FC-based shock-dynamics solver with neural-network localized artificial-viscosity assignment

FC-based shock-dynamics solver with neural-network localized artificial-viscosity assignment
复制标题

基于 FC 的冲击动力学求解器,具有神经网络局部人工粘度分配

DOI:
10.1016/j.jcpx.2022.100110
复制
发表时间:
2022
影响因子:
--
通讯作者:
Leibovici, Daniel V.
Leibovici, Daniel V.
中科院分区:
--
文献类型:
--
作者:
Bruno, Oscar P.;Hesthaven, Jan S.;Leibovici, Daniel V.

文献摘要

参考文献

相似文献

本文给出了任意边界条件下非周期区域内非线性守恒律数值解的谱格式。该方法依赖于非周期函数的频谱表示的傅里叶连续(FC)方法,以及通过激波检测神经网络(SDNN)产生的平滑局部人工粘性赋值。与以前的激波捕捉方案和人工粘性技术一样,FC-SDNN组合策略有效地控制了不连续区域内的虚假振荡。由于它使用了局部但光滑的人工粘性项,其支持仅限于流动不连续点附近,因此该算法具有谱精度和远离流动不连续的低耗散,并且在这些区域,它产生平滑的数值解--这从水平集线上基本上没有虚假振荡得到了证明。FC-SDNN粘度赋值不需要使用依赖于问题的算法参数,与其他方法相比,总体耗散显著降低,包括先前的熵粘性方法的傅里叶谱版本。用一维和二维非周期空间域中线性平流方程、Burgers方程和Euler方程的各种数值结果说明了该算法的特点。
This paper presents a spectral scheme for the numerical solution of nonlinear conservation laws innon-periodic domains under arbitrary boundary conditions. The approach relies on the use of the Fourier Continuation (FC) method for spectral representation of non-periodic functions in conjunction with smooth localized artificial viscosity assignments produced by means of a Shock-Detecting Neural Network (SDNN). Like previous shock capturing schemes and artificial viscosity techniques, the combined FC-SDNN strategy effectively controls spurious oscillations in the proximity of discontinuities. Thanks to its use of alocalized but smooth artificial viscosity term, whose support is restricted to a vicinity of flow-discontinuity points, the algorithm enjoys spectral accuracy and low dissipation away from flow discontinuities, and, in such regions, it produces smooth numerical solutions—as evidenced by an essential absence of spurious oscillations in level set lines. The FC-SDNN viscosity assignment, which does not require use of problem-dependent algorithmic parameters, induces a significantly lower overall dissipation than other methods, including the Fourier-spectral versions of the previous entropy viscosity method. The character of the proposed algorithm is illustrated with a variety of numerical results for the linear advection, Burgers and Euler equations in one and two-dimensional non-periodic spatial domains.
DOI: 10.1137/s1064827594273948
发表时间: 1997-09
期刊: SIAM J. Sci. Comput.
影响因子: --
作者:
D. Pathria
通讯作者: D. Pathria
DOI: 10.1016/j.jcp.2019.02.033
发表时间: 2019-08
期刊: J. Comput. Phys.
影响因子: --
作者:
O. Bruno;Max Cubillos;Edwin Jimenez
通讯作者: O. Bruno;Max Cubillos;Edwin Jimenez
DOI: 10.1016/j.jcp.2011.08.024
发表时间: 2011-10
期刊: J. Comput. Phys.
影响因子: --
作者:
Khosro Shahbazi;Nathan Albin;O. Bruno;J. Hesthaven
通讯作者: Khosro Shahbazi;Nathan Albin;O. Bruno;J. Hesthaven
DOI: 10.1007/s10915-017-0613-6
发表时间: 2017-03
影响因子: 2.5
作者:
Adeline Kornelus;D. Appelö
通讯作者: Adeline Kornelus;D. Appelö
DOI: 10.1016/j.jcp.2019.06.023
发表时间: 2019-10
期刊: J. Comput. Phys.
影响因子: --
作者:
A. Mazaheri;Chi-Wang Shu;V. Perrier
通讯作者: A. Mazaheri;Chi-Wang Shu;V. Perrier