Logging prediction model of coal seam gas content in Southern Yanchuan gas field

LIU Xiao; CHEN Zhenlong; YANG Song; LI Song; CHANG Chuang

doi:10.12097/gbc.2024.03.037

2025 Vol. 44, No. 5

Article Contents

Article navigation > Geological Bulletin of China > 2025 > 44(5): 792-800

Next Article Previous Article

LIU Xiao, CHEN Zhenlong, YANG Song, LI Song, CHANG Chuang. 2025. Logging prediction model of coal seam gas content in Southern Yanchuan gas field. Geological Bulletin of China, 44(5): 792-800. doi: 10.12097/gbc.2024.03.037

Citation:

LIU Xiao, CHEN Zhenlong, YANG Song, LI Song, CHANG Chuang. 2025. Logging prediction model of coal seam gas content in Southern Yanchuan gas field. Geological Bulletin of China, 44(5): 792-800. doi: 10.12097/gbc.2024.03.037

Logging prediction model of coal seam gas content in Southern Yanchuan gas field

1.
Linfen Coalbed Methane Company, SINOPEC, Linfen 041099, Shanxi, China
2.
Key Laboratory of Deep CBM Exploration and Production, SINOPEC, Nanjing 210011, Jiangsu, China
3.
School of Energy Resources, China University of Geosciences (Beijing), Beijing 100083, China

Received Date: 22 March 2024

Revised Date: 12 August 2024

Publish Date: 15 May 2025

Abstract

Abstract

Objective
The coal seam gas content is a core parameter for resource assessment and development, but current gas content prediction models generally suffer from issues such as insufficient accuracy and weak generalization ability, which hinder the exploration and development of coalbed methane.
Methods
Based on the logging response characteristics of coal seam gas content in Southern Yanchuan gas field, the MIV (Mean Impact Value) method was utilized to optimize the logging parameters. The BP neural network and random forest algorithm were introduced to establish a high−precision coal seam gas content prediction model.
Results
Compared to the traditional multiple linear regression model, both the BP neural network and random forest models achieved notably higher prediction accuracy, with the random forest model performing even better.
Conclusions
The Random Forest model is more suitable for predicting the coalbed methane content in the study area. Based on the model's prediction, the distribution range of gas content in the gas field is 4.84~21.83 m³/t, with an average of 11.63 m³/t. Spatially, the gas content of coal seam increases gradually from southeast to northwest, and its variation law is generally consistent with the buried depth pattern of coal seam. Vertically, with the increase of buried depth, the gas content of coal seam increases gradually, but the dispersion degree of gas content distribution increases.
- gas content /
- logging parameters /
- MIV /
- BP neural network /
- random forest

FullText(HTML)

