Three-dimensional Modeling and Application of Coalbed Methane Content Based on Potential-field Method and Machine Learning:No. 2 Coal Seam in the Zhongyu Block of the Qinshui Basin
Accurate evaluation of gas content in coalbed methane (CBM) reservoirs is essential for resource assessment and development planning. Current approaches often rely on predictive modeling but seldom incorporate geological models. With the increasing use of machine learning, conventional commercial modeling software is found to be inadequate for diverse predictive modeling requirements. A three-dimensional (3D) geological attribute modeling method for CBM gas content was developed by integrating potential-field theory with machine learning. The No. 2 coal seam of the Shanxi Formation in the Zhongyu block of the Qinshui Basin was selected as the case area. An implicit potential-field method combined with universal kriging interpolation was applied to construct the reservoir’s structural model. Support vector machines, random forests, stochastic gradient descent, and gradient boosting were then employed to predict gas content. K-fold cross-validation was used to optimize hyperparameters and assess model performance. Among the algorithms tested, stochastic gradient descent demonstrated the best predictive accuracy, with a coefficient of determination (R2) of 0.81 and an average relative error of 8.86%. A 3D gas content attribute model was subsequently generated by combining kriging interpolation with machine learning predictions. The resulting model shows significantly improved accuracy compared to conventional approaches and provides a clear, quantitative visualization of the spatial distribution of gas content. This integration of complex predictive models with geological modeling not only fully leverages the predictive capabilities of machine learning but also significantly enhances the accuracy of coal seam attribute models. The methodology presented here establishes a foundation for the future development of domestic software capable of 3D coalfield geological modeling.
HU Hui-ting, BI Wen-liang, YU Tian-teng, ZHOU Jun-jie, ZHANG Xiao, LIU Liang-liang, LI Guo-fu, ZHANG Sui-an.
Three-dimensional Modeling and Application of Coalbed Methane Content Based on Potential-field Method and Machine Learning:No. 2 Coal Seam in the Zhongyu Block of the Qinshui Basin[J].
Science Technology and Engineering, 2026, 26(19): 8119-8130 DOI:10.12404/j.issn.1671-1815.2505726
隐式建模即使用函数f(x,y,z)插值来表示曲面,并且同一曲面的函数值相同,然而如何准确表示各曲面间的空间关系并使其符合地质认识是关键。本文中所采用的基于势场理论的建模方法是由Lajaunie 等[23]提出,并在Calcagno等[24]进一步优化的基础上得以实现,如今该方法已成功应用于建模软件GeoModeller中,并且De la Varga等[25]基于此开源了Python 的建模库GemPy,由此使得该方法得到进一步的推广。在GemPy的基础上进一步开发软件进行煤层气储层含气量三维建模。
XuFengyin, HouWei, XiongXianyue, et al. The status and development strategy of coalbed methane industry in China[J]. Petroleum Exploration and Development, 2023, 50(4): 669-682.
HuoChao, GuoHaixiao, WangLei, et al. Research progress of deep coalbed methane exploration and development in china under the dual carbon background[J]. Science Technology and Engineering, 2025, 25(14): 5705-5720.
LiuDameng, JiaQifeng, CaiYidong. Research progress on coalbed methane reservoir geology and characterization technology in China[J]. Coal Science and Technology, 2022, 50(1): 196-203.
[7]
LiuB, ChangS, ZhangS, et al. Coalbed methane gas content and its geological controls: research based on seismic-geological integrated method[J]. Journal of Natural Gas Science and Engineering, 2022, 101: 104510.
ChenQiyu. Three-dimensional stochastic modeling method of geological entities based on multiple-point geostatistics[D]. Beijing: China University of Geosciences-Beijing, 2018.
LiuXudong, SangShuxun, ZhouXiaozhi, et al. Numerical simulation of CO2-ECBM based on a three-dimensional heterogeneous geological model of coal reservoirs[J]. Journal of China Coal Society, 2023, 48(7): 2773-2790.
MengZhaoping, GuoYansheng, ZhangJixing. Application and prediction model of coalbed methane content based on logging pa rameters[J]. Coal Science and Technology, 2014, 42(6): 25-30.
ChenTao, ZhangZhansong, ZhouXueqing, et al. Prediction model of coalbed methane content based on well logging parameter optimization[J]. Coal Geology & Exploration, 2021, 49(3): 227-235, 243.
[18]
ZouG, PengS, YinC, et al. Seismic studies of coal bed methane content in the west coal mining area of qinshui basin[J]. International Journal of Mining Science and Technology, 2013, 23(6): 795-803.
MaXiaohe, TanChengqian, GuoZekun. Analysis and prediction of main controlling factors of gas content in coalbed of M coalfield in Bowen Basin[J]. Science Technology and Engineering, 2023, 23(11): 4586-4595.
QinRuibao, YeJianping, LiLi, et al. Artificial-intelligence and machine-learning models of coalbed methane content based on geophysical logging data: a case study in Shizhuang South block of Qinshui Basin, China[J]. Geophysical Prospecting for Petroleum, 2023, 62(1): 68-79.
[23]
ZhouF, YaoG, TysonS. Impact of geological modeling processes on spatial coalbed methane resource estimation[J]. International Journal of Coal Geology, 2015, 146: 14-27.
