3D crosswell electromagnetic inversion based on radial basis function neural network

• Autorzy: Fang Sinan , Zhang Zhansong , Chen Wei , Pan Heping , Peng Jun

Identyfikatory

DOI

10.1007/s11600-020-00445-w

• Języki publikacji: EN

Abstrakty

EN

Crosswell electromagnetic (EM) method has fundamentally improved the horizontal detection ability of well logging and will become an increasingly promising approach for the secondary exploration of hydrocarbon reservoir. We applied orthogonal least squares (OLS) radial basis function neural network (RBFNN) based on improved Gram–Schmidt (G–S) procedure to three-dimensional (3D) crosswell EM inversion problems. In the inversion process of the simplifed crosswell model with single-grid conductivity anomalies and normal oil reservoir, compared the inversion results of other fve neural networks, OLS-RBFNN was proved to have the best global optimization ability and the fastest sample learning speed and the average inversion error of low conductivity anomalies model (4%) and oil reservoir model (9%) can meet the inversion requirements of crosswell EM method. Only the OLS-RBFNN could achieve ideal inversion results in the most concerned central area of crosswell model, and the inversion accuracy of this algorithm will be more outstanding when the model becomes more complex. Merely using the three-component time-domain crosswell EM data of two wells, the inversion of 3D medium conductivity in the crosswell dominant exploration area can be efectively realized through the nonlinear approximation of the OLS-RBFNN.

Słowa kluczowe

EN

EM; 3D inversion; radial basis function neural network; RBFNN; orthogonal least squares; OLS; improved gram; Schmidt; GS; crosswell electromagnetic

PL

EM; inwersja 3D; radialna sieć neuronowa; RBFNN; metoda najmniejszych kwadratów; OLS

• Wydawca: Instytut Geofizyki PAN ; Springer
• Czasopismo: Acta Geophysica
• Rocznik: 2020
• Tom: Vol. 68, no. 3
• Strony: 711--721
• Opis fizyczny: Bibliogr. 32 poz.

Twórcy

autor

Fang Sinan

• Key Laboratory of Exploration Technologies for Oil and Gas Resources, Ministry of Education, Yangtze University, Wuhan, China
• College of Geophysics and Petroleum Resources, Yangtze University, Wuhan, China

autor

Zhang Zhansong

• Key Laboratory of Exploration Technologies for Oil and Gas Resources, Ministry of Education, Yangtze University, Wuhan, China
• College of Geophysics and Petroleum Resources, Yangtze University, Wuhan, China

autor

Chen Wei

• Key Laboratory of Exploration Technologies for Oil and Gas Resources, Ministry of Education, Yangtze University, Wuhan, China
• College of Geophysics and Petroleum Resources, Yangtze University, Wuhan, China

autor

Pan Heping

• Institute of Geophysics and Geomatics, China University of Geosciences, Wuhan, China

