Abstract
The reconstruction of porous media is significant to some fields such as the study of seepage mechanics and reservoir engineering. Traditional methods have challenges in reconstruction quality due to their insufficient learning ability and suffer from lengthy computational time in large-quantity reconstruction tasks since they cannot reuse the parameters or models established previously. As a branch of deep learning, the variational auto-encoder (VAE) has shown excellent performance in extracting characteristics reflecting the underlying data manifold through a set of possible variables based on a latent model. Fisher information can help to balance the encoder and decoder in information control, used to estimate the variance of maximum likelihood estimate equation. Therefore, this paper proposes a method to reconstruct porous media based on VAE and Fisher information. Firstly, the structural characteristics of porous media are studied by the neural networks of the encoder to obtain the mean and variance of these characteristics. Then, the random sampling is carried out to reconstruct the intermediate results, and the optimization function of the encoder is combined with Fisher information to optimize the network. Finally, the intermediate results are input into the decoder to reconstruct porous media, and the optimization function of the decoder combined with Fisher information optimizes the reconstructed results. This method is evaluated by comparing the variogram, multiple-point connectivity, permeability and porosity of sandstone samples with some other typical reconstruction methods, showing its good reconstruction quality and efficiency.
Article Highlights
-
The reconstruction by our method is better than that of traditional methods.
-
The proposed method can reuse the parameters or models established previously.
-
The proposed method has a faster training speed.
Similar content being viewed by others
References
Al-Raoush, R., Apostolos, P.: Representative elementary volume analysis of porous media using X-ray computed tomography. Powder Technol. 200(1–2), 69–77 (2010)
Arns, C.H.: A comparison of pore size distributions derived by NMR and X-ray-CT techniques. Physica A 339(1–2), 159–165 (2004)
Bakke, S., Øren, P.E.: 3-D pore-scale modelling of sandstones and flow simulations in the pore networks. SPE J. 2(02), 136–149 (1997)
Bengio, Y., Lamblin, P., Popovici, D., Larochelle, H.: Greedy layer-wise training of deep networks. Proceedings of the 19th international conference on neural information processing system 153–160 (2006)
Blunt, M.J., Bijeljic, B., Dong, H., Gharbi, O., Iglauer, S., Mostaghimi, P., Paluszny, A., Pentland, C.: Pore-scale imaging and modelling. Adv. Water Resour. 51, 197–216 (2013)
Burda, Y., Grosse, R., Salakhutdinov, R.: Importance weighted autoencoders. arXiv: 1509.00519 (2015)
Chen, X., Kingma, D.P., Salimans, T., Duan, Y., Dhariwal, P., Schulman, J., Sutskever, I., Abbeel, P.: Variational lossy autoencoder. International conference on learning representation (2016)
Cormen, T.H., Leiserson, C.E., Rivest, R.L., Stein, C.: Introduction to algorithms. MIT Press (2009)
Costanza, Robinson, M.S., Estabrook, B.D., Fouhey, D.F.: Representative elementary volume estimation for porosity, moisture saturation, and air water interfacial areas in unsaturated porous media: Data quality implications. Water Resour. Res. 47(7), 07513.1–07513.12 (2011)
Dembo, A., Cover, T.M., Thomas, J.A.: Information theoretic inequalities. IEEE Trans. Inf. Theory 37(6), 1501–1518 (1991)
Fisher, R.A.: On the mathematical foundations of theoretical statistics. Philos. Trans. R. Soc. Lond. Ser. A 222, 309 (1922).
Fisher, R.A.: Theory of statistical estimation. Proc. Cambridge Philos. Soc. 22, 700–725 (1925)
Frieden, B.R.: Science from Fisher information: A Unification. Cambridge University Press (2004).
Hemes, S., Desbois, G., Urai, J.L., Desbois, G., Urai, J.L., Schröppel, B., Schwarz, J.O.: Multi-scale characterization of porosity in Boom Clay (HADES-level, Mol, Belgium) using a combination of X-ray μ-CT, 2D BIB-SEM and FIB-SEM tomography. Microporous Mesoporous Mater. 208, 1–20 (2015)
Hidajat, I., Rastogi, A., Singh, M., Mohanty, K.K.: Transport properties of porous media from thin-sections. SPE J. 7(1), 40–48 (2002)
Hinton, G.E., Osindero, S., Teh, Y.W.: A fast learning algorithm for deep belief nets. Neural Comput. 18(7), 1527–1554 (2006)
Ioannidis, M.A., Kwiecien, M., Chatzis, I.: Computer generation and application of 3d model porous media: from pore-level geostatistics to the estimation of formation factor. Petroleum computer conference (1995)
James, M.J.: Kullback-leibler divergence. International Encyclopedia of Statistical Science 720–722 (2011)
Karsanina, M.V., Gerke, K.M., Skvortsova, E.B., Mallants, D.: Universal spatial correlation functions for describing and reconstructing soil microstructure. Plos One 10(5), e0126515 (2015)
Kingma, D.P., Welling, M.: Auto-encoding variational bayes. International conference on learning representations 14–27 (2014)
Louizos, C., Swersky, K., Li, Y., Welling, M., Zemel, R.: The variational fair autoencoder. arXiv: 1511.00830 (2015)
