Parametric generation of conditional geological realizations using generative neural networks

DOI10.1007/s10596-019-09850-7zbMath1425.86018arXiv1807.05207OpenAlexW3098024297WikidataQ127483063 ScholiaQ127483063MaRDI QIDQ2009823

Yanyan Li

Publication date: 2 December 2019

Published in: Computational Geosciences (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1807.05207

zbMATH Keywords

parametrizationgenerative modelsgeological modelsdeep learningmultipoint geostatistics

Mathematics Subject Classification ID

Geostatistics (86A32) Neural nets and related approaches to inference from stochastic processes (62M45)

Related Items (10)

Conditioning generative adversarial networks on nonlinear data for subsurface flow model calibration and uncertainty quantification ⋮ Data-space inversion using a recurrent autoencoder for time-series parameterization ⋮ Recent developments combining ensemble smoother and deep generative networks for facies history matching ⋮ U-net generative adversarial network for subsurface facies modeling ⋮ An end-to-end three-dimensional reconstruction framework of porous media from a single two-dimensional image based on deep learning ⋮ Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow ⋮ Stochastic seismic waveform inversion using generative adversarial networks as a geological prior ⋮ Reconstruction of fluvial reservoirs using multiple-stage concurrent generative adversarial networks ⋮ GANSim: conditional facies simulation using an improved progressive growing of generative adversarial networks (GANs) ⋮ A deep-learning-based geological parameterization for history matching complex models

Uses Software

• PyTorch
• RMSprop
• Adam
• BigGAN
• Wasserstein GAN

Cites Work

• Kernel principal component analysis for stochastic input model generation
• Regularized kernel PCA for the efficient parameterization of complex geological models
• Comparing training-image based algorithms using an analysis of distance
• Compressed history matching: Exploiting transform-domain sparsity for regularization of nonlinear dynamic data integration problems
• Kernel principal component analysis for efficient, differentiable parametrization of multipoint geostatistics
• Sample estimate of the entropy of a random vector
• Machine learning for graph-based representations of three-dimensional discrete fracture networks
• Bayesian deep convolutional encoder-decoder networks for surrogate modeling and uncertainty quantification
• Conditional simulation of complex geological structures using multiple-point statistics
• Stochastic seismic waveform inversion using generative adversarial networks as a geological prior
• Identification of release sources in advection-diffusion system by machine learning combined with Green's function inverse method
• A new data-space inversion procedure for efficient uncertainty quantification in subsurface flow problems
• Monte Carlo simulation of permeability fields and reservoir performance predictions with SVD parameterization in RML compared with EnKF
• History matching of facies distribution with the EnKF and level set parameterization
• History matching of petroleum reservoirs using a level set technique
• Integral Probability Metrics and Their Generating Classes of Functions
• A new class of random vector entropy estimators and its applications in testing statistical hypotheses

This page was built for publication: Parametric generation of conditional geological realizations using generative neural networks