Assessment and XGBoost prediction of diagenetic facies in deepwater low-permeability Lingshui Formation of Baodao Sag B Gas Field
DOI:
https://doi.org/10.62813/see.2025.01.05Keywords:
Diagenetic facies evaluation; XGBoost algorithm; Distribution of diagenetic facies; Low-permeability reservoir; Deep water and deep burial; Baodao Sag, Diagenetic facies evaluation, XGBoost ensemble learning algorithm, Low-permeability reservoir, Deep water and deep burial, Baodao Sag B Gas FieldAbstract
To elucidate the distribution of favorable diagenetic facies in deepwater, deep-burial, and low-permeability sandstones, the Lingshui III Formation member in the Baodao Sag in Qiongdongnan Basin was analyzed using petrophysical, cast thin sections, mercury injection, XRD, and SEM data. Diagenetic facies were classified upon mineral composition, intensity, porosity, and property variations. XGBoost algorithm was then employed for identifying diagenetic facies. Combined with sedimentary facies, fault features, and burial depth, the planar distribution patterns were revealed. Results show that the reservoir exhibits medium-low porosity and low-extra low permeability, with dominant residual intergranular and secondary dissolution pores. Four diagenetic facies are identified, i.e., carbonate tight cementation, weak dissolution-strong compaction, strong dissolution unstable components, and weak dissolution-weak compaction. The latter two facies with better-developed pore throats and superior reservoir properties are favorable reservoirs. XGBoost model achieves 93.6% accuracy in predicting diagenetic facies with an increment of 41.8% over traditional methods. The strong dissolution facies primarily occurs in subaqueous distributary channels near northern deep faults, while the weak dissolution-weak compaction facies, located in structurally higher channels farther from these faults, is optimal for well placement. Findings herein are instructive for developing deepwater, deep-burial, low-permeability gas reservoirs.
References
Al-Anazi, A. and Gates, I.D., 2010. A support vector machine algorithm to classify lithofacies and model permeability in heterogeneous reservoirs. Engineering Geology, 114(3-4), 267-277.
Bestagini, P., Lipari, V. and Tubaro, S., 2017. A machine learning approach to facies classification using well logs. In SEG Technical Program Expanded Abstracts, 2137-2142.
Bhatt, A. and Helle, H.B., 2002. Determination of facies from well logs using modular neural networks. Petroleum Geoscience, 8(3), 217-228.
Cao, J., Wang, Q., Fan, L., et al., 2021. Characteristics and diagenetic facies analysis of deepwater tight sandstone reservoirs: a case study of Chang 7 oil reservoir in Xifeng area, Ordos Basin. Journal of Northwest University (Natural Science Edition), 51, 80-94. (In Chinese)
Chen, H. and Zhang, L., 2023. Development patterns of high-quality deep clastic reservoirs based on diagenetic facies: a case study of the Kepingtage Formation in Well Shun 9 area, Tarim Basin. Journal of Palaeogeography, 25, 1379-1393. (In Chinese)
Das, S., Singha, D.K., Mandal, P.P., et al., 2024. Identification of lithofacies from well log data in the upper Assam basin using machine learning techniques. Acta Geophysica, 72(5), 3191-3210.
Han, R., Wang, Z., Wang, W., et al., 2021. Lithology identification of igneous rocks based on XGboost and conventional logging curves, a case study of the eastern depression of Liaohe Basin. Journal of Applied Geophysics, 195, p.104480.
Guo, S., Liao, G., Liang, H., et al., 2021. Major breakthrough and significance of natural gas exploration in deepwater area of Well BD21, Qiongdongnan Basin. China Petroleum Exploration, 26, 49-59. (In Chinese)
Huang, L., Yan, J., Liu, M., et al., 2022. Logging identification and application of diagenetic facies in deep tight sandstone gas reservoirs: a case study of the third member of Xujiahe Formation in Dayi area, Western Sichuan Depression. Journal of China University of Mining & Technology, 51, 107-123. (In Chinese)
Li, M., Zhang, L., Li, Z., et al., 2021. Diagenetic facies classification and logging identification of tight sandstone in the lower conglomerate member of Lower Jurassic Ahe Formation, Tarim Basin: a case study of Yiqikelike area, Kuqa Depression. Natural Gas Geoscience, 32, 1559-1570. (In Chinese)
Liu, H., Zhang, C., Gai, S., et al., 2024. Diagenetic facies identification and distribution prediction of Jurassic ultra-deep tight sandstone reservoirs in Yongjin Oilfield, Junggar Basin. Petroleum Geology and Recovery Efficiency, 31, 13-22. (In Chinese)
Liu, J., 2022. Prediction of high-quality tight sandstone reservoirs based on artificial intelligence: a case study of Yanchang Formation in Zichang Area, Ordos Basin. Chang'an University. (In Chinese)
Ma, L., Xiao, H., Tao, J., et al., 2022. Intelligent lithology classification method based on gradient boosting decision tree algorithm. Petroleum Geology and Recovery Efficiency, 29, 21-29. (In Chinese)
Ma, Y.Z., 2011. Lithofacies clustering using principal component analysis and neural network: applications to wireline logs. Mathematical Geosciences, 43(4), 401-419.
Marchand, A.M., Apps, G., Li, W., et al., 2015. Depositional processes and impact on reservoir quality in deepwater Paleogene reservoirs, US Gulf of Mexico. AAPG Bulletin, 99(9), 1635-1648.
Sun, Y., Huang, Y., Liang, T., et al., 2020. Logging identification of complex carbonate lithology based on XGBoost algorithm. Lithologic Reservoirs, 32, 98-106. (In Chinese)
Wang, L., Tian, Y., Yao, B., et al., 2020. Machine learning analyses of low salinity effect in sandstone porous media. Journal of Porous Media, 23(7).
Xu, C., 2022. New progress and breakthrough directions of offshore oil and gas exploration in China. China Offshore Oil and Gas, 34, 9-16. (In Chinese)
Xu, C., Deng, Y., Wu, K., et al., 2023. Discovery and geological significance of Baodao 21-1 large gas field in highly active passive continental margin basin, northern South China Sea. Acta Petrolei Sinica, 44, 713-729. (In Chinese)
Xu, C. and You, L., 2022. North slope transition zone of Songnan-Baodao sag in Qiongdongnan Basin and its control on medium and large gas fields, South China Sea. Petroleum Exploration and Development, 49(6), 1229-1242.
You, L., Quan, Y., Tuo, L., et al., 2023. Natural gas origin and migration pathway system of Baodao 21-1 gas field in deepwater area, Qiongdongnan Basin. Oil & Gas Geology, 44, 1270-1278. (In Chinese)
Zhang, H., Ye, Q., Huan, J., et al., 2024. Complex lithofacies identification based on component indicative factors: a case study of deepwater deep-layer low-permeability gas reservoirs in Baodao Sag, South China Sea. Marine Geology Frontiers, 40, 87-95. (In Chinese)
Additional Files
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Ruxian Chen, Wei Zhou, Chong Zhang, Di Meng, Rui Yang
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.