Communications on Applied Mathematics and Computation ›› 2022, Vol. 4 ›› Issue (1): 250-270.doi: 10.1007/s42967-020-00108-z

Previous Articles     Next Articles

Negative Norm Estimates for Arbitrary Lagrangian-Eulerian Discontinuous Galerkin Method for Nonlinear Hyperbolic Equations

Qi Tao1, Yan Xu2, Xiaozhou Li3   

  1. 1 Beijing Computational Science Research Center, Beijing 100193, China;
    2 School of Mathematical Sciences, University of Science and Technology of China, Hefei 230026, China;
    3 School of Mathematical Sciences, University of Electronic Science and Technology of China, Chengdu 611731, China
  • Received:2020-08-31 Revised:2020-11-10 Online:2022-03-20 Published:2022-03-01
  • Contact: Yan Xu, Qi Tao, Xiaozhou Li E-mail:yxu@ustc.edu.cn;taoqi@csrc.ac.cn;xiaozhouli@uestc.edu.cn
  • Supported by:
    Q. Tao:Research supported by the fellowship of China Postdoctoral Science Foundation, no:2020TQ0030. Y. Xu:Research supported by National Numerical Windtunnel Project NNW2019ZT4-B08, Science Challenge Project TZZT2019-A2.3, NSFC Grants 11722112, 12071455. X. Li:Research supported by NSFC Grant 11801062.

Abstract: In this paper, we present the negative norm estimates for the arbitrary Lagrangian-Eulerian discontinuous Galerkin (ALE-DG) method solving nonlinear hyperbolic equations with smooth solutions. The smoothness-increasing accuracy-conserving (SIAC) flter is a postprocessing technique to enhance the accuracy of the discontinuous Galerkin (DG) solutions. This work is the essential step to extend the SIAC flter to the moving mesh for nonlinear problems. By the post-processing theory, the negative norm estimates are vital to get the superconvergence error estimates of the solutions after post-processing in the L2 norm. Although the SIAC flter has been extended to nonuniform mesh, the analysis of fltered solutions on the nonuniform mesh is complicated. We prove superconvergence error estimates in the negative norm for the ALE-DG method on moving meshes. The main diffculties of the analysis are the terms in the ALE-DG scheme brought by the grid velocity feld, and the time-dependent function space. The mapping from time-dependent cells to reference cells is very crucial in the proof. The numerical results also confrm the theoretical proof.

Key words: Arbitrary Lagrangian-Eulerian discontinuous Galerkin method, Nonlinear hyperbolic equations, Negative norm estimates, Smoothness-increasing accuracyconserving flter, Post-processing

CLC Number: