Superconvergence analysis of BDF-Galerkin FEM for nonlinear Schrodinger equation

被引:18
作者
Wang, Junjun [1 ]
Li, Meng [2 ]
Zhang, Yu [1 ]
机构
[1] Pingdingshan Univ, Sch Math & Stat, Pingdingshan 467000, Peoples R China
[2] Zhengzhou Univ, Sch Math & Stat, Zhengzhou 450001, Peoples R China
基金
中国博士后科学基金; 中国国家自然科学基金;
关键词
Nonlinear Schrodinger equation; BDF-FEM; Conservation; Prior estimate; Unconditional superconvergent result; FINITE-ELEMENT-METHOD; DIFFERENCE-SCHEMES; 4TH-ORDER COMPACT; ERROR ANALYSIS; DYNAMICS;
D O I
10.1007/s11075-021-01111-y
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
A nonlinear iteration scheme for nonlinear Schrodinger equation with 2-step backward differential formula (BDF) finite element method (FEM) is proposed. Energy stability is testified for the constructed scheme, which leads to the boundedness of parallel to U-h(n)parallel to(0) and root tau parallel to del U-h(n)parallel to(0). Auxiliary equation known as a time-discrete system is constructed to get rid of the restriction of tau. The solutions of the time-discrete equation in H-1-norm is deduced. On the one hand, root tau parallel to del U-n parallel to(0) reduces to the temporal error in H-2-norm. On the other hand, with the help of the boundedness about root tau parallel to del U-h(n)parallel to(0), the unconditional optimal estimate for spatial error is derived. Without any restriction of the time step, parallel to U-h(n)parallel to(0,infinity) is bounded through surmounting the difficulty of nonlinear term. By taking difference between two time levels n and n - 1 of the error equation, an unconditional superconvergence estimate is derived. At last, global superconvergence result is achieved through the known interpolated postprocessing technique. Here, tau is the time step, and U-n and U-h(n) denote the solutions of the time-discrete system and the finite element approximation equation, respectively. Furthermore, by introducing modified mass and energy functions, the numerical scheme is proved to preserve the total mass and energy in the discrete senses. Finally, numerical results are given to support the theoretical analysis.
引用
收藏
页码:195 / 222
页数:28
相关论文
共 30 条
[1]   ON FULLY DISCRETE GALERKIN METHODS OF 2ND-ORDER TEMPORAL ACCURACY FOR THE NONLINEAR SCHRODINGER-EQUATION [J].
AKRIVIS, GD ;
DOUGALIS, VA ;
KARAKASHIAN, OA .
NUMERISCHE MATHEMATIK, 1991, 59 (01) :31-53
[2]   FINITE-DIFFERENCE DISCRETIZATION OF THE CUBIC SCHRODINGER-EQUATION [J].
AKRIVIS, GD .
IMA JOURNAL OF NUMERICAL ANALYSIS, 1993, 13 (01) :115-124
[3]   UNIFORM ERROR ESTIMATES OF FINITE DIFFERENCE METHODS FOR THE NONLINEAR SCHRODINGER EQUATION WITH WAVE OPERATOR [J].
Bao, Weizhu ;
Cai, Yongyong .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 2012, 50 (02) :492-521
[4]   Dynamics of rotating Bose-Einstein condensates and its efficient and accurate numerical computation [J].
Bao, WZ ;
Du, Q ;
Zhang, YZ .
SIAM JOURNAL ON APPLIED MATHEMATICS, 2006, 66 (03) :758-786
[5]   Unconditional optimal error estimates for BDF2-FEM for a nonlinear Schrodinger equation [J].
Cai, Wentao ;
Li, Jian ;
Chen, Zhangxin .
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, 2018, 331 :23-41
[6]   Conservative super-convergent and hybrid discontinuous Galerkin methods applied to nonlinear Schrodinger equations [J].
Castillo, Paul ;
Gomez, Sergio .
APPLIED MATHEMATICS AND COMPUTATION, 2020, 371
[7]   Difference schemes for solving the generalized nonlinear Schrodinger equation [J].
Chang, QS ;
Jia, EH ;
Sun, W .
JOURNAL OF COMPUTATIONAL PHYSICS, 1999, 148 (02) :397-415
[8]   A compact split-step finite difference method for solving the nonlinear Schrodinger equations with constant and variable coefficients [J].
Dehghan, Mehdi ;
Taleei, Ameneh .
COMPUTER PHYSICS COMMUNICATIONS, 2010, 181 (01) :43-51
[9]   Unconditional Optimal Error Estimates of BDF-Galerkin FEMs for Nonlinear Thermistor Equations [J].
Gao, Huadong .
JOURNAL OF SCIENTIFIC COMPUTING, 2016, 66 (02) :504-527
[10]   A conservative Fourier pseudo-spectral method for the nonlinear Schrodinger equation [J].
Gong, Yuezheng ;
Wang, Qi ;
Wang, Yushun ;
Cai, Jiaxiang .
JOURNAL OF COMPUTATIONAL PHYSICS, 2017, 328 :354-370