毕卉 陈莎莎
摘 要:研究了基于偏迎风数值通量的四阶线性偏微分方程局部间断Galerkin方法的稳定性和误差估计问题。考虑在空间方向上,利用半离散形式的数值格式,通过使用广义Gauss-Radau投影,消除了数值通量产生的投影误差,利用Young不等式得到数值格式的最优误差估计。证明了当对流项选择偏迎风数值通量,方法的收敛阶为k+1阶。由于含有高阶空间导数的偏微分方程LDG方法的空间离散算子具有刚性,因此对于时间离散采用二阶隐式Crank-Nicolson方法,通过数值试验验证了理论分析结果的正确性。
关键词:四阶线性偏微分方程;局部间断Galerkin方法;误差估计;偏迎风通量;广义Gauss-Radau投影
DOI:10.15938/j.jhust.2021.04.022
中圖分类号:O29
文献标志码:A
文章编号:1007-2683(2021)04-0159-08
Abstract:This paper studies the stability and error estimates of the local discontinuous Galerkin method for fourth-order linear partial differential equations based on upwind-biased fluxes. Consider using the semi-discrete form of numerical format in the spatial direction and using the generalized Gauss-Radau projection, the projection error caused by the numerical flux is eliminated. The optimal error estimate of the numerical format is obtained by using Young inequality. It is proved that when the convective term is selected as the upwind-biased numerical fluxes, the convergence order of the method is order k+1. Because the spatial discrete operator of the partial differential equation LDG method with higher-order spatial derivatives is rigid, the second-order implicit Crank-Nicolson method is used for time dispersion, and the correctness of the theoretical analysis results is verified by numerical experiments.
Keywords:fourth-order linear PDEs; local discontinuous Galerkin methods; error estimates; upwind-biased fluxes; generalized Gauss-Radau projection
0 引 言
对流扩散方程是一类反映物质输运、分子扩散或黏性流体流动的数学模型,可以描述化学、流体力学、空气动力学等领域的众多物理现象,在天气预报、石油开采、半导体模拟等领域有着广泛的应用。因此,对流扩散方程的数值方法研究一直是偏微分方程数值解研究的重要课题之一。由于局部间断Galerkin(local discontinuous galerkin,简称LDG)方法具有良好的数值表现和数值实现的简便性,该方法己经成为求解高阶微分方程的热门方法之一。本文将求解以下的对流扩散方程,充分展示LDG方法的数值求解优势。
间断Galerkin有限元方法是由Reed和Hill[1]于1973年在求解中子运输方程时首次提出。Johnson和Pitkranta[2]将这个方法应用到标量线性双曲型方程上,并且研究了Lp范数意义下的误差估计问题。之后,Cockburn等在文[3-6]中针对双曲型守恒律方程提出了Runge-Kutta DG方法。由于Bassi和Rebay[7]应用DG方法成功地解决了可压缩的Navier-Stokes方程,Cockburn和Shu[8]受他们的启发,在解决对流扩散方程时第一次提出了局部间断Galerkin方法。LDG方法的主要思想是先把对流扩散方程化为等价的一阶偏微分方程组,再使用DG方法进行空间离散。局部间断Galerkin方法是DG方法的推广,用于求解含有高阶空间导数的偏微分方程。目前,LDG方法已经得到了广泛的发展和应用。Yan和Shu将LDG方法应用到三阶KdV方程[12],以及四阶和五阶偏微分方程[13]。Xu和Shu进一步将LDG推广到非线性波动方程[14-15]、Schrodinger方程[16]、Hunter-Saxton方程[18]及Surface diffusion和Willmore flow方程[19]等,更多见文[20]。如何选取合适的数值流通量来保证数值格式的稳定性是LDG方法的核心。
