A robust inversion of logging-while-drilling responses based on deep neural network

• Autorzy: Zhu Gaoyang , Gao Muzhi , Wang Bin

Identyfikatory

DOI

10.1007/s11600-023-01080-x

• Języki publikacji: EN

Abstrakty

EN

Resistivity inversion plays a significant role in recent geological exploration, which can obtain formation information through logging data. However, resistivity inversion faces various challenges in practice. Conventional inversion approaches are always time-consuming, nonlinear, non-uniqueness, and ill-posed, which can result in an inaccurate and inefficient description of subsurface structure in terms of resistivity estimation and boundary location. In this paper, a robust inversion approach is proposed to improve the efficiency of resistivity inversion. Specifically, inspired by deep neural networks (DNN) remarkable nonlinear mapping ability, the proposed inversion scheme adopts DNN architecture. Besides, the batch normalization algorithm is utilized to solve the problem of gradient disappearing in the training process, as well as the k-fold cross-validation approach is utilized to suppress overfitting. Several groups of experiments are considered to demonstrate the feasibility and efficiency of the proposed inversion scheme. In addition, the robustness of the DNN-based inversion scheme is validated by adding different levels of noise to the synthetic measurements. Experimental results show that the proposed scheme can achieve faster convergence and higher resolution than the conventional inversion approach in the same scenario. It is very significant for geological exploration in layered formations.

Słowa kluczowe

EN

deep neural networks; batch normalization; k-fold cross validation; logging while drilling; resistivity inversion; LWD; DNN

PL

głębokie sieci neuronowe; normalizacja wsadowa; k-krotna walidacja krzyżowa; rejestrowanie podczas wiercenia; inwersja rezystancji

• Wydawca: Instytut Geofizyki PAN ; Springer
• Czasopismo: Acta Geophysica
• Rocznik: 2024
• Tom: Vol. 72, no. 1
• Strony: 129--139
• Opis fizyczny: Bibliogr. 36 poz.

Twórcy

autor

Zhu Gaoyang

• College of Electronic and Information Engineering, Shandong University of Science and Technology, Qingdao 266590, China

autor

Gao Muzhi

• College of Control Science Engineering, China University of Petroleum (East China), Qingdao 266590, China

• https://orcid.org/0000-0002-5748-6131

autor

Wang Bin

• College of Control Science Engineering, China University of Petroleum (East China), Qingdao 266590, China

