当前位置: X-MOL 学术Numer. Algor. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A posteriori error estimations of the Petrov-Galerkin methods for fractional Helmholtz equations
Numerical Algorithms ( IF 1.7 ) Pub Date : 2021-06-28 , DOI: 10.1007/s11075-021-01147-0
Wenting Mao , Yanping Chen , Huasheng Wang

Here, we develop a Petrov-Galerkin spectral method for the fractional Helmholtz equations (FHEs) of even order ν = s + k,s ∈ (k − 1,k) and k. We define trial and test functions by related generalized Jacobi functions (GJFs). Moreover, we efficiently establish the well-posedness of problem and provide the rigorous priori error estimates. Furthermore, by auxiliary equation, we also obtain the super-approximation estimates. Notably, we propose a post-processing technique for the Petrov-Galerkin spectral method, and give the error estimates of corrected solution. In addition, we define a posteriori error estimators, and prove that they are asymptotically accurate. Finally, we demonstrate the sharpness of our error estimates by numerical experiments.



中文翻译:

分数阶亥姆霍兹方程的 Petrov-Galerkin 方法的后验误差估计

在这里,我们为偶数阶ν = s + k , s ∈ ( k − 1, k ) 和k的分数阶亥姆霍兹方程 (FHE) 开发了一种 Petrov-Galerkin 谱方法. 我们通过相关的广义雅可比函数 (GJF) 定义试验和测试函数。此外,我们有效地建立了问题的适定性并提供了严格的先验误差估计。此外,通过辅助方程,我们还获得了超近似估计。值得注意的是,我们为 Petrov-Galerkin 谱方法提出了一种后处理技术,并给出了修正解的误差估计。此外,我们定义了后验误差估计量,并证明它们是渐近准确的。最后,我们通过数值实验证明了我们的误差估计的锐度。

更新日期:2021-06-29
down
wechat
bug