully discrete approximations introduced in[9].The key tool employed to effect our analysis is the dispersive reconstruction devel-oped by Karakashian and Makridakis[20]for related discontinuous Galerkin methods.We conclude by providing a set of numerical experiments designed to validate the a posteriori theory and explore the effectivity of the resulting error indicators.

Study on generalized magneto-thermoelastic problems by FEM in time domain

This paper presents an investigation of temperature, displacement, stress, and induced magnetic field in a half space perfectly-conductive plate. Finite element equations regarding generalized magneto-thermoelasticity problems with two relaxation times （i.e., the G-L theory） are derived using the principle of virtual work. For avoiding numerical complication involved in inverse Laplace and Fourier transformation and low precision thereof, the equations are solved directly in time-domain. As a numerical example, the derived equation is used to investigate the generalized magneto-thermoelastic behavior of a semi-infinite plate under magnetic field and subjecting to a thermal shock loading. The results demonstrate that FEM can faithfully predict the deformation of the plate and the induced magnetic field, and most importantly can reveal the sophisticated second sound effect of heat conduction in two-dimensional generalized thermoelastic solids, which is usually difficult to model by routine transformation methods. A peak can be observed in the distribution of stress and induced front and the magnitude of magnetic field at the heat wave the peak decreases with time, which can not be obtained by transformation methods. The new method can also be used to study generalized piezo-thermoelastic problems.

Time Domain Simulation of Transient Responses of Very Large Floating Structures Under Unsteady External Loads

A time domain finite element method （FEM） for the analysis of transient elastic response of a very large floating structure （VLFS） subjected to arbitrary time-dependent external loads is presented. This method is developed directly in time domain and the hydrodynamic problem is formulated based on linear, inviscid and slightly compressible fluid theory and the structural response is analyzed on the thin plate assumption. The time domain finite element procedure herein is validated by comparing numerical results with available experimental data. Finally, the transient elastic response of a pontoon-type VLFS under the landing of an airplane is computed by the proposed time domain FEM. The time histories of the applied force and the position and velocity of an airplane during landing are modeled with data from a Boeing 747-400 jumbo jet.

Benchmarking of two three-dimensional numerical models in time/space domain to predict railway-induced ground vibrations

In the last 30 years,the scientific community has developed and proposed different models and numerical approaches for the study of vibrations induced by railway traffic.Most of them are formulated in the frequency/wave number domain and with a 2.5D approach.Three-dimensional numerical models formulated in the time/space domain are less frequently used,mainly due to their high computational cost.Notwithstanding,these models present very attractive characteristics,such as the possibility of considering nonlinear behaviors or the modelling of excess pore pressure and non-homogeneous and non-periodic geometries in the longitudinal direction of the track.In this study,two 3D numerical approaches formulated in the time/space domain are compared and experimentally validated.The first one consists of a finite element approach and the second one of a finite difference approach.The experimental validation in an actual case situated in Carregado(Portugal)shows an acceptable fitting between the numerical results and the actual measurements for both models.However,there are some differences among them.This study therefore includes some recommendations for their use in practical soil dynamics and geotechnical engineering.

Discontinuous-Galerkin-Based Analysis of Traffic Flow Model Connected with Multi-Agent Traffic Model

As the number of automobiles continues to increase year after year,the associated problem of traffic congestion has become a serious societal issue.Initiatives to mitigate this problem have considered methods for optimizing traffic volumes in wide-area road networks,and traffic-flow simulation has become a focus of interest as a technique for advance characterization of such strategies.Classes of models commonly used for traffic-flow simulations include microscopic models based on discrete vehicle representations,macroscopic models that describe entire traffic-flow systems in terms of average vehicle densities and velocities,and mesoscopic models and hybrid(or multiscale)models incorporating both microscopic and macroscopic features.Because traffic-flow simulations are designed to model traffic systems under a variety of conditions,their underlyingmodelsmust be capable of rapidly capturing the consequences of minor variations in operating environments.In other words,the computation speed of macroscopic models and the precise representation of microscopic models are needed simultaneously.Thus,in this study we propose a multiscale model that combines a microscopic model—for detailed analysis of subregions containing traffic congestion bottlenecks or other localized phenomena of interest-with a macroscopic model enabling simulation of wide target areas at a modest computational cost.In addition,to ensure analytical stability with robustness in the presence of discontinuities,we discretize our macroscopic model using a discontinuous Galerkin finite element method(DGFEM),while to conjoin microscopic and macroscopic models,we use a generating/absorbing sponge layer,a technique widely used for numerical analysis of long-wavelength phenomena in shallow water,to enable traffic-flow simulations with stable input and output regions.

