Abstract
A neural network approximation algorithm for solving inverse geoelectrics problems in the class of grid (block) models of media is presented. The algorithm is based on using neural networks for constructing an approximate inverse operator and enables formalized construction of solutions of inverse geoelectrics problem with a total number of sought-for medium parameters of ~ n · 103. The correctness of the problem of constructing neural network inverse operators is considered. A posteriori estimates of the degree of ambiguity of solutions of the resulting inverse problem are calculated. The operation of the algorithm is illustrated by examples of 2D and 3D inversions of synthetic and field geoelectric data obtained by the MTS method.
Similar content being viewed by others
References
Berdichevskii, M.N. and Dmitriev, V.I., Inverse Problems of Magnetotellurics: A Modern Formulation, Izv., Phys. Sol. Earth, 2004, vol. 40, no. 4, pp. 276–292.
Van’yan, L.L. and Butkovskaya, A.I., Magnitotelluricheskoe zondirovanie sloistykh sred (Magnetotelluric Sounding of LayeredMedia),Moscow: Nedra, 1980.
Vorontsov, K.V., Lektsii po iskusstvennym neironnym setyam (Lectures on Artificial Neural Networks), 2009; http://www.machinelearning.ru/wiki/images/c/cc/Voron-ML-NeuralNets.pdf.
Galushkin, A.I., Sintez mnogosloinykh sistem raspoznavaniya obrazov (Synthesis of Multilayer Pattern Recognition Systems),Moscow: Energiya, 1974.
Glasko, V.B., Gushchin, G.V., and Starostenko, V.I., On Application of the Tikhonov Method of Regularization to Solution of Nonlinear Systems of Equations, Zh. Vych. Mat. Mat. Fiz., 1976, vol. 16, no. 2, pp. 283–292.
Dmitriev, V.I., Obratnye zadachi geofiziki (Inverse Problems ofGeophysics),Moscow:MAKS Press, 2012.
Dmitriev, V.I. and Kokotushkin, G.A., Al’bom paletok dlya magnitotelluricheskogo zondirovaniya v neodnorodnykh sredakh (Album of Reticulations for Magnetotelluric Sounding in Inhomogeneous Media), Moscow:MGU, 1971.
Dmitrieva, M.V., Numerical Simulation of Physical Processes in Plasma of Tokamak Plants under Influence of ElectromagneticWaves of Alfven FrequencyRange, ExtendedAbstract of Cand. Sci.Diss.,Moscow, 1985.
Ivanov, V.K., On Ill-Posed Problems, Mat. Sb., 1963, vol. 61, no. 2, pp. 211–223.
Kabanikhin, S.I., Shishlenin, M.A., and Novikov, N.S., Algorithms for Determination of Elastic Parameters Basing on Areal Data Collection Systems (Direct Linear Processing of Seismic Data), Vest. Kiber., 2016, no. 2, pp. 83–91.
Romanov, V.G. and Kabanikhin, S.I., Obratnyye zadachi geoelektriki (Inverse Problems of Geoelectrics), Moscow: Nauka, 1991.
Strakhov, V.N. and Lapina, M.I., Fitting Method for Solving the Inverse Problem of Gravimetry, Dokl. AN SSSR, 1976, vol. 227, no. 2, pp. 344–347.
Spichak, V.V. and Popova, I.V., Application of the Neural Network Approach to the Reconstruction of Three- Dimensional Geoelectric Structure, Izv., Phys. Sol. Earth, 1998, vol. 34, no. 1, pp. 33–39.
Haykin, S., Neural Networks and Learning Machines, 3d ed., Pearson Education, 2009.
Fel’dman, I.S., Okulesskii, B.A., Suleimanov, A.K., Nikolaev, V.I., Kuncherov, V.A., and Chamo, S.S., Electrical Prospecting by the MTS Method as Part of Regional Oil and Gas Exploration in the European Part of Russia, Zap. Gornogo Inst., 2008, vol. 176, pp. 125–131.
