Pore-GNN: A graph neural network-based framework for predicting flow properties of porous media from micro-CT images
Abstract view|289|times PDF download|90|times
Abstract
This paper presents a hybrid deep learning framework that combines graph neural networks with convolutional neural networks to predict porous media properties. This approach capitalizes on the capabilities of pre-trained convolutional neural networks to extract n-dimensional feature vectors from processed three dimensional micro computed tomography porous media images obtained from seven different sandstone rock samples. Subsequently, two strategies for embedding the computed feature vectors into graphs were explored: extracting a single feature vector per sample (image) and treating each sample as a node in the training graph, and representing each sample as a graph by extracting a fixed number of feature vectors, which form the nodes of each training graph. Various types of graph convolutional layers were examined to evaluate the capabilities and limitations of spectral and spatial approaches. The dataset was divided into 70/20/10 for training, validation, and testing. The models were trained to predict the absolute permeability of porous media. Notably, the proposed architectures further reduce the selected objective loss function to values below 35 mD, with improvements in the coefficient of determination reaching 9%. Moreover, the generalizability of the networks was evaluated by testing their performance on unseen sandstone and carbonate rock samples that were not encountered during training. Finally, a sensitivity analysis is conducted to investigate the influence of various hyperparameters on the performance of the models. The findings highlight the potential of graph neural networks as promising deep learning-based alternatives for characterizing porous media properties. The proposed architectures efficiently predict the permeability, which is more than 500 times faster than that of numerical solvers.
Document Type: Original article
Cited as: Alzahrani, M. K., Shapoval, A., Chen, Z., Rahman, S. S. Pore-GNN: A graph neural network-based framework for predicting flow properties of porous media from micro-CT images. Advances in Geo-Energy Research, 2023, 10(1):39-55. https://doi.org/10.46690/ager.2023.10.05
Keywords
Full Text:
PDFReferences
Akai, T., Blunt, M. J., Bijeljic, B. Pore-scale numerical simulation of low salinity water flooding using the lattice boltzmann method. Journal of Colloid and Interface Science, 2020, 566: 444-453.
Al-Hashimi, O., Hashim, K., Loffill, E., et al. A comprehensive review for groundwater contamination and remediation: Occurrence, migration and adsorption modelling. Molecules (Basel, Switzerland), 2021, 26(19): 5913.
Alqahtani, N., Alzubaidi, F., Armstrong, R. T., et al. Machine learning for predicting properties of porous media from 2D X-ray images. Journal of Petroleum Science and Engineering, 2020, 184: 106514.
Alqahtani, N., Armstrong, R. T., Mostaghimi, P. Deep learning convolutional neural networks to predict porous media properties. Paper SPE 191906 Presented at the SPE Asia Pacific Oil and Gas Conference and Exhibition, Australia, 23-25 October, 2018.
Alqahtani, N. J., Chung, T., Wang, Y., et al. Flow-based characterization of digital rock images using deep learning. SPE Journal, 2021, 26(4): 1800-1811.
Alsobhani, A., ALabboodi, H. M. A., Mahdi, H. Speech recognition using convolution deep neural networks. Journal of Physics: Conference Series, 2021, 1973: 012166.
Ashual, O., Wolf, L. Specifying object attributes and relations in interactive scene generation. arXiv, 2019, 1909.05379.
Bhatnagar, P. L., Gross, E. P., Krook, M. A model for collision processes in gases. I. small amplitude processes in charged and neutral one-component systems. Physical Review, 1954, 94: 511-525.
Brantson, E. T., Ju, B., Appau, P. O., et al. Development of hybrid low salinity water polymer flooding numerical reservoir simulator and smart proxy model for chemical enhanced oil recovery (CEOR). Journal of Petroleum Science and Engineering, 2020, 187: 106751.
Cai, C., Vlassis, N., Magee, L., et al. Equivariant geometric learning for digital rock physics: Estimating formation factor and effective permeability tensors from morse graph. International Journal for Multiscale Computational Engineering, 2023, 21: 1-24.
Cheng, C., Herrmann, J., Wagner, B., et al. Long-term evolution of fracture permeability in slate: An experimental study with implications for enhanced geothermal systems (EGS). Geosciences, 2021, 11(11): 443.
Defferrard, M., Bresson, X., Vandergheynst, P. Convolutional neural networks on graphs with fast localized spectral filtering. arXiv, 2017,1606.09375.
Deng, J., Dong, W., Socher, R., et al. ImageNet: A largescale hierarchical image database. Paper Presented at 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA, 20-25 June, 2009.
d’ Humires, D. Multiple-relaxation-time lattice boltzmann models in three dimensions. Philosophical Transactions of the Royal Society of London, 2002, 360(1792): 437-451.
Ding, K., Wang, S., Luo, Y. Supervised biological network alignment with graph neural networks. Bioinformatics, 2023, 39: i465-i474.
Ding, H., Wang, Y., Shapoval, A., et al. Macro- and microscopic studies of “smart water” flooding in carbonate rocks: An image-based wettability examination. Energy & Fuels, 2019, 33(8): 6961-6970.