起伏地表弹性波传播的间断Galerkin有限元数值模拟方法

间断Galerkin有限元法(DG-FEM)作为一种有效的高阶有限元法受到了国内外学者的广泛关注.本文基于任意高阶间断Galerkin有限元法对弹性波方程进行空间离散,并将离散后所得的非齐次线性常微分方程系统齐次化,最后结合针对齐次问题的强稳定性保持龙格库塔(SSP Runge-Kutta)算法,将DG-FEM推广至时间任意高阶精度.另外,借鉴近最佳匹配层(NPML)的思想,基于复频移(CFS)拉伸坐标变换推导了一种新的PML吸收边界条件(简称为CFS-NPML),该CFS-NPML能够与DG-FEM算法很好地结合,形成有效的起伏地表地震波传播数值模拟技术.数值试验结果表明,DG-FEM具有高阶精度,可以适应任意复杂起伏地表和复杂构造情况下的弹性波传播数值模拟.同时,CFS-NPML对包括面波等震相的人为边界反射都具有良好的吸收效果.

任意单元间断Galerkin有限元计算方法研究

基于龙格库塔间断Galerkin(RKDG)有限元法的构造思想,通过局部坐标变换,发展了非正交单元DG有限元计算方法;借鉴非结构网格有限体积隐式计算方法,发展了适应于DG有限元方法的隐式计算方法;借鉴一维和二维(三角形单元)DG有限元限制器构造方法,提出了非正交三棱柱单元限制器方法;利用上述方法数值模拟了球头和双椭球的高超声速粘性绕流,得到了清晰的流场结构,并得到了较好的压力和热流分布,表明该方法在复杂高超声速流动的数值模拟方面具有广阔的应用前景。

Navier-Stokes方程间断Galerkin有限元方法研究

通过引入全局提升算子和局部提升算子,发展了求解Navier-Stokes方程的间断Galerkin(discontinuousGalerkin,DG)有限元方法的一般框架,并在此框架下给出了几种典型黏性离散格式的具体表达形式.对局部提升算子的求解给出了详细的计算步骤.同时还给出了一种简单有效的计算方法来对物面边界进行高阶近似.为了能够对NS方程进行精度测试,采用对原始系统添加源项的方法构造精确解.二维Euler和NS系统的精度测试表明该方法达到了DG方法的理论精度.二维圆柱无黏绕流的计算结果表明关于物面边界的高阶近似方法能够保持DG方法原有的精度.卡门涡街数值模拟则进一步验证了该方法的正确性并且显示出DG方法较高的计算精度和分辨率.

高阶间断Galerkin方法求解三维欧拉方程的研究

在三维非结构网格上,对高阶间断Galerkin方法求解定常三维欧拉方程进行研究。文中使用Roe格式通量函数计算网格单元边界上的数值通量;时间方向采用显式Runge-Kutta方法推进;并引入激波探测器和斜率限制器技术,成功地抑制流场解在间断处的数值振荡。对M6机翼跨音速无粘流场进行数值模拟,结果表明:计算结果和实验值吻合较好,和同等精度的有限体积法相比,具有更低的数值耗散和更强的激波捕捉能力。

固体非傅立叶温度场的时域间断Galerkin有限元法

运用时域间断Galerkin有限元法[1],对高频非傅立叶热波动问题[2-3]进行分析。其主要特点是:取温度及温度的时间导数为基本未知量,对其分别进行3次Hermite插值和线性插值。在保证节点温度自动保持连续的基础上,温度的时间导数在离散时域存在间断。数值结果表明所提出的方法能够滤掉虚假的数值震荡,能够良好地模拟固体中的非傅立叶热波动行为。

一种自适应影响域半径无网格Galerkin法

针对某些力学问题的数值求解需要结点的局部加密,在采用背景积分网格积分方式的基础上,提出一种影响域半径随结点疏密程度而变化的自适应影响域半径无网格Galerkin法。在方法中,无网格结点与背景积分网格的结点重合,结点的影响域半径即可根据该结点周围的网格的最大边长来选取。算例显示,该文方法是可行而有效的。

基于局部间断Galerkin方法的p型有限元

将基于三变量能量原理的局部间断Galerkin方法 (localdiscontinuousGalerkin ,LDG)应用于 p型单元的构造 .该方法采用间断的单元试解 ,不需要满足普通有限元所必须的协调条件 ,就能使构造高阶的插值函数变得更加灵活和容易 .在此基础上 ,对应力和应变场采用Legendre正交多项式进行插值 ,避免了柔度矩阵的求逆过程 .数值算例表明这种方法构造出的 p型单元不仅升阶过程简单 ,而且具有较高的精度 .

时域间断Galerkin有限元法在激光热加工过程中应用

针对半无限体和薄膜结构受多种激光热源作用下的非傅里叶热传导过程,采用时域间断Galerkin有限元法进行数值仿真.其主要特点是在时域内对温度及其时间导数分别进行三次Hermite插值和线性插值.对于半无限激光热源热传导问题,其计算结果与解析解吻合良好.算例表明,时域间断Galerkin有限元法在高频激光脉冲问题中,没有虚假的数值振荡,具有广泛的工程应用性.