Preconditioning Markov chain Monte Carlo method for geomechanical subsidence using multiscale method and machine learning technique

DOI10.1016/J.CAM.2021.113420zbMATH Open1467.74096arXiv2007.00066OpenAlexW3127637752MaRDI QIDQ2020498FDOQ2020498

Authors: Maria Vasilyeva, Aleksei Tyrylgin, Donald L. Brown, Anirban Mondal 

Publication date: 23 April 2021

Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)

Abstract: In this paper, we consider the numerical solution of the poroelasticity problem with stochastic properties. We present a Two-stage Markov Chain Monte Carlo method for geomechanical subsidence. In this work, we study two techniques of preconditioning: (MS) multiscale method for model order reduction and (ML) machine learning technique. The purpose of preconditioning is the fast sampling, where a new proposal is first testes by a cheap multiscale solver or using fast prediction of the neural network and the full fine grid computations will be conducted only if the proposal passes the first step. To construct a reduced order model, we use the Generalized Multiscale Finite Element Method and present construction of the multiscale basis functions for pressure and displacements in stochastic fields. In order to construct a machine learning based preconditioning, we generate a dataset using a multiscale solver and use it to train neural networks. The Karhunen-Loeve expansion is used to represent the realization of the stochastic field. Numerical results are presented for two- and three-dimensional model examples.

Full work available at URL: https://arxiv.org/abs/2007.00066

Recommendations

• Preconditioning Markov Chain Monte Carlo Simulations Using Coarse-Scale Models
• A Hierarchical Multilevel Markov Chain Monte Carlo Algorithm with Applications to Uncertainty Quantification in Subsurface Flow
• Multiscale finite-element method for linear elastic geomechanics
• Multilevel Markov Chain Monte Carlo
• Multiscale finite element methods for stochastic porous media flow equations and application to uncertainty quantification

zbMATH Keywords

Karhunen-Loeve expansiongeneralized multiscale finite element methodporoelastic heterogeneous mediumtwo-stage Markov chain Monte Carlo method

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Fluid-solid interactions (including aero- and hydro-elasticity, porosity, etc.) (74F10) Finite element methods applied to problems in solid mechanics (74S05) Soil and rock mechanics (74L10) Stochastic and other probabilistic methods applied to problems in solid mechanics (74S60) Inhomogeneity in solid mechanics (74E05)

Cites Work

• Automated solution of differential equations by the finite element method. The FEniCS book
• Multiscale Finite Element Methods
• Convergence of iterative coupling for coupled flow and geomechanics
• Preconditioning Markov Chain Monte Carlo Simulations Using Coarse-Scale Models
• Application of the two-stage Markov chain Monte Carlo method for characterization of fractured reservoirs using a surrogate flow model
• Multiscale two-stage solver for Biot's poroelasticity equations in subsurface media
• Reservoir Geomechanics
• Constrained energy minimization based upscaling for coupled flow and mechanics
• Learning macroscopic parameters in nonlinear multiscale simulations using nonlocal multicontinua upscaling techniques
• Computational multiscale methods for linear poroelasticity with high contrast
• A generalized multiscale finite element method for poroelasticity problems. I: Linear problems.
• A generalized multiscale finite element method for poroelasticity problems. II: Nonlinear coupling.
• Computational Multiscale Methods for Linear Heterogeneous Poroelasticity
• Multiscale finite volume method for finite-volume-based simulation of poroelasticity
• Uncertainty quantification in injection and soil characteristics for Biot's poroelasticity model
• A parametric acceleration of multilevel Monte Carlo convergence for nonlinear variably saturated flow
• Machine learning for accelerating macroscopic parameters prediction for poroelasticity problem in stochastic media
• Generalized multiscale finite element method for the poroelasticity problem in multicontinuum media
• Data-driven uncertainty quantification for predictive flow and transport modeling using support vector machines

Cited In (11)

• A Bayesian reinforcement learning approach in Markov games for computing near-optimal policies
• Prediction of discretization of online GMsFEM using deep learning for Richards equation
• A prefetching technique for prediction of porous media flows
• Prediction of numerical homogenization using deep learning for the Richards equation
• Bayesian decision making using partial data for fractured poroelastic media
• Generalized Multiscale Finite Element Treatment of a Heterogeneous Nonlinear Strain-limiting Elastic Model
• A computational macroscopic model of piezomagnetoelectric materials using generalized multiscale finite element method
• Non-local multi-continuum method (NLMC) for Darcy-Forchheimer flow in fractured media
• A Lyapunov approach for stable reinforcement learning
• Advancing wave equation analysis in dual-continuum systems: a partial learning approach with discrete empirical interpolation and deep neural networks
• Canonical Monte Carlo multispin cluster method

Uses Software

• EnKF
• FEniCS
• SyFi
• TensorFlow
• Keras
• ImageNet
• AlexNet

This page was built for publication: Preconditioning Markov chain Monte Carlo method for geomechanical subsidence using multiscale method and machine learning technique

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2020498)