References

[1]	Bharadwaj P K B, Kanagachidam−baresan G R. 2021. Pattern recognition and machine learning. Google Scholar
[2]	Breiman. 2001. Random Forests[J]. Mach−Learn, 45(1): 5−32. doi: 10.1023/A:1010933404324 CrossRef Google Scholar
[3]	Chang C, Li S, Tang D Z, et al. 2023. In−situ stress calculation for coal reservoirs based on log parameters: A case study of the southern Yanchuan block[J]. Coal Geology & Exploration, 51(5): 23−32 (in Chinese with English abstract). Google Scholar
[4]	Chen S D, Hou W, Tang D Z, et al. 2024. Effects of depth on gas−bearing properties of coal reservoirs and their coup−ling relationships with coalbed methane accumulation[J]. Coal Geology & Exploration, 52(2): 52−59 (in Chinese with English abstract). Google Scholar
[5]	Chen T, Zhang Z S, Zhou X Q, et al. 2021. Prediction model of coalbed methane content based on well logging parameter optimization[J]. Coal Geology & Exploration, 49(3): 227−235 (in Chinese with English abstract). Google Scholar
[6]	Dietterich T G. 2000. Ensemble methods in machine learning[C]//Proc International Work Shop on Multiple Classifier Systems. Google Scholar
[7]	Feng P, Li S, Tang D Z, et al. 2022. Application of support vector machine in prediction of coal seam stress[J]. Geoscience, 6(5): 1333−1340 (in Chinese with English abstract). Google Scholar
[8]	Fu X, Qin Y, Wang G, et al. 2009. Evaluation of gas content of coalbed methane reservoirs with the aid of geophysical logging technology[J]. Fuel, 88: 2269−2277. doi: 10.1016/j.fuel.2009.06.003 CrossRef Google Scholar
[9]	Ge X, Liu D, Cai Y, et al. 2018. Gas content evaluation of coalbed methane reservoir in the Fukang area of Southern Junggar Basin, Northwest China by multiple geophysical logging methods[J]. Energies, 11: 1867. doi: 10.3390/en11071867 CrossRef Google Scholar
[10]	Ghosh S, Chatterjee R, Paul S, et al. 2014. Designing of plug−in for estimation of coal proximate parameters using statistical analysis and coal seam correlation[J]. Fuel, 134: 63−73. doi: 10.1016/j.fuel.2014.05.023 CrossRef Google Scholar
[11]	Hawkins J M, Schraufnagel R A, Olszewski A J. 1992. Estimating coalbed gas content and sorption isotherm using well log data[C]//SPE Annual Technical Conference and Exhibition. SPE: SPE−24905−MS. Google Scholar
[12]	Hou X, Liu S, Zhu Y, et al. 2020. Evaluation of gas contents for a multi−seam deep coalbed methane reservoir and their geological controls: In situ direct method versus indirect method[J]. Fuel, 265: 116917. doi: 10.1016/j.fuel.2019.116917 CrossRef Google Scholar
[13]	Kim A G. 1977. Estimating methane content of bituminous coalbeds from adsorption data[M]. Department of the Interior, Bureau of Mines. Google Scholar
[14]	Li D D, Wang Z G, Jiang W P, et al. 2022. Approach study on coal gas content quantitative assessment based on well logging parameters—a case study of coal No. 15 in Shouyang block[J]. Coal Geology of China, 34(8): 34−40 (in Chinese with English abstract). Google Scholar
[15]	Lin H, Sun X Y, Song X X, et al. 2023. A model for shale gas well production prediction based on improved artificial neural network[J]. Petroleum Reservoir Evaluation and Development, 13(3): 467−473 (in Chinese with English abstract). Google Scholar
[16]	Li Q, Liu Y S, Huang Z Q, et al. 2016. Exploring mechanism and numerical simulation of rock breaking process using PDC bit[J]. Mechanical Science and Technology, 35(2): 203−209 (in Chinese with English abstract). Google Scholar
[17]	Li S, Qin Y, Tang D Z, et al. 2023. A comprehensive review of deep coalbed methane and recent developments in China[J]. International Journal of Coal Geology, 279: 104369. doi: 10.1016/j.coal.2023.104369 CrossRef Google Scholar
[18]	Liu X. 2024. Comparison of seam network morphology in coal reservoirs under different fracturing scales: A case of Yanchuannan CBM Gas Field[J]. Petroleum Reservoir Evaluation and Development, 14(3): 510−518 (in Chinese with English abstract). Google Scholar
[19]	Li Z C, Du W F, Hu J K, et al. 2018. Interpretation method of gas content in logging of Linxing block in Ordos Basin[J]. Coal Journal, 43(S2): 490−498 (in Chinese with English abstract). Google Scholar
[20]	Meng Z P, Tian Y D, Li G F. 2020. Geology theory and method of coalbed methane development[M]. Beijing: Science Press (in Chinese with English abstract). Google Scholar
[21]	Meng Z P, Tian Y D, Lei Y. 2008. Prediction models of coal bed gas content based on BP neural networks and its applications[J]. Journal of China University & Mining and Technology, (4): 456−461. (in Chinese with English abstract). Google Scholar