Ershadnia, R., Wallace, C. D., Soltanian, M. R. CO2 geological sequestration in heterogeneous binary media: Effects of geological and operational conditions. Advances in Geo-Energy Research, 2020, 4(4): 392-405.
Essaid, H. I., Bekins, B. A., Cozzarelli, I. M. Organic contaminant transport and fate in the subsurface: Evolution of knowledge and understanding. Water Resources Research, 2015, 51: 4861-4902.
Fey, M., Lenssen, J. E. Fast graph representation learning with pyTorch geometric. arXiv, 2019, 1903.02428.
Gärttner, S., Alpak, F. O., Meier, A., et al. Estimating permeability of 3D micro-CT images by physics-informed CNNs based on DNS. Computational Geosciences, 2023, 27: 245-262.
Gerke, K. M., Sizonenko, T. O., Karsanina, M. V., et al. Improving watershed-based pore-network extraction method using maximum inscribed ball pore-body positioning. Advances in Water Resources, 2020, 140: 103576.
Gilmer, J., Schoenholz, S. S., Riley, P. F., et al. Neural message passing for quantum chemistry. arXiv, 2017, 1704.01212.
Goldberger, J., Hinton, G. E., Roweis, S., et al. Neighbourhood components analysis. Paper Presented at Advances in Neural Information Processing Systems 17, Columbia, Canada, December, 2004.
Gostick, J., Khan, Z., Tranter, T., et al. PoreSpy: A python toolkit for quantitative analysis of porous media images. Journal of Open Source Software, 2019, 4: 1296.
Graczyk, K. M., Matyka, M. Predicting porosity, permeability, and tortuosity of porous media from images by deep learning. Scientific Reports, 2020, 10: 21488.
Guo, Y., Zhang, L., Zhu, G., et al. A pore-scale investigation of residual oil distributions and enhanced oil recovery methods. Energies, 2019, 12(19): 3732.
Hamilton, W. L., Ying, R., Leskovec, J. Inductive representation learning on large graphs. arXiv, 2018, 1706.02216.
Hussain, S. T., Rahman, S. S., Azim, R. A., et al. Multiphase fluid flow through fractured porous media supported by innovative laboratory and numerical methods for estimating relative permeability. Energy & Fuels, 2021, 35(21): 17372-17388.
Hussain, S. T., Regenauer-Lieb, K., Zhuravljov, A., et al. Asymptotic hydrodynamic homogenization and thermodynamic bounds for upscaling multiphase flow in porous media. Advances in Geo-Energy Research, 2023, 9(1): 38-53.
Iglauer, S., Pentland, C. H., Busch, A. CO2 wettability of seal and reservoir rocks and the implications for carbon geosequestration. Water Resources Research, 2015, 51(1): 729-774.
Ijeje, J. J., Gan, Q., Cai, J. Influence of permeability anisotropy on heat transfer and permeability evolution in geothermal reservoir. Advances in Geo-Energy Research, 2019, 3(1): 43-51.
Jiang, J., Guo, B. Graph convolutional networks for simulating multi-phase flow and transport in porous media. arXiv, 2023, 2307.04449.
Jolliffe, I. T., Cadima, J. Principal component analysis: A review and recent developments. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2016, 374(2065): 20150202.
Kashefi, A., Mukerji, T. Point-cloud deep learning of porous media for permeability prediction. Physics of Fluids, 2021, 33(9): 097109.
Kingma, D. P., Ba, J. Adam: A method for stochastic optimization. arXiv, 2017, 1412.6980.
Kipf, T. N., Welling, M. Semi-supervised classification with graph convolutional networks. arXiv, 2017, 1609.02907.
Li, X., Li, B., Liu, F., et al. Advances in the application of deep learning methods to digital rock technology. Advances in Geo-Energy Research, 2023, 8(1): 5-18.
Li, B., Nie, X., Cai, J., et al. U-Net model for multicomponent digital rock modeling of shales based on CT and QEMSCAN images. Journal of Petroleum Science and Engineering, 2022, 216: 110734.
Liang, J., Deng, Y., Zeng, D. A deep neural network combined CNN and GCN for remote sensing scene classification. Earth Observations and Remote Sensing, 2020, 13: 4325-4338.
Liu, Q., Kampffmeyer, M., Jenssen, R., et al. SCG-Net: Self-constructing graph neural networks for semantic segmentation. arXiv, 2021, 2009.01599.
Lucas-Oliveira, E., Araujo-Ferreira, A. G., Trevizan, W. A., et al. Sandstone surface relaxivity determined by NMR T2 distribution and digital rock simulation for permeability evaluation. Journal of Petroleum Science and Engineering, 2020, 193: 107400.
Mohammadi, M., Riazi, M. Applicable investigation of SPH in characterization of fluid flow in uniform and nonuniform periodic porous media. Sustainability, 2022, 14(21): 14320.
Morris, C., Ritzert, M., Fey, M., et al. Weisfeiler and leman go neural: Higher-order graph neural networks. arXiv, 2021, 1810.02244.
