Discovering variable fractional orders of advection-dispersion equations from field data using multi-fidelity Bayesian optimization

Discovering variable fractional orders of advection-dispersion equations from field data using multi-fidelity Bayesian optimization
复制标题

使用多保真贝叶斯优化从现场数据发现平流弥散方程的可变分数阶

DOI:
10.1016/j.jcp.2017.07.052
复制
发表时间:
2017-11
影响因子:
4.1
通讯作者:
George Em Karniadakis
George Em Karniadakis
中科院分区:
物理与天体物理2区
文献类型:
--
作者:
Guofei Pang;Paris Perdikaris;Wei Cai;George Em Karniadakis

文献摘要

参考文献

被引文献

相似文献

分数平流扩散方程(FADE)可以准确地描述地下水中溶质的运移,但其分数阶必须先验确定。在这里,我们采用多保真贝叶斯优化来获得各种条件下的分数阶,并且与之前发布的数据相比,我们获得了更准确的结果。此外,本方法非常有效,因为我们使用不同级别的分辨率来构建随机代理模型并量化其不确定性。我们考虑两种不同的问题设置。在第一个设置中,考虑合成数据和现场数据,我们获得一维 FADE 的可变分数阶。在第二个设置中,我们使用合成数据识别二维 FADE 的恒定分数阶。我们采用使用两级和三级高斯过程回归模型的多分辨率模拟来构建代理。
The fractional advection–dispersion equation (FADE) can describe accurately the solute transport in groundwater but its fractional order has to be determineda priori. Here, we employ multi-fidelity Bayesian optimization to obtain the fractional order under various conditions, and we obtain more accurate results compared to previously published data. Moreover, the present method is very efficient as we use different levels of resolution to construct a stochastic surrogate model and quantify its uncertainty. We consider two different problem set ups. In the first set up, we obtain variable fractional orders of one-dimensional FADE, considering both synthetic and field data. In the second set up, we identify constant fractional orders of two-dimensional FADE using synthetic data. We employ multi-resolution simulations using two-level and three-level Gaussian process regression models to construct the surrogates.
DOI: 10.1137/140965144
发表时间: 2015-06
期刊: SIAM J. Sci. Comput.
影响因子: --
作者:
Liang Yan;Ling Guo
通讯作者: Liang Yan;Ling Guo
DOI: 10.1198/tech.2003.s770
发表时间: 2003-08
期刊: Technometrics
影响因子: 2.5
作者:
E. Ziegel
通讯作者: E. Ziegel
DOI: 10.1103/physreve.59.5026
发表时间: 1999-05
期刊: Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics
影响因子: --
作者:
M. Meerschaert;D. Benson;Boris Bäumer
通讯作者: M. Meerschaert;D. Benson;Boris Bäumer
DOI: --
发表时间: 2004-12
期刊: --
影响因子: --
作者:
A. Schwaighofer;Volker Tresp;Kai Yu
通讯作者: A. Schwaighofer;Volker Tresp;Kai Yu
DOI: 10.1016/j.advwatres.2009.01.008
发表时间: 2009-04
影响因子: 4.7
作者:
Yong Zhang;D. Benson;D. Reeves
通讯作者: Yong Zhang;D. Benson;D. Reeves