Predicting channel sandstone thickness through a VIF-NRBO-XGBoost model

Abstract

High-precision sand thickness data are fundamentally important for optimizing exploration strategies in petroleum geology. In the Chengbei work area of the Jiyang Depression, the stratigraphic channels are chaotically developed, with channels of varying sizes in different strata overlapping, intersecting, and exhibiting narrow widths. The actual well-seismic relationship is poor. Therefore, individual seismic attributes in this area exhibit extremely low correlation with channel sandstone thickness. Conventional attributes such as root mean square amplitude show no distinct channel characteristics, necessitating the integration of multiple seismic attributes for effective prediction. Moreover, the high multicollinearity among seismic attributes introduces significant interference in prediction results. Therefore, this study integrates the Pearson correlation coefficient and variance inflation factor (VIF) to optimize seismic attribute selection, effectively eliminating redundant attributes and those with low correlation. To further enhance prediction accuracy and address the significant bias inherent in single-model predictions, this study introduces the ensemble learning XGBoost model, which integrates predictions from multiple weak learners to improve the precision of sandstone thickness estimate. The Newton-Raphson-based optimization algorithm was employed to fine-tune the XGBoost parameters. Results from test wells demonstrate a remarkable improvement in prediction accuracy, achieving reliable sandstone thickness estimation despite poor well-seismic correlations. This research provides valuable insights and offers a widely applicable methodology for predicting the thickness of complex channel sand bodies.

Keywords

River channel sand body / Thickness prediction / Variance inflation factor / Newton-Raphson based optimization optimization / XGBoost

Funding

This study is supported by the National Science and Technology Major Project Management Office for New Oil and Gas Exploration and Development (NO.  2024ZD1400102); and Shengli Oilfield Jiyang Depression Typical Lithologic Body Project Fund (NO. 30200020-24-ZC0613-0062).

References

[1] Wang S, Gu Z, Guo P, Zhao W. Comparative laboratory wettability study of sandstone, tuff, and shale using 12- MHz NMR T1- T 2 fluid typing: Insight of shale. SPE J. 2024; 29(9):4781-4803. doi: 10.2118/221496-PA

[2] Ehsan M, Chen R, Manzoor U, et al. Unlocking thin sand potential: A data-driven approach to reservoir characterization and pore pressure mapping. Geomech Geophys Geoenerg Georesour. 2024; 10:160. doi: 10.1007/s40948-024-00871-w

[3] Liu L, Jin J, Liu J, et al. Mechanical properties of sandstone under in-situ high-temperature and confinement conditions. Int J Miner Metall Mater. 2025; 32(4):778-787. doi: 10.1007/s12613-024-3047-9

[4] Manzoor U, Ehsan M, Hussain M, Bashir Y. Improved reservoir characterization of thin beds by advanced deep learning approach. Appl Comput Geosci. 2024; 23:100188. doi: 10.1016/j.acags.2024.100188

[5] Dormann CF, Elith J, Bacher S, et al. Collinearity: A review of methods to deal with it and a simulation study evaluating their performance. Ecography. 2012; 36:27-46. doi: 10.1111/j.1600-0587.2012.07348.x

[6] Widess MB. How thin is a thin bed? Geophysics. 1973; 38(6):1176-1180. doi: 10.1190/1.1440403

[7] Chung HM, Lawton DC. Amplitude responses of thin beds: Sinusoidal approximation versus Ricker approximation. Geophysics. 1995; 60(1):19-307. doi: 10.1190/1.1443750

[8] Long J. Neural network BP modeling of the relation between thin bed thickness and amplitude-frequency. Oil Geophys Prospect. 1995; 30(6):817-822.

[9] Zhuang D, Xiao C. Thin-bed thickness estimation using neural network. Oil Geophys Prospect. 1996; 31(3):394-399.

[10] Gridley JA, Partyka GA. Processing and Interpretational Aspects of Spectral Decomposition: 67th Annual International Meeting of the Society of Exploration Geophysicists. Expanded Abstracts; 1997. p. 1055-1058.

[11] Partyka GA, Gridley JA, Lopez JA. Interpretational aspects of spectral decomposition in reservoir characterization. Lead Edge. 1999; 18(3):353-360.

[12] Ye T, Su J, Liu X. Application of seismic frequency division interpretation technology in predicting continental sandstone reservoir in the west of Sichuan province. Geophys Prospect Petrol. 2008; 47(1):72-76.