Neumann, R. F., Barsi-Andreeta, M., Lucas-Oliveira, E., et al. High accuracy capillary network representation in digital rock reveals permeability scaling functions. Scientific Reports, 2021, 11(1): 11370.
Paszke, A., Gross, S., Massa, F., et al. PyTorch: An imperative style, high-performance deep learning library. arXiv, 2019, 1912.01703.
Pearson, K. LIII. On lines and planes of closest fit to systems of points in space. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 1901, 2(11): 559-572.
Perrin, J. C., Krause, M., Kuo, C. W., et al. Core-scale experimental study of relative permeability properties of CO2 and brine in reservoir rocks. Energy Procedia, 2009, 1(1): 3515-3522.
Pfaff, T., Fortunato, M., Sanchez-Gonzalez, A., et al. Learning mesh-based simulation with graph networks. arXiv, 2021, 2010.03409.
Prodanovic, M., Esteva, M., Hanlon, M. Digital rocks portal: A sustainable platform for imaged dataset sharing, translation and automated analysis. AGU Fall Meeting Abstracts, 2015, 2015: MR43A-02.
Qi, C. R., Su, H., Mo, K., et al. PointNet: Deep learning on point sets for 3D classification and segmentation. arXiv, 2017, 1612.00593.
Rabbani, A., Babaei, M., Shams, R., et al. DeePore: A deep learning workflow for rapid and comprehensive characterization of porous materials. Advances in Water Resources, 2020, 146: 103787.
Redmon, J., Divvala, S., Girshick, R., et al. You only look once: Unified, real-time object detection. arXiv, 2016, 1506.02640.
Ronneberger, O., Fischer, P., Brox, T. U-Net: Convolutional networks for biomedical image segmentation. arXiv, 2015, 1505.04597.
Ryazanov, A. V., van Dijke, M. I., Sorbie, K. S. Pore-network prediction of residual oil saturation based on oil layer drainage in mixed-wet systems. Paper SPE 129919 Presented at the SPE Improved Oil Recovery Symposium, Tulsa Oklahoma, 24-28 April, 2010.
Santos, J. E., Gigliotti, A., Bihani, A., et al. MPLBM-UT: Multiphase LBM library for permeable media analysis. SoftwareX, 2022, 18: 101097.
Santos, J. E., Xu, D., Jo, H., et al. PoreFlow-Net: A 3D convolutional neural network to predict fluid flow through porous media. Advances in Water Resources, 2020, 138: 103539.
Shapoval, A., Alzahrani, M., Xue, W., et al. Oil-water interactions in porous media during fluid displacement: Effect of potential determining ions (PDI) on the formation of in-situ emulsions and oil recovery. Journal of Petroleum Science and Engineering, 2022, 210: 110079.
Shapoval, A., Zhuravljov, A., Lanetc, Z., et al. Pore-scale evaluation of physicochemical interactions by engineered water injections. Transport in Porous Media, 2023, 148: 605-625.
Taigman, Y., Yang, M., Ranzato, M., et al. DeepFace: Closing the gap to human-level performance in face verification. Paper Presented at 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA, 23-28 June, 2014.
Tran, D., Bourdev, L., Fergus, R., et al. Learning spatiotemporal features with 3D convolutional networks. Paper Presented at 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile, 7-13 December, 2015.
Wang, W., Gang, J. Application of convolutional neural network in natural language processing. Paper Presented at 2018 International Conference on Information Systems and Computer Aided Education (ICISCAE), Changchun, China, 6-8 July, 2018.
Wang, Z., Li, H., Lan, X., et al. Formation damage mechanism of a sandstone reservoir based on micro-computed tomography. Advances in Geo-Energy Research, 2021, 5(1): 25-38.
Wang, Y., Shabaninejad, M., Armstrong, R. T., et al. Physical accuracy of deep neural networks for 2D and 3D multimineral segmentation of rock micro-CT images. arXiv, 2020, 2002.05322.
Wang, Y., Song, R., Liu, J., et al. Pore scale investigation on scaling-up micro-macro capillary number and wettability on trapping and mobilization of residual fluid. Journal of Contaminant Hydrology, 2019, 225: 103499.
Wu, C., Pfrommer, J., Beyerer, J., et al. Object detection in 3D point clouds via local correlation-aware point embedding. Paper Presented at the 2020 Joint 9th International Conference on Informatics, Electronics & Vision (ICIEV) and 2020 4th International Conference on Imaging, Vision & Pattern Recognition (icIVPR), Kitakyushu, Japan, 26-29 August, 2020.
Ying, R., He, R., Chen, K., et al. Graph convolutional neural networks for web-scale recommender systems. Paper Presented at the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, London, United Kingdom, 19-23 August, 2018.
Yun, W., Liu, Y., Kovscek, A. R. Deep learning for automated characterization of pore-scale wettability. Advances in Water Resources, 2020, 144: 103708
DOI: https://doi.org/10.46690/ager.2023.10.05
Refbacks
- There are currently no refbacks.
Copyright (c) 2023 The Author(s)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.