Computing Residual Diffusivity by Adaptive Basis Learning via Super-Resolution Deep Neural Networks
Computing Residual Diffusivity by Adaptive Basis Learning via Super-Resolution Deep Neural Networks
复制标题
通过超分辨率深度神经网络的自适应基础学习计算残余扩散率
DOI:
10.1007/978-3-030-38364-0_25
复制
发表时间:
2020
期刊:
影响因子:
--
通讯作者:
Yu, Yifeng
中科院分区:
文献类型:
--
作者:
Lyu, Jiancheng;Xin, Jack;Yu, Yifeng
It is expensive to compute residual diffusivity in chaotic in-compressible flows by solving advection-diffusion equation due to the formation of sharp internal layers in the advection dominated regime. Proper orthogonal decomposition (POD) is a classical method to construct a small number of adaptive orthogonal basis vectors for low cost computation based on snapshots of fully resolved solutions at a particular molecular diffusivity. The quality of POD basis deteriorates if it is applied to. To improve POD, we adapt a super-resolution generative adversarial deep neural network (SRGAN) to train a nonlinear mapping based on snapshot data at two values of. The mapping models the sharpening effect on internal layers asbecomes smaller. We show through numerical experiments that after applying such a mapping to snapshots, the prediction accuracy of residual diffusivity improves considerably that of the standard POD.
登录
查看更多内容
DOI:
10.1137/18m1165219
发表时间:
2017
期刊:
SIAM J. Numer. Anal.
影响因子:
--
作者:
Zhongjian Wang;J. Xin;Zhiwen Zhang
通讯作者:
Zhiwen Zhang
DOI:
--
发表时间:
2013
期刊:
影响因子:
--
作者:
J. Xin;Yifeng Yu
通讯作者:
Yifeng Yu
影响因子:
2.5
作者:
S. Heinze
通讯作者:
S. Heinze
DOI:
10.1103/physreva.43.774
发表时间:
1991
期刊:
Physical review. A, Atomic, molecular, and optical physics
影响因子:
--
作者:
Camassa;Wiggins
通讯作者:
Wiggins