Efficient uncertainty quantification for dynamic subsurface flow with surrogate by theory-guided neural network

DOI10.1016/j.cma.2020.113492zbMath1506.76178arXiv2004.13560OpenAlexW3021923779MaRDI QIDQ2020800

Haibin Chang, Nanzhe Wang, Dongxiao Zhang

Publication date: 26 April 2021

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2004.13560

zbMATH Keywords

uncertainty quantificationsubsurface flowsurrogate modelingtheory-guided neural network

Mathematics Subject Classification ID

Fluid-solid interactions (including aero- and hydro-elasticity, porosity, etc.) (74F10) Flows in porous media; filtration; seepage (76S05)

Related Items (9)

Weak form theory-guided neural network (TgNN-wf) for deep learning of subsurface single- and two-phase flow ⋮ Surrogate and inverse modeling for two-phase flow in porous media via theory-guided convolutional neural network ⋮ Uncertainty quantification in scientific machine learning: methods, metrics, and comparisons ⋮ A physics-informed convolutional neural network for the simulation and prediction of two-phase Darcy flows in heterogeneous porous media ⋮ Transfer learning based physics-informed neural networks for solving inverse problems in engineering structures under different loading scenarios ⋮ HRW: Hybrid Residual and Weak Form Loss for Solving Elliptic Interface Problems with Neural Network ⋮ A connection element method: both a new computational method and a physical data-driven framework -- take subsurface two-phase flow as an example ⋮ Theory-guided auto-encoder for surrogate construction and inverse modeling ⋮ Use of multifidelity training data and transfer learning for efficient construction of subsurface flow surrogate models

Uses Software

• MLMSRBF
• Adam

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Reservoir description with integrated multiwell data using two-dimensional wavelets
• Constrained probabilistic collocation method for uncertainty quantification of geophysical models
• A deep-learning-based surrogate model for data assimilation in dynamic subsurface flow problems
• Compressed history matching: Exploiting transform-domain sparsity for regularization of nonlinear dynamic data integration problems
• An efficient, high-order perturbation approach for flow in random porous media via Karhunen-Loève and polynomial expansions.
• Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
• Modeling uncertainty in steady state diffusion problems via generalized polynomial chaos
• Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
• Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
• Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
• Simulator-free solution of high-dimensional stochastic elliptic partial differential equations using deep neural networks
• A self-adaptive deep learning algorithm for accelerating multi-component flash calculation
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Monte Carlo simulation of permeability fields and reservoir performance predictions with SVD parameterization in RML compared with EnKF
• A Stochastic Radial Basis Function Method for the Global Optimization of Expensive Functions
• Large-Scale Machine Learning with Stochastic Gradient Descent
• The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
• Predicting the output from a complex computer code when fast approximations are available

This page was built for publication: Efficient uncertainty quantification for dynamic subsurface flow with surrogate by theory-guided neural network