Bibliografia

• 1. Colombo, Alyaev S, Shahriari M, Pardo D, Omella AJ, Larsen DS, Jahani N et al (2021a) Modeling extra-deep electromagnetic logs using a deep neural network. Geophysics 86(3):E269-E281
• 2. Colombo D, Turkoglu E, Li W, Sandoval-Curiel E, Rovetta D (2021b) Physics-driven deep-learning inversion with application to transient electromagnetics. Geophysics 86(3):E209-E224
• 3. Hao P, Sun X, Nie Z, Yue X, Zhao Y (2022) A robust inversion of induction logging responses in anisotropic formation based on supervised descent method. IEEE Geosci Remote Sens Lett 19:1-5
• 4. Hardman RH, Shen LC (1986) Theory of induction sonde in dipping beds. Geophysics 51(3):800
• 5. Heriyanto M, Srigutomo W (2017) 1-D DC resistivity inversion using singular value decomposition and levenberg-marquardt’s inversion schemes. In: Journal of Physics: Conference Series, vol 877, No (1), pp 012066
• 6. Hu YY, Guo R, Jin YC, Wu XQ, Li MK, Abubakar A et al (2020) A Supervised descent learning technique for solving directional electromagnetic logging-while-drilling inverse problems. IEEE Trans Geosci Remote Sens 58(11):8013-8025
• 7. Ioffe, S., and Szegedy, C. (2015). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. In: International conference on machine learning, pp. 448-456. arXiv preprint arXiv:1502.03167
• 8. Jianchang M, Anil KJ (1996) Artificial neural networks: a tutorial. Computer 29:31-44
• 9. Li MY, Yue XG, Hong DC, Han W (2015) Simulation and analysis of the symmetrical measurements of a triaxial induction tool. IEEE Geosci Remote Sens Lett 12(1):122-124
• 10. Li H, He ZH, Zhang YT, Feng J, Jian ZY, Jiang YB (2022) A study of health management of LWD tool based on data-driven and model-driven. Acta Geophys 70(2):669-676
• 11. Liu DJ, Li H, Zhang YY, Zhu GX, Ai QH (2014) A study on directional resistivity logging-while-drilling based on self-adaptive hp-FEM. Acta Geophys 62(6):1328-1351
• 12. Liu B, Guo Q, Li S, Liu B, Jiang P (2020) Deep learning inversion of electrical resistivity data. IEEE Trans Geosci Remote Sens 58(8):5715-5728
• 13. Noh K, Pardo D, Torres-Verdin C (2022) 2.5-D Deep learning inversion of LWD and deep-sensing em measurements across formations with dipping faults. IEEE Geosci Remote Sens Lett 19:1-5
• 14. Pardo D, Torres-Verdín C (2015) Fast 1D inversion of logging-whiledrilling resistivity measurements for improved estimation of formation resistivity in high-angle and horizontal wells. Geophysics 80(2):E111-E124
• 15. Shen QY, Wu XQ, Chen JF, Han Z, Huang YQ (2018) Solving geosteering inverse problems by stochastic Hybrid Monte Carlo method. J Petrol Sci Eng 161:9-16
• 16. Raj AS, Srinivas Y, Oliver DH, Muthuraj D (2014) A novel and generalized approach in the inversion of geoelectrical resistivity data using artificial neural networks (ANN). J Earth Syst Sci 123(2):395-411
• 17. Shahriari M, Pardo D, Picon A, Galdran A, Del Ser J, Torres-Verdin C (2020) A deep learning approach to the inversion of borehole resistivity measurements. Comput Geosci 24(3):971-994
• 18. Shahriari M, Pardo D, Rivera JA, Torres-Verdin C, Picon A, Del Ser J et al (2021) Error control and loss functions for the deep learning inversion of borehole resistivity measurements. Int J Numer Meth Eng 122(6):1629-1657
• 19. Shahriari M, Hazra A, Pardo D (2022) A deep learning approach to design a borehole instrument for geosteering. Geophysics 87(2):D83-D90
• 20. Singh UK, Tiwari RK, Singh SB (2013) Neural network modeling and prediction of resistivity structures using VES Schlumberger data over a geothermal area. Comput Geosci 52(MAR.):246-257
• 21. Tembely M, AlSumaiti AM, Alameri W (2020) A deep learning perspective on predicting permeability in porous media from network modeling to direct simulation. Comput Geosci 24(4):1541-1556
• 22. Thiel M, Omeragic D (2017) High-fidelity real-time imaging with electromagnetic logging-while-drilling measurements. IEEE Trans Comput Imaging 3(2):369-378
• 23. Veettil DRA, Clark K (2020) Bayesian geosteering using sequential monte carlo methods. Petrophysics 61(1):99-111
• 24. Wang GL, Barber T, Wu P, Allen D, Abubakar A (2017) Fast inversion of triaxial induction data in dipping crossbedded formations. Geophysics 82(2):D31-D45
• 25. Wang L, Li H, Fan Y (2019) Bayesian inversion of logging-whiledrilling extra-deep directional resistivity measurements using parallel tempering markov chain monte carlo sampling. IEEE Trans Geosci Remote Sens 57(10):8026-8036
• 26. Xing G, Wang H, Ding Z (2008) A new combined measurement method of the electromagnetic propagation resistivity logging. IEEE Geosci Remote Sens Lett 5(3):430-432
• 27. Xu Y, Sun K, Xie H, Zhong X, Hong X (2018) Borehole resistivity measurement modeling using machine-learning techniques. Petrophysics 59(6):778-785
• 28. Yadav S, Shukla S (2016) Analysis of k-fold cross-validation over holdout validation on colossal datasets for quality classification In:
• 29. IEEE 6th International advance computing conference (IACC), pp 78-83
• 30. Yang S, Hong D, Huang WF, Liu QH (2017) A stable analytic model for tilted-coil antennas in a concentrically cylindrical multilayered anisotropic medium. IEEE Geosci Remote Sens Lett 14(4):480-483
• 31. Yu X, Efe MO, Kaynak O (2002) A general backpropagation algorithm for feedforward neural networks learning. IEEE Trans Neural Netw 13(1):251-254
• 32. Zhang L (2000) Application of neural networks to interpretation of well logs. The University of Arizona, Tucson
• 33. Zhang Z (2011) 1-D modeling and inversion of triaxial induction logging tool in layered anisotropic medium. University of Houston, Houston
• 34. Zhong L, Jing L, Bhardwaj A, Shen LC, Liu RC (2008) Computation of triaxial induction logging tools in layered anisotropic dipping formations. IEEE Trans Geosci Remote Sens 46(4):1148-1163
• 35. Zhu G, Chen X, Kong F, Kang L (2017) A continued fraction method for modeling and inversion of triaxial induction logging tool. In: 2017 IEEE microwaves, radar and remote sensing symposium (MRRS)). pp 210-204
• 36. Zhu GY, Gao MZ, Kong FM, Li K (2019) Application of Logging while drilling tool in formation boundary detection and geo-steering. Sensors 19(12):2754