Learning macroscopic parameters in nonlinear multiscale simulations using nonlocal multicontinua upscaling techniques

Learning macroscopic parameters in nonlinear multiscale simulations using nonlocal multicontinua upscaling techniques
复制标题

使用非局部多连续放大技术学习非线性多尺度模拟中的宏观参数

DOI:
10.1016/j.jcp.2020.109323
复制
发表时间:
2020
影响因子:
4.1
通讯作者:
Wheeler, Mary
Wheeler, Mary
中科院分区:
物理与天体物理2区
文献类型:
--
作者:
Vasilyeva, Maria;Leung, Wing T.;Chung, Eric T.;Efendiev, Yalchin;Wheeler, Mary

文献摘要

参考文献

被引文献

相似文献

在这项工作中,我们提出了一种新的基于机器学习算法的非局部非线性粗网格逼近。我们考虑非均质和裂隙多孔介质中的非饱和和两相渗流问题,其中的数学模型被表示为一般的多连续介质模型。我们使用有限体积法和嵌入的离散裂缝模型构造了精细网格近似。这些复杂的非线性系统的宏观模型需要非局部多连续体方法,这是在早期工作[8]中发展起来的。这些严格的技术需要复杂的局部计算,这涉及到解决受约束的过采样区域中的局部问题。这些局部问题的解可以通过在多个输入参数(边界项和源项)的粗(过采样)区域上求解原始问题并计算由非线性非局部多连续体方法得到的有效性质来代替。有效的性质依赖于许多变量(过采样区域和连续体的数目),因此它们的计算需要某种类型的机器学习技术。在本文中,我们的贡献是两方面的。首先,我们介绍了宏观模型,并讨论了如何使用深度学习算法有效地计算宏观参数。该方法可以被认为是局部机器学习,补充了我们以前的全局机器学习方法[36],[35]。我们考虑了一种粗网格近似,它使用两种提升技术,其中单阶段提升透过率和使用机器学习算法的非局部非线性提升透过率。给出了非均质和裂隙多孔介质中两个模型问题的计算结果,表明该方法具有较高的精度,并提供了快速的粗网格计算。
In this work, we present a novel nonlocal nonlinear coarse grid approximation using a machine learning algorithm. We consider unsaturated and two-phase flow problems in heterogeneous and fractured porous media, where mathematical models are formulated as general multicontinuum models. We construct a fine grid approximation using the finite volume method and embedded discrete fracture model. Macroscopic models for these complex nonlinear systems require nonlocal multicontinua approaches, which are developed in earlier works [8]. These rigorous techniques require complex local computations, which involve solving local problems in oversampled regions subject to constraints. The solutions of these local problems can be replaced by solving original problem on a coarse (oversampled) region for many input parameters (boundary and source terms) and computing effective properties derived by nonlinear nonlocal multicontinua approaches. The effective properties depend on many variables (oversampled region and the number of continua), thus their calculations require some type of machine learning techniques. In this paper, our contribution is two fold. First, we present macroscopic models and discuss how to effectively compute macroscopic parameters using deep learning algorithms. The proposed method can be regarded as local machine learning and complements our earlier approaches on global machine learning [36], [35]. We consider a coarse grid approximation using two upscaling techniques with single phase upscaled transmissibilities and nonlocal nonlinear upscaled transmissibilities using a machine learning algorithm. We present results for two model problems in heterogeneous and fractured porous media and show that the presented method is highly accurate and provides fast coarse grid calculations.
裂缝多孔介质中多连续流问题的非局部多连续升级
DOI: --
发表时间: 2018
影响因子: 2.4
作者:
M. Vasilyeva;Eric T. Chung;Siu Wun Cheung;Yating Wang;G. Prokopev
通讯作者: G. Prokopev
模型降阶应用于耦合流动和地质力学
DOI: --
发表时间: 2018
期刊: ECMOR XVI - 16th European Conference on the Mathematics of Oil Recovery
影响因子: --
作者:
H. Florez;E. Gildin
通讯作者: E. Gildin
DOI: 10.1088/1742-6596/1158/4/042034
发表时间: 2019-02
期刊: Journal of Physics: Conference Series
影响因子: --
作者:
M. Vasilyeva;A. Tyrylgin
通讯作者: M. Vasilyeva;A. Tyrylgin
DOI: --
发表时间: 2018-10
期刊: ArXiv
影响因子: --
作者:
M. Vasilyeva;A. Tyrylgin
通讯作者: M. Vasilyeva;A. Tyrylgin
基于约束能量最小化的耦合流动和力学的升级
DOI: --
发表时间: 2018
影响因子: 4.1
作者:
M. Vasilyeva;Eric T. Chung;Y. Efendiev;Jihoon Kim
通讯作者: Jihoon Kim