摘""要:"由于二阶奇异摄动问题的解具有边界层,用传统的数值方法很难得到理想的逼近效果. 为了高效地求解此类问题,对区间做剖分,分别使用移位的映射Legendre多项式和传统Legendre多项式作为基函数,提出新型混合谱元法. 该方法能够更好地模拟解在边界层附近的奇异行为. 数值结果验证了所提方法的高效性.
关键词:"谱元法;"奇异摄动;"映射Legendre多项式;"边界层
中图分类号:"O 241.82 """文献标志码:"A """文章编号:"1000-5137(2024)04-0457-08
New spectral element method for the singular perturbation problem of second order
LIU Zhipeng TAO Dongya
(1.School of Economics and Trade,"Changzhou Technical Institute of Tourism amp; Commerce,"Changzhou 213032,"Jiangsu,"China;"2.School of Mathematics and Statistics,"Jiangsu Normal University,"Xuzhou 221116,"Jiangsu,"China)
Abstract:"As the solution of singular perturbation problem of second order has a boundary layer,"it is difficult to obtain ideal approximation results for traditional numerical methods. In order to solve these problems efficiently,"the interval is divided and a new hybrid spectral element method is proposed by using the shifted mapped Legendre polynomials and the traditional Legendre polynomials as the basis functions. The new method can be used to better simulate the singular behavior of solutions near the boundary layer. Numerical results verify the effectiveness of the proposed method.
Key words:"spectral element method;"singular perturbation;"mapped Legendre polynomial;"boundary layer
0 "引"言
奇异摄动问题在流体动力学、最优控制及化学反应等领域中有着广泛的应用[1-2]. 考虑如下二阶奇异摄动问题:
谱方法是一种高精度算法,以整体光滑的正交多项式组合作为基底来逼近问题的解. 相比于有限差分法和有限元法,对于充分光滑的解,谱方法只需要很少的自由度就能提供非常精确的逼近结果[11-16]. 因此,谱方法在统计物理、流体力学、数值天气预报和海洋工程等许多领域中有着广泛的应用. 对于问题(1),谱方法主要有两种途径来处理. 一种方法是对原方程做变换,"例如:TANG等[17]使用合适的坐标变换将原方程的奇性减弱,然后对新方程使用传统的Chebyshev拟谱方法进行求解,取得了很好的逼近效果;"为了提高边界层附近配置点的稠密度,LIU等[4,18]使用奇异变换,更好地模拟了真解在边界层附近的奇异行为,但是变换后的方程更复杂,分析相应的理论也更困难. 另一种方法是将传统多项式做变换,构造一类新的基函数[19-22]来逼近局部振荡的解. 此种做法的好处在于新的基函数及其相应逼近结果一旦建立,就能被应用到一大类相似的问题中去. SHEN等[23]介绍了三种常规变换和相应的映射Legendre多项式,建立了相应的逼近结果,并且推导出基于映射参数上界的最优误差估计,在此工作基础上,构造了经过奇异变换后的映射Jacobi多项式,"以更少的自由度获得了更高精度的数值解[24].
1 "二阶奇异问题的新的谱元方法
本章将引入移位的传统Legendre多项式和映射Legendre多项式,然后对于二阶奇异摄动问题(1)提出新型的混合谱元方法.
1.1 准备工作
1.2 移位的传统Legendre多项式
1.3 移位的映射Legendre多项式
1.4 全局基函数
1.5 问题(1)的新型混合谱元法
2 "数值算例
3 "结语
本文介绍了数值求解二阶奇异摄动问题的新型混合谱元法,在首尾区间使用移位的映射Legendre多项式作为基函数,在其他区间使用移位的传统Legendre多项式作为基函数. 数值结果表明:相比于已有方法,该方法能够有效地刻画解在边界层附近的奇异行为,在数值误差的精度上有明显的优势. 在今后的工作中,可以通过缩短步长和增加多项式次数来获得理想的逼近效果.
参考文献:
[1] BECKETT G,"MACKENZIE J A. Convergence analysis of finite difference approximations on equidistributed grids to a singularly perturbed boundary value problem [J]. Applied Numerical Mathematics,"2000,"35:87-109.
[2] GIE G M,"JUNG C Y,"TEMAM R. Recent progresses in boundary layer theory [J]. Discrete and Continuous Dynamical Systems,"2016,"36:2521-2583.