LiYong, GaoShuang, WuPeng, et al. Evaluation and correction of prediction model for free gas content in deep coalbed methane: a case study of deep coal seams in the eastern margin of Ordos Basin[J]. Acta Petrolei Sinica, 2023, 44(11): 1892-1902.
BaiZe, TanMaojin, BaiYang, et al. Coalbed methane content prediction based on joint optimization of log interpretation model and mean impact value method[J]. Progress in Geophysics, 2024, 39(5): 1863-1873.
ZhuQingzhong, MengZhaoping, HuangPing, et al. Isothermal adsorption characteristics of coal reservoir and its gas content prediction in Qinnan-Xiadian Block of Qinshui Basin[J]. Coal Geology & Exploration, 2016, 44(4): 69-72.
ZhangPingsong, LiJie, LiShenglin, et al. Application status of 3D geological modeling in the development of coal mine intelligence[J]. Science Technology and Engineering, 2022, 22(5): 1725-1740.
[32]
LaurentG, AilleresL, GroseL, et al. Implicit modeling of folds and overprinting deformation[J]. Earth and Planetary Science Letters, 2016, 456: 26-38.
LuChengang, ZhangSui’an, BaiTiefeng, et al. Improved modeling of 3D gas content attributes of CBM in Junlian Region of Southern Sichuan[J]. Coal Science and Technology, 2019, 47(8): 219-225.
ChenQiyu, LiuGang, HeZhenwen, et al. Current situation and prospect of structure-attribute integrated 3D geological modeling technology for geological big data[J]. Bulletin of Geological Science and Technology, 2020, 39(4): 51-58.
[37]
WellmannF, CaumonG. 3-D structural geological models: concepts, methods, and uncertainties[J]. Advances in Geophysics, Elsevier, 2018, 59: 1-121.
ZhongDeyun, WangLiguan. Implicit and automatic modeling method for complex coal seam geological bodies integrating fault potential fields[J]. Journal of China Coal Society, 2025, 50(3): 1705-1716.
[40]
LajaunieC, CourriouxG, ManuelL. Foliation fields and 3D cartography in geology: principles of a method based on potential interpolation[J]. Mathematical Geology, 1997, 29(4): 571-584.
[41]
CalcagnoP, ChilesJ P, CourriouxG, et al. Geological modelling from field data and geological knowledge: part I. modelling method coupling 3D potential-field interpolation and geological rules[J]. Physics of the Earth and Planetary Interiors, 2008, 171(1): 147-157.
[42]
De la VargaM, SchaafA, WellmannF. GemPy 1.0: open-source stochastic geological modeling and inversion[J]. Geoscientific Model Development, 2019, 12(1): 1-32.
[43]
LorensenW E, ClineH E. Marching cubes: a high resolution 3D surface construction algorithm[J]. Computer Graphics, 1987, 21(4): 163-169.
QinYong, LiuHuanjie, SangShuxun, et al. The studies on gasbearing characteristics of the Upper Paleozoic coal seam s in the southern Shanxi, China I. gas-bearing seams in the indicated areas[J]. Coal Geology & Exploration, 1997, 25(4): 25-30.
QinYong, LiuHuanjie, FanBinghuan, et al. Study on gas potential of the upper Paleozoic in South Shanxi Province(part2): prediction of coalbed gas inferred area and prediction area[J]. Coal Geology & Exploration, 1997, 25(6): 20-24.
[48]
ZouZ, LiuD, CaiY, et al. Geological factors and reservoir properties affecting the gas content of coal seams in the Gujiao Area, Northwest Qinshui Basin, China[J]. Energies, 2018, 11(5): 1044.
LiuChanglong. From machine learning to deep learning: effective development based on scikit-learn and TensorFlow[M]. Beijing: Electronic Industry Press, 2019: 80-94.
YaoHuifang, ZhaoMingkun, ChenQiang. Research on classification of tight sandstone gas reservoir in coal measures based on machine learning: a case from DJ Block of Ordos Basin[J]. Coal Science and Technology, 2022, 50(6): 260-270.
YangYanhui, WangYuting, LiuZhong, et al. Strategy and practice of high-efficiency development of high-rank coalbed methane in the southern Qinshui Basin[J]. China Petroleum Exploration, 2024, 29(4): 17-30.
WangZhangfei, FangHuihuang, SangShuxun, et al. Three-dimensional visualization of pore structure of low permeability coal seam in lianghuai coal field and its connectivity[J]. Science Technology and Engineering, 2024, 24(26): 11195-11207.
HeWeimin. Three-point method for calculation of the occurrence of buried active fault[J]. Geotectonica et Metallogenia, 2022, 46(1): 36-47.
[61]
BerettaF S, CostaJ F C L, KoppeJ C. Reducing coal quality attributes variability using properly designed blending piles helped by geostatistical simulation[J]. International Journal of Coal Geology, 2010, 84(2): 83-93.
[62]
OliverM A, WebsterR. A tutorial guide to geostatistics: computing and modelling variograms and kriging[J]. Catena, 2014, 113: 56-69.
[63]
GroseL, AilleresL, LaurentG, et al. LoopStructural 1. 0: time-aware geological modelling[J]. Geoscientific Model Development, 2021, 14(6): 3915-3937.
[64]
JessellM, GuoJ, LiY, et al. Into the noddyverse: a massive data store of 3D geological models for machine learning and inversion applications[J]. Earth System Science Data, 2022, 14(1): 381-392.