[13] Sun L, Zheng X, Shou H, Li J, Li Y. Quantitative prediction of channel sand bodies based on seismic peak attributes in the frequency domain and its application. Appl Geophys. 2010; 7(1):10-17. doi: 10.1007/s11770-010-0009-y

[14] Barnes AE, Fink L, Laughlin K. Improving Frequency Domain Thin Bed Analysis. SEG Technical Program Expanded Abstracts. United States: SEG. 2004. p. 1929-1932. doi: 10.1190/1.1851175

[15] Wang Z, Gao D, Lei X, Wang D, Gao J. Machine learning-based seismic spectral attribute analysis to delineate a tight-sand reservoir in the Sulige gas field of central Ordos Basin, western China. Mar Petrol Geol. 2020; 113:104136. doi: 10.1016/j.marpetgeo.2019.104136

[16] Li W, Yue D, Wang W, et al. sing multiple frequency-decomposed seismic attributes with machine learning for thickness prediction and sedimentary facies interpretation in fluvial reservoirs. J Petrol Sci Eng. 2019; 177:1087-1102. doi: 10.1016/j.petrol.2019.03.017

[17] Yue D, Li W, Wang W, et al. Fused spectral-decomposition seismic attributes and forward seismic modelling to predict sand bodies in meandering fluvial reservoirs. Mar Petrol Geol. 2019; 99:27-44. doi: 10.1016/j.marpetgeo.2018.09.031

[18] Chopra S, Marfurt KJ. Seismic attributes - A historical perspective. Geophysics. 2005; 70(5):1SO-Z82. doi: 10.1190/1.2098670

[19] Liu S. A grain size profile prediction method based on combined model of extreme gradient boosting and artificial neural network and its application in sand control design. SPE J. 2024; 29(6):2988-3002. doi: 10.2118/219484-PA

[20] Zhang D, Qian L, Mao B, et al. A data-driven design for fault detection of wind turbines using random forests and XGboost. IEEE Acess. 2018; 6:21020-21031. doi: 10.1109/ACCESS.2018.2818678

[21] Yu B, Qiu W, Chen C, et al. SubMito-XGBoost: Predicting protein submitochondrial localization by fusing multiple feature information and eXtreme gradient boosting. Bioinformatics. 2020; 36(4):1074-1081. doi: 10.1093/bioinformatics/btz734

[22] Kounlavong K, Sadik L, Keawsawasvong S, Jamsawang P. Novel hybrid XGBoost-based soft computing models for predicting penetration resistance of buried pipelines in cohesive soils. Ocean Eng. 2024; 311(2):118948. doi: 10.1016/j.oceaneng.2024.118948

[23] Liu X, Zhang J, Bai Q, et al. Research and application of sandstone thickness prediction method based on NRBO-XGBoost optimization method. Comput Techn Geophys Geochem Explor. 2024; 46(2):146-153. doi: 10.3969/j.issn.1001-1749.2024.02.03 (in Chinese)

[24] Liu L, Li W, Du Y, et al. Reservoir prediction method of fusing frequency-decomposed seismic attributes using Stacking ensemble learning. Oil Geophys Prospect. 2024; 59(1):12-22. doi: 10.13810/j.cnki.issn.1000-7210.2024.01.002 (in Chinese)

[25] Cheng J, Sun J, Yao K, Xu M, Cao Y. A variable selection method based on mutual information and variance inflation factor. Spectrochim Acta A Mol Biomol Spectrosc. 2022; 268:120652. doi: 10.1016/j.saa.2021.120652

[26] Chen T, Guestrin C. XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. USA; 2016. p. 785-794. doi: 10.1145/2939672.2939785

[27] Sowmya R, Premkumar M, Jangir P. Newton raphson-based optimizer: A new population-based metaheuristic algorithm for continuous optimization problems. Eng Appl Artif Intell. 2024;128(C):107532. doi: 10.1016/j.engappai.2023.107532

[28] Ahmadianfar I, Bozorg-Haddad O, Chu X. Gradient-based optimizer: A new metaheuristic optimization algorithm. Information Sciences. 2020; 540:131-159. doi: 10.1016/j.ins.2020.06.037

[29] Lian Y, Luo J, Xue W, Zuo G, Zhang S. Cause-driven streamflow forecasting framework based on linear correlation reconstruction and long short-term memory. Water Resour Manag. 2022; 36:1661-1678. doi: 10.1007/s11269-022-03097-1