[3] HENNER E,"WILHELM H. A new method of stabilization for singular perturbation problems with spectral methods [J]. SIAM Journal on Numerical Analysis,"1992,"29:107-122.
[4] LIU W B ,"SHEN J. A new efficient spectral Galerkin method for singular perturbation problems [J]. Journal of Scientific Computuing,"1996,"11:411-437.
[5] YUN D F,"WEN Z H,"HON Y C. Adaptive least squares finite integration method for higher-dimensional singular perturbation problems with multiple boundary layers [J]. Applied Mathematics and Computation,"2015,"271:232-250.
[6] GUO W,"STYNES M. Pointwise error estimates for a streamline diffusion scheme on a Shishkin mesh for a convection-diffusion problem [J]. IMA Journal of Numerical Analysis,"1997,"17:29-59.
[7] LUBUMA J M S,"PATIDAR K C. Non-standard methods for singularly perturbed problems possessing oscillatory/layer solutions [J]. Applied Mathematics and Computation,"2007,"187:1147-1160.
[8] SCHWAB C,"SURI M. The p and hp versions of the finite element method for problems with boundary layers [J]. Mathematics of Computation,"1996,"65:1403-1429.
[9] KADALBAJOO M K,"ARORA P,"GUPTA V. Collocation method using artificial viscosity for solving stiff singularly perturbed turning point problem having twin boundary layers [J]. Computers amp; Mathematics with Applications,"2011,"61:1595-1607.
[10] RAO S C S,"KUMAR M. Exponential B-spline collocation method for self-adjoint singularly perturbed boundary value problems [J]. Applied Numerical Mathematics,"2008,"58:1572-1581.
[11] GOTTLIEB D,"ORSZAG S A. Numerical Analysis of Spectral Methods:"Theory and Applications [M]. Philadelphia:"CBMS-NSF Regional Conference Series in Applied Mathematics,"1977.
[12] GUO B Y. Jacobi approximations in certain Hilbert spaces and their applications to singular differential equations [J]. Journal of Mathematical Analysis and Applications,"2000,"243:373-408.
[13] GUO B Y,"ZHANG C. Generalized Hermite spectral method matching different algebraic decay at infinities [J]. Journal of Scientific Computuing,"2015,"65:648-671.
[14] SHEN J. Direct solvers of second- and fourth-order equations using Legendre polynomials [J]. SIAM Journal on Scientific Computing,"1994,"15:1489-1505.
[15] SHEN J,"TANG T. Spectral and High-order Methods with Applications [M]. Beijing:"Science Press,"2006.
[16] SHEN J,"TANG T,"WANG L L. Algorithms,"Analysis and Applications [M]. Heidelberg:"Springer Series in Computational Mathematics,"2011.
[17] TANG T,"TRUMMER M R. Boundary layer resolving pseudospectral methods for singular perturbation problems [J]. SIAM Journal on Scientific Computing,"1996,"17:430-438.
[18] LIU W B,"TANG T. Error analysis for a Galerkin-spectral method with coordinate transformation for solving singularly perturbed problems [J]. Applied Numerical Mathematics,"2001,"38:315-345.
[19] CHEN S,"SHEN J. Enriched spectral methods and applications to problems with weakly singular solutions [J]. Journal of Scientific Computing,"2018,"77:1468-1489.
[20] HONG Y,"JUNG C Y. Enriched spectral method for stiff convection-dominated equations [J]. Journal of Scientific Computing,"2018,"74:1325-1346.
[21] SHEN J,"WANG Y W. Müntz-Galerkin methods and applications to mixed Dirichlet-Neumann boundary value problems [J]. SIAM Journal on Scientific Computing,"2016,"38:A2357-A2381.
[22] SUN T,"LIU R Q,"WANG L L. Generalised Müntz spectral Galerkin methods for singularly perturbed fractional differential equations [J]. East Asian Journal on Applied Mathematics,"2018,"8:611-633.
[23] SHEN J,"WANG L L. Error analysis for mapped Legendre spectral and pseudospectral methods [J]. SIAM Journal on Numerical Analysis,"2004,"42:326-349.
[24] WANG L L,"SHEN J. Error analysis for mapped Jacobi spectral methods [J]. Journal of Scientific Computing,"2005,"24:183-218.
(责任编辑:包震宇)