autor

Peng Jun

• Changjiang Institute of Survey Technical Research, MWR, Wuhan, China

Bibliografia

• 1. Abedinia O, Amjady N (2016) Short-term load forecast of electrical power system by radial basis function neural network and new stochastic search algorithm. Int Trans Electr Energy Syst 26(7):1511–1525
• 2. Alumbaugh DL, Morrison HF (1995a) Monitoring subsurface changes over time with cross-well electromagnetic tomography. Geophys Prospect 43(7):873–902
• 3. Alumbaugh DL, Morrison HF (1995b) Theoretical and practical considerations for cross-well electromagnetic tomography assuming a cylindrical geometry. Geophysics 60(3):846–870
• 4. Antonio J, Tadeu A (2004) The use of monopole and dipole sources in crosswell surveying. J Appl Geophys 56(4):231–245
• 5. Böhm G, Carcione JM, Gei D, Picotti S, Michelini A (2015) Cross-well seismic and electromagnetic tomography for CO2 detection and monitoring in a saline aquifer. J Petrol Sci Eng 133:245–257
• 6. Byun J, Yu J, Seol SJ (2010) Crosswell monitoring using virtual sources and horizontal wells. Geophysics 75:SA37–SA43
• 7. Carcione JM, Gei D, Picotti S, Michelini A (2012) Cross-hole electromagnetic and seismic modeling for CO2 detection and monitoring in a saline aquifer. J Petrol Sci Eng 100:162–172
• 8. Chen S, Cowan CF, Grant PM (1991) Orthogonal least squares learning algorithms for radial basis function networks. IEEE Trans Neural Netw 2(2):302–309
• 9. Donadille JM, Al-Ofi SM (2012) Crosswell electromagnetic response in a fractured medium. Geophysics 77(3):D53–D61
• 10. Fang SN, Pan HP, Du T, Konaté KK, Deng CX, Qin Z, Guo B, Peng L, Ma HL, Li G et al (2016) Crosswell electromagnetic modeling from impulsive source: optimization strategy for dispersion suppression in convolutional perfectly matched layer. Sci Rep 6:32613
• 11. Fang SN, Zhang ZS, Wang Z, Pan HP, Du T (2019) The dominant exploration area from three-dimensional crosswell electromagnetic modeling in the time domain. Energy Sour Part A Recovery Util Environ Eff. https://doi.org/10.1080/15567036.2019.1670293
• 12. Gunnink JL, Bosch JHA, Siemon B, Roth B, Auken E (2012) Combining ground-based and airborne EM through artificial neural networks for modelling glacial till under saline groundwater conditions. Hydrol Earth Syst Sci 16(8):3061–3074
• 13. Guo C, Zhang C, Zhu L (2017) Predicting the total porosity of shale gas reservoirs. Pet Sci Technol 35(10):1022–1031
• 14. Hu ZZ, He ZX, Yang WC et al (2015) Constrained inversion of magnetotelluric data with the artificial fish swarm optimization method. Chin J Geophys 58(7):2578–2587
• 15. Kaftan I, Salk M, Senol Y (2011) Evaluation of gravity data by using artificial neural networks case study: seferihisar geothermal area (Western Turkey). J Appl Geophys 75(4):711–718
• 16. Li JH, He ZX (2014) Three-dimensional cross-well electromagnetic inversion using the least-square method. Oil Geophys Prospect 49(3):586–595
• 17. Li MK, Abubakar A, Habashy TM (2010) Application of a two-and-a-half dimensional model-based algorithm to crosswell electromagnetic data inversion. Inverse Prob 26:074013
• 18. Liu JF, Hu WB, Hu XY (2015) Two-dimensional magnetotelluric inversion using differential ant-stigmergy algorithm. Oil Geophys Prospecting 50(3):548–555
• 19. Liu B, Liu ZY, Song J et al (2017) Joint inversion method of 3D electrical resistivity detection based on inequality constraints. Chin J Geophys 60(2):820–832
• 20. Maclennan K, Karaoulis M, Revil A (2014) Complex conductivity tomography using low-frequency crosswell electromagnetic data. Geophysics 79(1):E23–E38
• 21. Montahaei M, Oskooi B (2014) Magnetotelluric inversion forazimuthally anisotropic resistivities employing artificial neural networks. Acta Geophys 62(1):12–43
• 22. Moody J, Darken C (1989) Fast learning in networks of locally-tuned processing units. Neural Comput 1(2):281–294
• 23. Nie X, Zou CC, Li ZH et al (2016) Numerical simulation of the electrical properties of shale gas reservoir rock based on digital core. J Geophys Eng 13(4):481–490
• 24. Sharma SP (2012) VFSARES-a very fast simulated annealing fortran program for interpretation of 1-D DC resistivity sounding data from various electrode arrays. Comput Geosci 42:177–188
• 25. Shaw R, Srivastava S (2007) Particle swarm optimization: a new tool to invert geophysical data. Geophysics 72(2):F75–F83
• 26. Singh UK, Tiwari RK, Singh SB (2010) Inversion of 2-D DC resistivity data using rapid optimization and minimal complexity neural network. Nonlinear Processes Geophys 17:65–76
• 27. Wang H, Liu ML, Xi ZZ et al (2018) Magnetotelluric inversion based on BP neural network optimized by genetic algorithm. ChinJ Geophys 61(4):1563–1575
• 28. Wei BJ (2006) A combined 1D/2D inversion algorithm of cross_hole electromagnetic fields. Chin J Geophys 49(1):264–274
• 29. Yang LQ, Chen W, Liu W et al (2020) Random noise attenuation based on residual convolutional neural network in seismic datasets. IEEE Access 8(1):30271–30286
• 30. Zhang YC, Liu C, Shen LC (1995) Monitoring saltwater injection using conductivity images obtained by electromagnetic crosshole measurements. Radio Sci 30(5):1405–1415
• 31. Zhang YC, Liu C, Shen LC (1996) An interative algorithm for conductivity image reconstruction from crosswell-EM measurements. Geophysics 61(4):987–997
• 32. Zhu LQ, Zhang C, Wei Y et al (2017) Permeability prediction of the tight sandstone reservoirs using hybrid intelligent algorithm and Nuclear magnetic resonance logging data. Arab J Sci Eng 42(4):1643–1654

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2021)