Enhancing well log curve synthesis with selective attention long short-term memory network

• Autorzy: Zhou Yuankai , Li Huanyu

Identyfikatory

DOI

10.1007/s11600-024-01367-7

• Języki publikacji: EN

Abstrakty

EN

In geological exploration projects, well log curves, as the primary carriers of information, are prone to data defects due to geological conditions, logging equipment, and unexpected events. This paper proposes a low-cost curve synthesis method based on deep learning. The method in this paper is based on a recurrent neural network, which can preserve contextual information in signals, crucial for logging data that vary with depth. An attention mechanism is employed to enhance the vanilla long short-term memory network, enabling it to capture larger spatial dependencies, but introducing a significant amount of matrix operations. To simplify this computation, a selector is designed to reduce the time complexity from O(n²) to O(n log n) . Two application scenarios are considered: predicting missing logging parameters using complete logging parameters and predicting missing segments of a well based on the original well data. Through validation and analysis, the proposed method demonstrates higher accuracy. This accurate, efficient, and cost-effective prediction method holds practical value in engineering applications.

Słowa kluczowe

EN

well logging; deep learning; self-attention mechanism; curve synthesis

PL

profilowanie odwiertów; głębokie uczenie; mechanizm samouwagi

• Wydawca: Instytut Geofizyki PAN ; Springer
• Czasopismo: Acta Geophysica
• Rocznik: 2025
• Tom: Vol. 73, no. 1
• Strony: 347--358
• Opis fizyczny: Bibliogr. 29 poz.

Twórcy

autor

Zhou Yuankai

• Division of Geophysical and Geochemical Exploration, China National Nuclear Corp Beijing Research Institute of Uranium Geology, Beijing, China

• https://orcid.org/0009-0003-7828-2271

autor

Li Huanyu

• College of Computer Science and Technology, Zhejiang University, Hangzhou, China

Bibliografia

• 1. Alaei HK, Salahshoor K (2012) The design of new soft sensors based on PCA and a neural network for parameters estimation of a petroleum reservoir. Pet Sci Technol 30(22):2294-2305
• 2. Ao Y, Li H, Zhu L, Ali S, Yang Z (2019) The linear random forest algorithm and its advantages in machine learning assisted logging regression modeling. J Pet Sci Eng 174:776-789
• 3. Bao L, Cao X, Yu C, Zhang G, Zhou W (2020) A deep neural network based feature learning method for well log interpretation. In: International conference on internet of things as a service. Springer, Cham, pp 543-556
• 4. Bhatt A, Helle HB (2002) Committee neural networks for porosity and permeability prediction from well logs. Geophys Prospect 50(6):645-660
• 5. Cai Y, Zhang J, Li Z, Guo Q, Song J, Fan H, Liu W, Qi F, Zhang M (2015) Outline of uranium resources characteristics and metallogenetic regularity in China. Acta Geol Sin Engl Ed 89(3):918-937
• 6. Czubek JA (1972) Pulsed neutron method for uranium well logging. Geophysics 37(1):160-173
• 7. Gao X, Lu WK, Li FY, Jiang XD (2013) The application of robust principal component analysis for weak seismic signal enhancement. In: 75th EAGE conference and exhibition incorporating SPE EUROPEC 2013. European Association of Geoscientists and Engineers, pp 348
• 8. Gei D, Brancolini G, De Santis L, Geletti R (2023) Well-log integration and seismic-to-well tie off George V Land (Antarctica). Geophys Prospect 72:685-704
• 9. Graves A (2012) Long short-term memory. In: Graves A (ed) Supervised sequence labelling with recurrent neural networks. Springer, Berlin, pp 37-45
• 10. Hethcoat MG, Edwards DP, Carreiras JM, Bryant RG, Franca FM, Quegan S (2019) A machine learning approach to map tropical selective logging. Remote Sens Environ 221:569-582
• 11. Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735-1780
• 12. Humphreys DR, Barnard RW, Bivens HM, Jensen DH, Stephenson WA, Weinlein JH (1983) Uranium logging with prompt fission neutrons. Int J Appl Radiat Isot 34(1):261-268
• 13. Kaźmierczuk M, Jarzyna J (2006) Improvement of lithology and saturation determined from well logging using statistical methods. Acta Geophys 54:378-398
• 14. Niculescu BM, Andrei GINA (2016) Principal component analysis as a tool for enhanced well log interpretation. Rev Roum Geophys 60:49-61
• 15. Rolon L, Mohaghegh SD, Ameri S, Gaskari R, McDaniel B (2009) Using artificial neural networks to generate synthetic well logs. J Nat Gas Sci Eng 1(4-5):118-133
• 16. Singh H, Ray MR (2021) Synthetic stream flow generation of River Gomti using ARIMA model. In: Advances in civil engineering and infrastructural development: select proceedings of ICRA- CEID 2019. Springer, Singapore, pp 255-263
• 17. Tang XM, Cheng CHA (2004) Quantitative borehole acoustic methods, vol 24. Elsevier, Amsterdam
• 18. Tung PS, Giang NN, Nhat ND, Dung TQ (2022) Application of neural networks in synthetic log generation. Int J Oil Gas Coal Technol 30(2):157-174
• 19. Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser PI (2017) Attention is all you need. In: Guyon I, Von Luxburg U, Bengio S, Wallach H, Fergus R, Vishwanathan S, Garnett R (eds) Advances in neural information processing systems, vol 30. Curran Associates Inc, New York
• 20. Wang G, Miao A, Gao H, Qiao P, Yi C, Li X (2015) Petrogeochemical characteristics of Nalinggou uranium deposit, Ordos Basin. Uranium Geol 31(Suppl. 1):273-282
• 21. Wu Y, Yuan M, Dong S et al (2018) Remaining useful life estimation of engineered systems using vanilla LSTM neural networks. Neurocomputing 275:167-179
• 22. Wu Q, Li Z, Wang Y, Cao C, Qiao B, Huang Y, Yu X (2023) Combination of seismic attributes using clustering and neural networks to identify environments with sandstone-type uranium mineralization. Acta Geophys 71:1-17
• 23. Yu Y, Si X, Hu C, Zhang J (2019) A review of recurrent neural networks: LSTM cells and network architectures. Neural Comput 31(7):1235-1270
• 24. Zaremba W, Sutskever I, Vinyals O (2014) Recurrent neural network regularization. arXiv preprint https://arxiv.org/abs/1409.2329
• 25. Zerrouki AA, Aifa T, Baddari K (2014) Prediction of natural fracture porosity from well log data by means of fuzzy ranking and an artificial neural network in Hassi Messaoud oil field, Algeria. J Pet Sci Eng 115:78-89
• 26. Zhang D, Yuntian C, Jin M (2018) Synthetic well logs generation via recurrent neural networks. Pet Explor Dev 45(4):629-639
• 27. Zhou H, Zhang S, Peng J, Zhang S, Li J, Xiong H, Zhang W (2021) Informer: beyond efficient transformer for long sequence timeseries forecasting. In: Proceedings of the AAAI conference on artificial intelligence, vol 35, no 12, pp 11106-11115
• 28. Zhu Q, Feng X, Li J, Sima X, Tang C, Xu Z, Wen S (2019) Mineralogy, geochemistry, and fluid action process of uranium deposits in the Zhiluo Formation, Ordos Basin, China. Ore Geol Rev 111:102984
• 29. Zhu L, Zhou X, Liu W, Kong Z (2023) Total organic carbon content logging prediction based on machine learning: a brief review. Energy Geosci 4(2):100098

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).