Mahmud, K., Mariethoz, G., Caers, J., Tahmasebi, P., Baker, A.: Simulation of Earth textures by conditional image quilting. Water Resour. Res. 50, 3088–3107 (2014)
Makhzani, A., Shlens, J., Jaitly, N., Goodfellow, L., Frey, B.: Adversarial autoencoders. arXiv: 1511.05644 (2015)
Nash, C., Williams, C.K.I.: The shape variational autoencoder: a deep generative model of part-segmented 3D objects. Computer Graphics Forum 36(5), 1–12 (2017)
Nordahl, K., Ringrose, P.S.: Identifying the representative elementary volume for permeability in heterolithic deposits using numerical rock models. Math. Geosci. 40(7), 753–771 (2008)
Oda, M.: A method for evaluating the representative elementary volume based on joint survey of rock masses. Can. Geotech. J. 25(3), 440–447 (1988)
Okabe, H., Blunt, M.J.: Pore space reconstruction using multiple-point statistics. J. Petrol. Sci. Eng. 46(1–2), 121–137 (2005)
Okabe, H., Blunt, M.J.: Prediction of permeability for porous media reconstructed using multiple-point statistics. Phys. Rev. E 70(6), 066135 (2004)
Øren, P.E., Bakke, S.: Process based reconstruction of sandstones and prediction of transport properties. Transp. Porous Media 46(2–3), 311–343 (2002)
Phuong, M., Welling, M., Kushman, N., Tomioka, R., Nowozin S.: The mutual autoencoder: Controlling information in latent code representations. ICLR 2018 conference blind submission (2018)
Quiblier, J.A.: A new three-dimensional modeling technique for studying porous media. Colloid Interface Sci. 98, 84–102 (1984)
Rezende, D.J., Mohamed, S.: Variational inference with normalizing flows. Comput. Sci. 34(6), 421–427 (2015)
Rosso, O.A., Olivares, F., Plastino, A.: Noise versus chaos in a causal fisher-shannon plane. Papers Phys. 7, 070006 (2015)
Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning representations by back-propagating errors. Nature 323(6088), 533–536 (1986)
Samuel, R.B., Luke, V., Oriol, V., Andrew, M.D., Samy, B.: Generating sentences from a continuous space. arxiv:1511.06349 (2015)
Schrödinger, E.: About heisenberg uncertainty relation. Proc. Prussian Acad. Sci. Phys. Math. XIX. 293 (1930)
Shannon, C., Weaver, W.: The mathematical theory of communication. University of Illinois Press (1949)
Stam, A.J.: Some inequalities satisfied by the quantities of information of fisher and shannon. Inf. Control 2(2), 101–112 (1959)
Strebelle, S.: Conditional simulation of complex geological structures using multiple-point statistics. Math. Geol. 34(1), 1–21 (2002)
Stingaciu, L.R., Weihermuller, L., Haberpohlmeier, S.: Determination of pore size distribution and hydraulic properties using nuclear magnetic resonance relaxometry: a comparative study of laboratory methods. Water Resour. Res. 46(11), 2387–2392 (2010)
Tahmasebi, P., Hezarkhani, A., Sahimi, M.: Multiple-point geostatistical modeling based on the cross-correlation functions. Comput. Geosci. 16(3), 779–797 (2012)
Tahmasebi, P., Sahimi, M.: Reconstruction of three-dimensional porous media using a single thin section. Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 85(6), 066709 (2012)
Tomutsa, L., Silin, D., Radmilovic, V.: Analysis of chalk petrophysical properties by means of submicron-scale pore imaging and modeling. SPE Reservoir Eval. Eng. 10(3), 285–293 (2007)
Vignat, C., Bercher, J.F.: Analysis of signals in the Fisher-Shannon information plane. Phys. Lett. A 312(1–2), 27–33 (2003)
Vincent, P., Larochelle, H., Bengio, Y., Manzagol P.A.: Extracting and composing robust features with denoising autoencoders. Proceedings of the 25th international conference on machine learning 1096–1103 (2008)
Wirth, R.: Focused Ion Beam (FIB): A novel technology for advanced application of micro- and nanoanalysis in geosciences and applied mineralogy. Eur. J. Mineral. 16(6), 863–876 (2004)
Wirth, R.: Focused Ion Beam (FIB) combined with SEM and TEM: advanced analytical tools for studies of chemical composition, microstructure and crystal structure in geomaterials on a nanometre scale. Chem. Geol. 261(3–4), 217–229 (2009)
Zhang, T., Du, Y., Huang, T., Yang, J., Lu, F., Li, X.: Reconstruction of porous media using ISOMAP-based MPS. Stoch. Env. Res. Risk Assess. 30(1), 395–412 (2016)
Zhang, T., Switzer, P., Journel, A.: Filter-based classification of training image patterns for spatial simulation. Math. Geol. 38(1), 63–80 (2006)
Zhao, S., Song, J., Ermon, S.: InfoVAE: Information maximizing variational autoencoders. arxiv:1706.02262 (2017)
Zheng, H., Yao, J., Zhang, Y., Tsang, I.W. and Wang, J.: Understanding vaes in fisher-shannon plane. In Proceedings of the AAAI conference on artificial intelligence 33(1), 5917–5924 (2019).
Funding
This work is supported by the National Natural Science Foundation of China (Nos. 41672114, 41702148).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Zhang, T., Tu, H., Xia, P. et al. Reconstruction of Porous Media Using an Information Variational Auto-Encoder. Transp Porous Med 143, 271–295 (2022). https://doi.org/10.1007/s11242-022-01769-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11242-022-01769-5