Identifying the potash reservoirs from seismic data by using convolution neural network, constrained by the waveform characteristics of potash reservoirs

• Autorzy: Zhu Lei , Zhang Fanchang , Xu Xunyong , Xu Wanglin

Identyfikatory

DOI

10.1007/s11600-023-01064-x

• Języki publikacji: EN

Abstrakty

EN

Exploration of potash resources under complex geological condition is particularly important. However, it is difficult to establish characteristic equations for direct prediction, since there is no direct relation between potash content (PC) and seismic response. To solve this problem, this paper proposed a potash reservoir prediction method by a specially designed convolution neural network (CNN) structure to train the special waveform and petrophysical characteristics of potash reservoirs. Considering that the potash reservoirs and petrophysical characteristics are not a one-to-one mapping, the prediction procedure is divided into two parts. First, a CNN is constructed for potash reservoir prediction, according to the spatial waveform characteristics of potash reservoirs. The mapping between potash reservoirs and waveform characteristics is used to obtain the potash reservoir probability data by the soft-max function. Then, another CNN for PC prediction is built based on the petrophysical characteristics of potash reservoirs. Meanwhile, according to the Hadamard criterion, the petrophysical characteristics of potash reservoir are constrained by the waveform characteristics. The two CNN models are used to directly predict the PC synergistically. Consequently, the bidirectional mapping problem can be alleviated and a loss function of the PC prediction CNN constrained with the waveform is obtained. Finally, by tuning the PC prediction CNN through the loss function, PC prediction is performed. The correlation between the predicted and true PC values can reach more than 80%.

Słowa kluczowe

EN

potash reservoir prediction; CNN; waveform characteristics; petrophysical characteristics; waveform constrained loss function

• Wydawca: Instytut Geofizyki PAN ; Springer
• Czasopismo: Acta Geophysica
• Rocznik: 2023
• Tom: Vol. 71, no. 6
• Strony: 2699--2714
• Opis fizyczny: Bibliogr. 44 poz., rys., tab.

Twórcy

autor

Zhu Lei

• Shandong Provincial Key Laboratory of Deep Oil & Gas, China University of Petroleum (East China), Qingdao, China

autor

Zhang Fanchang

• Shandong Provincial Key Laboratory of Deep Oil & Gas, China University of Petroleum (East China), Qingdao, China

