Reconstruction of three-dimensional porous media using generative adversarial neural networks
By Lukas Mosser, Olivier Dubrule, Martin J. Blunt. Physical Review E, 2017
481 citations recorded by the source corpus.
To evaluate the variability of multiphase flow properties of porous media at the pore scale, it is necessary to acquire a number of representative samples of the void-solid structure. While modern x-ray computer tomography has made it possible to extract three-dimensional images of the pore space, assessment of the variability in the inherent material properties is often experimentally not feasible. We present a method to reconstruct the solid-void structure of porous media by applying a generative neural network that allows an implicit description of the probability distribution represented by three-dimensional image data sets. We show, by using an adversarial learning approach for neural networks, that this method of unsupervised learning is able to generate representative samples of porous media that honor their statistics. We successfully compare measures of pore morphology, such as the Euler characteristic, two-point statistics, and directional single-phase permeability of synthetic realizations with the calculated properties of a bead pack, Berea sandstone, and Ketton limestone. Results show that generative adversarial networks can be used to reconstruct high-resolution three-dimensional images of porous media at different scales that are representative of the morphology of the images used to train the neural network. The fully convolutional nature of the trained neural network allows the generation of large samples while maintaining computational efficiency. Compared to classical stochastic methods of image reconstruction, the implicit representation of the learned data distribution can be stored and reused to generate multiple realizations of the pore structure very rapidly.
Subfields: Geophysics, Ocean Engineering
View Paper: https://www.semanticscholar.org/paper/fed829b1ccf5750e2eca9f5833453d97d04506bc
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Geophysics
In Ocean Engineering
Citations in the corpus
- Training‐Image Based Geostatistical Inversion Using a Spatial Generative Adversarial Neural Network (2017, methodology): “Training of GANs is notoriously unstable, in the sense that image generation quality may not necessarily increase as training progresses (e.g., Salimans et al. , 2016; Sønderby et al. , 2016; Mosser et al. , 2017).”
- Deep Generative Models in Engineering Design: A Review (2021, background): “GAN [60]i [61]i [62]i [63]i [64] [65]i [66]vi [67]v [68]i [69]vi [70]c”
- How good is my GAN? (2018, unknown): “GANs are also used to learn complex 3D distributions and replace computationally intensive simulations in physics [39, 42] and neuroscience [38].”
- Deep Fluids: A Generative Network for Parameterized Fluid Simulations (2018, methodology): “ calorimeter interactions with electromagnetic showers using GANs. GANs have also been employed to generate [RLM17] and deconvolve [SZZ17] galaxy images, and reconstruct threedimensional porous media [MDB17]. As we focus on generative networks for known parameterizations, we will not employ learned,”
- Machine learning for data-driven discovery in solid Earth geoscience (2019, unknown): “A GAN is a system of two neural networks with opposing objectives: a generator network that uses training data to learn a model to generate realistic synthetic data and a discriminator network that learns to distinguish the synthetic data from real training data [see (94) for a clear explanation].”
- Data-driven synthesis of smoke flows with CNN-based feature descriptors (2017, unknown): “ms. E.g., they were successfully used for efficient and noise-free renderings [Bako et al. 2017; Chaitanya et al. 2017], the illumination of volumes [Kallweit et al. 2017], for modeling porous media [Mosser et al. 2017], and for character control [Peng et al. 2017]. First works also exist that targe”