Moving mesh method with variational multiscale finite element method for convection-diffusion-reaction equations
被引:0
|
作者:
Zhang, Xiaohua
论文数: 0引用数: 0
h-index: 0
机构:
Yunnan Normal Univ, Sch Math, Kunming 650500, Yunnan, Peoples R China
Yunnan Normal Univ, Yunnan Key Lab Modern Analyt Math & Applicat, Kunming 650500, Yunnan, Peoples R ChinaYunnan Normal Univ, Sch Math, Kunming 650500, Yunnan, Peoples R China
Zhang, Xiaohua
[1
,2
]
Xu, Xinmeng
论文数: 0引用数: 0
h-index: 0
机构:
China Three Gorges Univ, Three Gorges Math Res Ctr, Yichang 443002, Hubei, Peoples R ChinaYunnan Normal Univ, Sch Math, Kunming 650500, Yunnan, Peoples R China
Xu, Xinmeng
[3
]
机构:
[1] Yunnan Normal Univ, Sch Math, Kunming 650500, Yunnan, Peoples R China
[2] Yunnan Normal Univ, Yunnan Key Lab Modern Analyt Math & Applicat, Kunming 650500, Yunnan, Peoples R China
[3] China Three Gorges Univ, Three Gorges Math Res Ctr, Yichang 443002, Hubei, Peoples R China
The solutions tend to have large gradients, discontinuities, or sharp layers for convection-dominated convection-diffusion-reaction equations. Many studies have demonstrated the advantages of moving mesh and variational multiscale finite element method (VMFEM) in solving such problems. In this paper, we propose a new approach called the moving mesh variational multiscale finite element method (MM-VMFEM), which combines the benefits of both moving mesh and variational multiscale methods for solving convection-diffusion-reaction equations with dominant convection effects. The moving mesh method decouples the mesh equations from the underlying partial differential equations (PDEs) and constructs a connection between the physical and logical domains using harmonic mapping. This moving mesh method can change only the underlying PDEs algorithm partly when solving different differential equations and can keep the mesh topology unchanged after multiple numerical integration. The VMFEM decomposes the unknown scalar and test functions into coarse- and fine-scales, and the use of bubble functions to automatically obtain the stabilization parameters allows the VMFEM to be naturally combined with mesh adaptivity. The effectiveness of the proposed MM-VMFEM is verified by four numerical examples. The experimental results show that the MM-VMFEM can improve the computational accuracy and reduce the numerical spurious oscillations for convection-dominated problems.
机构:
Chongqing Jiaotong Univ, Sch Sci, Chongqing 400074, Peoples R China
Beijing Computat Sci Res Ctr, Beijing 100084, Peoples R ChinaChongqing Jiaotong Univ, Sch Sci, Chongqing 400074, Peoples R China
Du ShaoHong
Xie XiaoPing
论文数: 0引用数: 0
h-index: 0
机构:
Sichuan Univ, Sch Math, Chengdu 610064, Peoples R ChinaChongqing Jiaotong Univ, Sch Sci, Chongqing 400074, Peoples R China
机构:
Nanjing Normal Univ, Sch Math Sci, Jiangsu Key Lab NSLSCS, Nanjing 210023, Jiangsu, Peoples R China
Shihezi Univ, Coll Sci, Dept Math, Shihezi 832003, Peoples R ChinaNanjing Normal Univ, Sch Math Sci, Jiangsu Key Lab NSLSCS, Nanjing 210023, Jiangsu, Peoples R China
Qian, Lingzhi
Cai, Huiping
论文数: 0引用数: 0
h-index: 0
机构:
Shihezi Univ, Coll Sci, Dept Math, Shihezi 832003, Peoples R ChinaNanjing Normal Univ, Sch Math Sci, Jiangsu Key Lab NSLSCS, Nanjing 210023, Jiangsu, Peoples R China
Cai, Huiping
Guo, Rui
论文数: 0引用数: 0
h-index: 0
机构:
Shihezi Univ, Coll Sci, Dept Math, Shihezi 832003, Peoples R ChinaNanjing Normal Univ, Sch Math Sci, Jiangsu Key Lab NSLSCS, Nanjing 210023, Jiangsu, Peoples R China
Guo, Rui
Feng, Xinlong
论文数: 0引用数: 0
h-index: 0
机构:
Xinjiang Univ, Coll Math & Syst Sci, Urumqi 830046, Peoples R ChinaNanjing Normal Univ, Sch Math Sci, Jiangsu Key Lab NSLSCS, Nanjing 210023, Jiangsu, Peoples R China