传统上,线性双曲方程DG方法的数值通量通常选择纯迎风数值通量。但是,对于复杂的系统或非线性问题,纯迎风数值通量很难构造。因此,研究更一般的数值通量(如偏迎风偏通量)是必要且重要的。最近,Meng等[21]研究了基于偏迎风通量线性守恒律方程DG方法的收敛性,证明了最优半离散DG方法的收敛阶为k+1阶。Cao等[22]基于偏迎风通量,研究了一维线性双曲型方程的DG方法超收敛性,发现DG解及其导数在一些特殊点处可以得到k+1阶和k+2阶超收敛。本文研究了基于偏迎风数值通量的线性四阶偏微分方程LDG方法的误差估计。
论文第2节给出了线性四阶方程的LDG方法,证明了基于偏迎风通量四阶线性偏微分方程LDG方法的收敛阶为k+1阶;在第3节中通过数值实验,验证了结果的正确性;在第4节中,给出了结论和未来的工作。
进而对方法的误差和收敛阶进行分析。在表1和表2中,时间步长选择τ=0.01h2;在表3和表4中,时间步长选择τ=0.01h4,计算终止时刻分别取为T=1及T=10。通过对收敛阶的计算发现,当数值流通量取为偏迎风通量时,对于不同的θ值,收敛阶可以达到k+1阶精度,特别是对于长效时间,如T=10时,收敛阶仍可达到k+1阶,验证了定理2的结论。
以上算例表明,当对流项数值通量选取偏迎风通量,扩散项数值流通量选择交替流通量,局部间断Galerkin有限元解对真解有较好的逼近效果,且具有較好的长效性,这为进一步研究该方法的稳定性提供了数值保障。同时通过对误差和收敛阶的计算,得到当使用初值的p*h投影和P1以及P2多项式时在不同时刻的收敛阶均可达到k+1阶,验证了本文对四阶线性对流扩散方程的LDG方法的误差估计结果。
3 结 论
本文讨论了四阶线性对流扩散方程LDG方法的稳定性以及误差估计问题。证明了当对流项选择偏迎风通量,扩散项选择交错流通量时,LDG方法在Pk(k≥1)分片多项式有限元空间中的误差估计的阶为k+1。通过数值实验,验证了误差估计的理论分析是正确的。接下来将进一步研究高阶方程基于偏迎风数值流通量的数值稳定性问题,从理论上分析求解高阶微分方程LDG方法的数值特性。
参 考 文 献:
[1] REED W H, HILL T R. Triangular Mesh Methods for the Neutron Transport Equation[R]. Los Alamos Report LA-UR-73-479:1973.
[2] PitkRanta, C Johnson. An Analysis of the Discontinuous Galerkin Method for a Scalar Hyperbolic Equation[J]. Mathematics of Computation, 1986, 46(173):1.
[3] COCKBURN B, LIN S Y & SHU C W. TVB Runge-Kutta Local Projection Discontinuous Galerkin Finite Element method for Conservation Laws III:One-dimensional Systems[J]. Journal of Computational Physics, 1989, 84(1):903.
[4] COCKBURN B, HOU C W, SHU C W. The Runge-kutta Local Projection Discontinuous Galerkin Finite Element Method for Conservation Laws. iv:the Multidimensional Case. Mathematics of Computation,1990, 54(190):545.
[5] COCKBURN B, SHU C W. TVB Runge-Kutta Local Projection Discontinuous Galerkin Finite Element Method for Conservation Laws II:General Framework[J]. Mathematics of Computation, 1989, 52(186):411.
[6] COCKBURN B, SHU C W. The Runge-Kutta Discontinuous Galerkin Method for Conservation Laws V[J]. Journal of Computational Physics, 1997, 141(2):199.
[7] BASSI F, REBAY S. A high-order Accurate Discontinuous Finite Element Method for the Numerical Solution of the Compressible Navier-Stokes Equations[J]. Journal of Computational Physics, 1997, 131(2):267.
[8] SHU C W. The Local Discontinuous Galerkin Method for Time-Dependent Convection-Diffusion Systems[J]. SIAM Journal on Numerical Analysis, 1998, 35(6):2440.
[9] CHENG Y, MENG X, ZHANG Q. Application of Generalized Gauss-Radau Projections for the Local Discontinuous Galerkin Method for Linear Convection-diffusion Equations[J]. Mathematics of Computation, 2016, 86(305):1233.
[10]CHENG Y, ZHANG Q. Local Analysis of the Local Discontinuous Galerkin Method with Generalized Alternating Numerical Flux for One-Dimensional Singularly Perturbed Problem[J]. Journal of Scientific Computing, 2017, 72(2):792.
[11]CHENG Y, ZHANG Q, Wang H J. Local Analysis of the Local Discontinuous Galerkin Method with Generalized Alternating Numerical Flux for Two-Dimensional Singularly Perturbed Problem.Int.J. Numer. Anal. Mod. 2018, 15:785.
[12]YAN J, SHU C W. A Local Discontinuous Galerkin Method for KdV Type Equations[J]. SIAM Journal on Numerical Analysis, 2002, 40(2):769.
[13]YAN J, SHU C W. Local Discontinuous Galerkin Methods for Partial Differential Equations with Higher Order Derivatives[J]. Journal of Scientific Computing, 2002, 17(1):27.
[14]SHU C W, XU Y. Local Discontinuous Galerkin Methods For Three Classes Of Nonlinear Wave Equations[J]. Journal of Computational Mathematics, 2004, 22(2):250.
[15]XU Y, SHU C W. Local Discontinuous Galerkin Methods for Two Classes of Two-dimensional Nonlinear Wave Equations[J]. Physica D, 2005, 208(1-2):21.
[16]XU Y, SHU C W. Local Discontinuous Galerkin Methods for Nonlinear Schrodinger Equations[J]. Journal of Computational Physics, 2005, 20S(1):72.
[17]SHU C W. A Local Discontinuous Galerkin Method for the Camassa-Holm Equation[J]. SIAM Journal on Numerical Analysis, 2008, 46(4):1998.
[18]XU Y, SHU C W. Local Discontinuous Galerkin Method for the Hunter-Saxton Equation and Its Zero-Viscosity and Zero-Dispersion Limits[J]. SIAM Journal on Scientific Computing, 2009, 31(2):1249.
[19]XU Y, SHU C W. Local Discontinuous Galerkin Method for Surface Diffusion and Willmore Flow of Graphs[J]. Journal of Scientific Computing, 2009, 40(1/3):375.
[20]XU Y, SHU C W. Local Discontinuous Galerkin Methods for High-order Time-dependent Partial Differential Equations. Commun. Comput. Phys., 2010, 7:1.
[21]MENG X, SHU C W, Wu B Y. Optimal Error Estimates for Discontinuous Galerkin Methods Based on Upwind-biased Fluxes for Linear Hyperbolic Equations[J]. Mathematics of Computation, 2016, 85:1225.
[22]CAO W, LI D, YANG Y et al. Superconvergence of Discontinuous Galerkin Methods Based on Upwind-biased Fluxes for 1D Linear Hyperbolic Equations[J]. ESAIM:Mathematical Modelling and Numerical Analysis, 2017, 51(2):467.
[23]XU Y, SHU C W. Optimal Error Estimates of the Semidiscrete Local Discontinuous Galerkin Methods for High Order Wave Equations[J]. SIAM Journal on Numerical Analysis, 2012, 50:79.
[24]毕卉, 钱琛庚. 显式Runge-Kutta局部间断Galerkin方法的稳定性分析[J]. 哈尔滨理工大学学报, 2017, 22(6):109.
BI Hui, QIAN Chengeng. Stability Analysis of the Explicit Runge-kutta Local Discontinuous Galerkin Method[J].Journal of Harbin University of Scince and Technology, 2017,22(6):109.
(編辑:温泽宇)