Application of Artificial Intelligence in the Identification and Characterization of Fracture–Cavity Carbonate Reservoirs

Xingyu Hou; Linwei Li; Yanxiong Liu; Yan Chen

doi:10.54691/3e2vww04

Authors

Xingyu Hou
Linwei Li
Yanxiong Liu
Yan Chen

DOI:

https://doi.org/10.54691/3e2vww04

Keywords:

Artificial Intelligence; Carbonate Reservoirs; Multi-source Information Fusion.

Abstract

Carbonate reservoirs are among the most important carriers of global hydrocarbon resources. However, the accurate identification and quantitative characterization of their core storage space—the fracture–cavity system—remain a long-standing global challenge in petroleum exploration and development. Over the past decade, rapid advances in artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), have provided transformative tools to address the characterization difficulties arising from the strong heterogeneity and multiscale nature of carbonate reservoirs.This review systematically summarizes the applications of AI technologies in the study of fracture–cavity carbonate reservoirs from 2015 to 2025. The geological and engineering backgrounds, as well as the necessity of introducing AI techniques, are first outlined. Taking technological evolution as the main thread, research progress is reviewed from traditional machine learning to deep learning and further to intelligent methods based on multimodal data fusion. The review focuses on key research directions, including intelligent identification of fractures and cavities, parameter prediction, three-dimensional modeling, and data integration, with an emphasis on methodologies and representative applications. Current challenges are discussed in depth from three core dimensions: data, models, and knowledge-driven constraints. Finally, future research trends are prospected, highlighting the path toward interpretable, strongly generalizable, high-fidelity, and fully integrated intelligent workflows.

Downloads

Download data is not yet available.

References

[1] Zhang Chengsen, Xiao Chengwen, Liu Xingli, et al. Application of long-range acoustic logging in the evaluation of fracture–cavity carbonate reservoirs. Xinjiang Petroleum Geology, 2011, 32(03): 325–328.

[2] Liang Qimin , Chen Rou , Yan Weiwei. Research progress and challenges oflow mechanisms and well testing models in carbonate reservoirs.Frontiers in Energy Research.2023,10.

[3] Al-Obaidani, H.S., Farfour, M. Seismic characterization of carbonate stringers using machine learning techniques: an example from the western flank of South Oman Salt Basin. Sci Rep 14, 30844 (2024).

[4] Zongjie Li, Haiying Li, Jun Liu, et al, 3D seismic intelligent prediction of fault-controlled fractured-vuggy reservoirs in carbonate reservoirs based on a deep learning method, Journal of Geophysics and Engineering, Volume 21, Issue 2, April 2024, Pages 345–358.

[5] Kouassi, A.K.F., Pan, L., Wang, X. et al. Identification of Karst Cavities from 2D Seismic Wave Impedance Images Based on Gradient-Boosting Decision Trees Algorithms (GBDT): Case of Ordovician Fracture-Vuggy Carbonate Reservoir, Tahe Oilfield, Tarim Basin, China. Energies 2023, 16, 643.

[6] Pei, J.; Zhang, Y. Prediction of Reservoir Fracture Parameters Based on the Multi-Layer Perceptron Machine-Learning Method: A Case Study of Ordovician and Cambrian Carbonate Rocks in Nanpu Sag, Bohai Bay Basin, China. Processes 2022, 10, 2445.

[7] Liu B, Yasin Q, Sohail GM, et al. Seismic characterization of fault and fractures in deep buried carbonate reservoirs using CNN‐LSTM based deep neural networks. Geoenergy Sci Eng. 2023; 229: 212126.

[8] Feng X, Zhao K, Zhang J, et al. Carbonate reservoir fracture-cavity system identification based on the improved YOLOv5s deep learning algorithm. Energy Sci Eng. 2024; 12: 2643-2660.

[9] Yuslandi, R., Sautter, B., Poppelreiter, M.C. The various controls on the platform architecture of carbon-ate buildups in the Central Luconia province, offshore Sarawak. Petroleum and Coal,2019. 61, 1425–1437.