• https://orcid.org/0000-0001-6802-6344

autor

Xu Xunyong

• PetroChina Tarim Oilfield Company, Xinjinag, China

autor

Xu Wanglin

• Research Institute of Petroleum Exploration and Development, Beijing, China

Bibliografia

• 1. Annan AP, Davis JL, Gendzwill D (1988) Radar sounding in potash mines, Saskatchewan, Canada. Geophysics 53(12):1556–1564
• 2. Baar CA (1970) Geological problems in Saskatchewan potash mining due to peculiar conditions during deposition of potash beds. Fourth Symp Salt 1:101–118
• 3. Bai Q, Yuan JH, Wang ZJ (2014) Industrial advances of soluble potash resources in China and overseas. Resour Ind 16(02):37–46
• 4. Bao RH, Liu XG (2017) Global potash resources and development and utilization. Fertil Health 44(3):66–69
• 5. Bao MF, Bao RH, Qi ZY (2017) Changes in global potash resources, production and marketing in 2017. Phosphate Compd Fertil 34(3):1–4
• 6. Biswas R, Sen MK, Das V, Mukerji T (2019) Pre-stack inversion using a physics-guided convolutional neural network. SEG Technical Program Expanded Abstracts. pp 4967–4971
• 7. Chen AQ, Yang S, Xu SL et al (2019) Sedimentary model of marine evaporites and implications for potash deposits exploration in China. Carbonates Evaporites 34(1):83–99
• 8. Chen XC, Kadry S, Meqdad MN, Crespo RG (2022) CNN supported framework for automatic extraction and evaluation of dermoscopy images. J Supercomput 78:17114–17131
• 9. Choromanska A, Henaff M, Mathieu M, Arous GB, LeCun Y (2015) The loss surfaces of multilayer networks. In: Artificial Intelligence and Statistics. pp 192–204
• 10. Cova D, Xie PG, Trinh PT (2020) Automated first break picking with constrained pooling networks. SEG Technical Program Expanded Abstracts. pp 1481–1485
• 11. Das V, Pollack A, Wollner U, Mukerji T (2019) Convolutional neural network for seismic impedance inversion CNN for seismic impedance inversion. Geophysics 84(6):R869–R878
• 12. Ding T, Liu CL, Zhao YJ et al (2019) Chlorine isotope analysis of Triassic salt rock and geological significance of ancient salt lake in Sichuan Basin. China Carbonates Evaporites 34(3):909–915
• 13. Gendzwill DJ (1969) Underground applications of seismic measurements in a Saskatchewan potash mine. Geophysics 34(6):906–915
• 14. Glorot X, Bordes A, Bengio Y (2011) Deep sparse rectifier neural networks. In: Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics. pp 315–323
• 15. Hinton GE, Osindero S, Teh YW (2006) A fast learning algorithm for deep belief nets. Neural Comput 18(7):1527–1554
• 16. Kovin ON (2012) Some results of acoustic reflection testing in Russian potash mines. J Environ Eng Geophys 5(1):39–45
• 17. Lan NY, Zhang FC (2022) Seismic data recovery using deep targeted denoising priors in an alternating optimization framework. Geophysics 87(4):1–58
• 18. Leiphart DJ, Hart BS (2001) Comparison of linear regression and a probabilistic neural network to predict porosity from 3-D seismic attributes in Lower Brushy Canyon channeled sandstones, southeast New Mexico. Geophysics 66(5):1349–1358
• 19. Li WG (1994) Comprehensive utilization of our potash resource. J Salt Lake Sci 2(3):65–68
• 20. Li WC (2018) Classifying geological structure elements from seismic images using deep learning. SEG Technical Program Expanded Abstracts. pp 4643–4648
• 21. Lin YT (1996) How to resolve the shortage of sylvite in China. Conserv Util Miner Resour 2:14–17
• 22. Lu P, Morris M, Brazell S, Comiskey C, Xiao Y (2018) Using generative adversarial networks to improve deep-learning fault interpretation networks. Lead Edge 37(8):578–583
• 23. Meng ZP, TianLei YDY (2008) Prediction models of coal bed gas content based on BP neural networks and its applications. J China Univ Min Technol 37(4):456–461
• 24. Mohamed IA, Hemdan M, Hosny A, Rashidy M (2019) High-resolution water-saturation prediction using geostatistical inversion and neural network methods. Interpretation 7(2):T455–T465
• 25. Mu YZ, Nie Z, Bu LZ (2016) Progress in study of potash resources of oil (gas) field brine in China. Adv Earth Sci 31(2):147–160
• 26. Pesowski MS, Larson RK (2000) Seismic exploration methods applied to potash mining: Risk analysis and mine planning. Seg Tech Progr Expand Abstr 19(1):1105–1110
• 27. Pu YT, Zhang XL (2018) Application of deep learning in first break picking of seismic data. SEG Global Meeting Abstracts. pp 19–21
• 28. Sang KH, Yin XY, Zhang FC (2021) Machine learning seismic reservoir prediction method based on virtual sample generation. Pet Sci 18(6):1662–1674
• 29. Shahraeeni MS, Curtis A, Chao G (2012) Fast probabilistic petrophysical mapping of reservoirs from 3D seismic data. Geophysics 77(3):O1–O19
• 30. Sharma N, Gupta S, Mehta P, Cheng XQ, Shankar A, Singh P, Nayak SR (2022) Offline signature verification using deep neural network with application to computer vision. J Electron Imaging 31(4):041210–041210
• 31. Stove G, Robinson M, Fourie L (2019) Identification and delineation of potash deposits in Saskatchewan, Canada using pulsed radar technology. Geophysics 85(1):1–46
• 32. Wang CN, Yu JQ, Chen L (2007) A review on the exploration of global potash resources with an emphasis on the past and present status of China with a methodological perspective. J Salt Lake Res 15(3):56–72
• 33. Wang BF, Zhang N, Lu WK (2019) Deep-learning-based seismic data interpolation: A preliminary result. Geophysics 84(1):V11–V20
• 34. Whittaker S D, Sharma R, Hallau D (2010) Characterization of a Slim‐Hole Gamma‐Ray Sonde for Potash exploration applications in a simple test pit environment. In: Symposium on the Application of Geophysics to Engineering and Environmental Problems Proceedings. pp 71–96
• 35. Wiener JM, Rogers JA, Rogers JR, Moll RF (1991) Predicting carbonate permeabilities from wireline logs using a back-propagation neural network. SEG Technical Program Expanded Abstracts. pp 285–288
• 36. Xiong W, Ji X, Ma Y, Wang YX, AlBinHassan NM, Ali MN, Luo Y (2018) Seismic fault detection with convolutional neural network. Geophysics 83(5):O97–O103
• 37. Yu SW, Ma JW et al (2019) Deep learning for denoising. Geophysics 84(6):V333–V350
• 38. Zhang FC, Yin XY, Wu GC (1997) Impedance inversion by using annealing neural network. J Univ Petrol China 21(6):16–28
• 39. Zhang FC, Liu HQ, Niu XM et al (2014) High resolution seismic inversion by convolution inversion. Oil Geophys Prospect 49(6):1165–1169
• 40. Zhang G, Wang Z, Chen Y (2018) Deep learning for seismic lithology prediction. Geophys J Int 215(2):1368–1387
• 41. Zhang X, Zhu ZJ, Wei YY (2019) Research on the effect of tectonism on the form and preservation of marine potash in Triassic Jialingjiang formation in Dianjiang Salt Basin, eastern Sichuan Basin. J Geomech 25(S1):072–077
• 42. Zhao X, Lu P, Zhang YY (2019) Swell-noise attenuation: a deep learning approach. Lead Edge 38(12):934–942
• 43. Zheng MP, Qi W, Zhang YS (2006) Present situation of potash resources and direction of potash search in China. Geol Bull China 25(11):1239–1246
• 44. Zheng MP, Hou XH, Zhang YS (2018) Progress in investigation of potash resources in western China. China Geol 1(3):392–401

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2024).