Shimelevich, M.I. and Obornev, E.A., An ApproximationMethod for Solving the InverseMTS Problem with the Use of Neural Networks, Izv., Phys. Sol. Earth, 2009, vol. 45, no. 12, pp. 1055–1071.
Shimelevich, M.I., Obornev, E.A., Obornev, I.E., and Rodionov, E.A., NumericalMethods for Estimating the Degree of Practical Stability of Inverse Problems in Geoelectrics, Izv., Phys. Sol. Earth, 2013, vol. 49, no. 3, pp. 356–362.
Shimelevich, M.I., Obornev, E.A., Obornev, I.E., and Rodionov, E.A., The Neural Network Approximation Method for Solving Multidimensional Nonlinear Inverse Problems of Geophysics, Izv., Phys. Sol. Earth, 2017, vol. 53, no. 4, pp. 588–597.
Cybenko, G., Approximation by Superpositions of a Sigmoidal Function, Math. Control, Signals, Sys., 1989, vol. 2, pp. 303–314.
Dolenko, S., Guzhva, A., Obornev, E., Persiantsev, I., and Shimelevich, M., Comparison of Adaptive Algorithms for Significant Feature Selection in Neural Network Based Solution of the Inverse Problem of Electrical Prospecting, Proc. ICANN 2009, part 2, Berlin: Springer-Verlag, 2009, pp. 397–405 (Lect. Notes in Computer Science, 5769).
Eisenstat, S., Schultz, M., and Sherman, A., Algorithms and Data Structures for Sparse Symmetric Gaussian Elimination, SIAM J. Sci. Stat. Comput., 1981, vol. 2, no. 2, pp. 225–237.
Hidalgo, H., Gymez-Trevico, E., and Swiniarski, R., Neural Network Approximation of a Inverse Functional, Proc. IEEE Int. Conf. on Neural Networks, vol. 5, 1994, pp. 3387–3392.
Kabanikhin, S.I., Definitions and Examples of Inverse and Ill-Posed Problems, J. Inv. Ill-Posed Problems, 2008, vol. 16, pp. 317–357.
Lönnblad, L., Peterson, C., and Rögnvalsson, T., Pattern Recognition in High Energy Physics with Artificial Neural Networks—JETNET 2.0, Comput. Phys. Comms., 1992, vol. 70, no. 1, pp. 167–182.
Mackie, R.L., Smith, J.T., and Madden, T.R., Three-Dimensional Electromagnetic Modeling Using Finite Difference Equations: TheMagnetotelluric Example, Radio Sci., 1994, vol. 29, pp. 923–935.
Poulton, M., Sternberg, B., and Glass, C., Neural Network Pattern Recognition of Subsurface EM Images, J. Appl. Geophys., 1992, vol. 29, no. 1, pp. 21–36.
Raiche, A., A Pattern Recognition Approach to Geophysical Inversion Using Neural Nets, Geophys. J. Int., 1991, vol. 105, pp. 629–648.
Shimelevitch, M. and Obornev, E., The Method of Neuron Network in Inverse Problems MTZ, Abstr. of the 14th Workshop on Electromagnetic Induction in the Earth, Sinaia, Romania, 1998.
Werbos, P.J., Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences, PhD thesis, Cambridge,MA: Harvard University, 1974.
Yagola, A., Error Estimation for Ill-Posed Problems with Priori Information, Glob. J. Technol. Optim., 2010, vol. 1, pp. 88–92.
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © M.I. Shimelevich, E.A. Obornev, I.E. Obornev, E.A. Rodionov, 2018, published in Sibirskii Zhurnal Vychislitel’noi Matematiki, 2018, Vol. 21, No. 4, pp. 451–468.
Rights and permissions
About this article
Cite this article
Shimelevich, M.I., Obornev, E.A., Obornev, I.E. et al. An Algorithm for Solving Inverse Geoelectrics Problems Based on the Neural Network Approximation. Numer. Analys. Appl. 11, 359–371 (2018). https://doi.org/10.1134/S1995423918040080
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1995423918040080