[22]	Mullen M J. 1989. Coalbed Methane Resource evaluation from wireline logs in the northeastern San Juan Basin: A case study[C]//SPE Rocky Mountain Petroleum Technology Conference/Low-Permeability Reservoirs Symposium. SPE: SPE-18946-MS. Google Scholar
[23]	Qin R B, Ye J P, Li L, et al. 2023. 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, 62(1): 68−79 (in Chinese with English abstract). Google Scholar
[24]	Rumelhart, D E, Hinton, G E, Williams R J. 1986. Learning representations by back propagating errors[J]. Nature, 323(6088): 533−536. doi: 10.1038/323533a0 CrossRef Google Scholar
[25]	Roslin A, Esterle J S. 2015. Electro facies analysis using high−resolution wireline geophysical data as a proxy for inertinite−rich coal distribution in Late Permian Coal Seams, Bowen Basin[J]. International Journal of Coal Geology, 152: 10–18. Google Scholar
[26]	Shao X, Sun Y, Sun J, et al. 2013. Log interpretation for coal petrologic parameters: A case study of Hancheng mining area, Central China[J]. Petroleum Exploration and Development, 40: 599–605. Google Scholar
[27]	Shen L, Wang C Z, Ning C Q, et al. 2023. Well−log lithofacies classification based on machine learning for Chang−7 member in Longdong area of Ordos Basin[J]. Petroleum Reservoir Evaluation and Development, 13(4): 525−536 (in Chinese with English abstract). Google Scholar
[28]	Song J, Su X, Wang Q, et al. 2017. A new method for calculating gas content of coal reservoirs with consideration of a micro−pore overpressure environment[J]. Natural Gas Industry B, 4: 182−188. doi: 10.1016/j.ngib.2017.08.003 CrossRef Google Scholar
[29]	Wang X, Han J Q, Zan L, et al. 2024. Logging evaluation of shale oil in the second member of Funing Formation of Qintong Sag, Subei Basin[J]. Petroleum Reservoir Evaluation and Development, 14(3): 364−372 (in Chinese with English abstract). Google Scholar
[30]	Wei L, Zhang Y, Li W. 2000. Research on the method of electric character logging in the third series coal of Baoqing area[J]. Coal Technology, 19: 36–50. Google Scholar
[31]	Wu D P, Yue X Y, Wu C P. 2000. Application of neural network technology in coalbed gas logging evaluation[J]. Fault Block Oil and Gas Fields, 7(5): 48−49 (in Chinese with English abstract). Google Scholar
[32]	Xu H, Tang D Z, Tao S, et al. 2024. Differences in geological conditions of deep and shallow coalbed methane and their formation mechanisms[J]. Coal Geology & Exploration, 52(2): 33−39 (in Chinese with English abstract). Google Scholar
[33]	Yang, Y, Shao B, Wang, et al. 2015. Dynamic crushing mechanism of well−wall in engraving stage of sidetracking based on Lagrange explicit algorithm[J]. Journal of Coal Society, 40(7): 1491−1497 (in Chinese with English abstract). Google Scholar
[34]	Zhang R, Liu S. 2017. Experimental and theoretical characterization of methane and CO₂ sorption hysteresis in coals based on Langmuir desorption[J]. International Journal of Coal Geology, 171: 49−60. doi: 10.1016/j.coal.2016.12.007 CrossRef Google Scholar
[35]	常闯, 李松, 汤达祯, 等. 2023. 基于测井参数的煤储层地应力计算方法研究——以延川南区块为例[J]. 煤田地质与勘探, 51(5): 23−32. Google Scholar
[36]	陈世达, 侯伟, 汤达祯, 等. 2024. 煤储层含气性深度效应与成藏过程耦合关系[J]. 煤田地质与勘探, 52(2): 52−59. Google Scholar
[37]	陈涛, 张占松, 周雪晴, 等. 2021. 基于测井参数优选的煤层含气量预测模型[J]. 煤田地质与勘探, 49(3): 227−235. doi: 10.3969/j.issn.1001-1986.2021.03.029 CrossRef Google Scholar
[38]	冯鹏, 李松, 汤达祯, 等. 2022. 支持向量机在煤层地应力预测中的应用[J]. 现代地质, 36(5): 1333−1340. Google Scholar
[39]	李丹丹, 王振国, 降文萍, 等. 2022. 基于测井参数的煤层含气量定量评价方法研究——以寿阳区块15煤为例[J]. 中国煤炭地质, 34(8): 34−40. doi: 10.3969/j.issn.1674-1803.2022.08.06 CrossRef Google Scholar
[40]	林魂, 孙新毅, 宋西翔, 等. 2023. 基于改进人工神经网络的页岩气井产量预测模型研究[J]. 油气藏评价与开发, 13(4): 467−473. Google Scholar
[41]	李琴, 刘永升, 黄志强, 等. 2016. PDC钻头切削破岩机理及数值模拟研究[J]. 机械科学与技术, 35(2): 203−209. Google Scholar
[42]	刘晓. 2024. 不同压裂规模下煤储层缝网形态对比研究——以延川南煤层气田为例[J]. 油气藏评价与开发, 14(3): 510−518. Google Scholar
[43]	李泽辰, 杜文凤, 胡进奎, 等. 2018. 鄂尔多斯盆地临兴区块测井含气量解释方法[J]. 煤炭学报, 43(S2): 490−498. Google Scholar
[44]	孟召平, 田永东, 李国富. 2020. 煤层气开发地质学理论与方法[M]. 北京: 科学出版社. Google Scholar
[45]	孟召平, 田永东, 雷旸. 2008. 煤层含气量预测的BP神经网络模型与应用[J]. 中国矿业大学学报, (4): 456−461. doi: 10.3321/j.issn:1000-1964.2008.04.005 CrossRef Google Scholar
[46]	秦瑞宝, 叶建平, 李利, 等. 2023. 基于机器学习的煤层含气量测井评价方法——以沁水盆地柿庄南区块为例[J]. 石油物探, 62(1): 68−79. doi: 10.3969/j.issn.1000-1441.2023.01.005 CrossRef Google Scholar
[47]	谌丽, 王才志, 宁从前, 等. 2023. 基于机器学习的鄂尔多斯盆地陇东地区长7段岩相测井识别方法[J]. 油气藏评价与开发, 13(4): 525−536. Google Scholar
[48]	王欣, 韩建强, 昝灵, 等. 2024. 苏北盆地溱潼凹陷阜宁组二段页岩油测井评价研究[J]. 油气藏评价与开发, 14(3): 364−372. Google Scholar
[49]	吴东平, 岳晓燕, 吴春萍. 2000. 神经网络技术在煤层气测井评价中的应用[J]. 断块油气田, 7(5): 48−49. Google Scholar
[50]	许浩, 汤达祯, 陶树, 等. 2024. 深、浅部煤层气地质条件差异性及其形成机制[J]. 煤田地质与勘探, 52(2): 33−39. doi: 10.12363/issn.1001-1986.23.10.0693 CrossRef Google Scholar
[51]	杨勇, 邵兵, 王风锐, 等. 2015. 基于Lagrange显式算法的井壁侧钻过程动态破碎规律[J]. 煤炭学报, 40(7): 1491−1497. Google Scholar