[10] Mousavi S, Afghah F, Acharya UR SleepEEGNet: Automated sleep stage scoring with sequence to sequence deep learning approach. PLoS ONE.2019, 14(5): e0216456.

[11] Ketineni, S. P, Ertekin.et al Structuring an integrative approach for field development planning using artificial intelligence and its application to an offshore oilfield. In Society of Petroleum Engineers - SPE Annual Technical Conference and Exhibition, ATCE 2015 pp. 2040-2060.

[12] Hussein, Outcrop-to-Subsurface Approaches for Fracture Characterization in Carbonate Reservoirs, Kurdistan Region, Iraq, Egyptian Journal of Petroleum:Vol.34: Iss.1, Article5.

[13] Wu Yunlong, Liu Meiwei, Wang Guodong, et al. Identification and characterization of multi-scale pores and fractures in carbonate reservoirs: A case study of the Maokou Formation carbonate gas reservoir. *Journal of Chongqing University of Science and Technology (Natural Science Edition)*, 2025, 27(01): 74–82.

[14] Jun Yao, Qin Huang Z .Fractured Vuggy Carbonate Reservoir Simulation[M].Springer, Berlin, Heidelberg: 2016.

[15] Tian H ,Zhang J ,Xin Y , et al.Reservoir characteristics and forming conditions for the Middle Triassic Leikoupo Formation in the western Sichuan Basin, China[J].Journal of Natural Gas Geoscience, 2019, 4 (2):101-110.

[16] Pan, J.-G & Wei, P.-S & Cai, Z.-X & Yang, et al Reservoir architectural system in the Middle-Lower Ordovician carbonate rock of Tazhong areas in Tarim. Diqiu Kexue - Zhongguo Dizhi Daxue Xuebao/Earth Science - Journal of China University of Geosciences. 2012,37. 751-762.

[17] Zhang Ying, Cui Yanfei. Study on economic evaluation models for the development of fracture–cavity carbonate reservoirs. Shandong Chemical Industry, 2018, 47(13): 85–88.

[18] Ma,Y.S,Cai,X.Y,Li,H.L,Zhu,D.et al.New understanding of the development mechanism of deep-super deep carbonate reservoirs and the direction of ultra-deep oil and gas exploration. Geol. J. 30.06 2023, 1–13.

[19] Al-Obaidani, H.S, Farfour, M. Seismic characterization of carbonate stringers using machine learning techniques: an example from the western flank of South Oman Salt Basin. Sci Rep 14, 30844 2024.

[20] Hou, J; Zhao, L; Zeng, X;et al. Characterization and Evaluation of Carbonate Reservoir Pore Structure Based on Machine Learning. Energies 2022, 15, 7126.

[21] Xiao Pengfei. Application of pre-stack seismic fluid identification technology in fracture–cavity carbonate reservoirs. Petroleum Geophysical Prospecting, 2020, 59(03): 450–461.

[22] Wang Yan , Xie Pengfei , Zhang Huitao , et al.Fracture-vuggy carbonate reservoir characterization based on multiple geological information fusion, Frontiers in Earth Science, Volume 11 – 2023, 10.3389.

[23] Zhang Binxin, Zhang Guanjie, Liu Jingshou, et al. Research progress on identification and carving methods for fracture–cavity carbonate reservoirs. Progress in Geophysics, 2024, 39(2): 580–593.

[24] Li Guohui, Yuan Jingyi, Luo Haoyu, et al. Quantitative carving techniques for fracture–cavity carbonate reservoirs in the Harahatang area, Tarim Basin. China Petroleum Exploration, 2015, 20(4): 24–29.

[25] Hendry, J,Burgess, P,Hunt, D. et al.Seismic Characterization of Carbonate Platforms and Reservoirs. Geological Society, London, Special Publications,2021. 509, 1–28.

[26] Alatefi, S, Abdel Azim, R,Alkouh, A.et al.Integration of Multiple Bayesian Optimized Machine Learning Techniques and Conventional Well Logs for Accurate Prediction of Porosity in Carbonate Reservoirs. Processes 2023, 11, 1339.