迁移深度神经网络的页岩总孔隙度预测

doi:10.11885/j.issn.1674-5086.2021.06.11.03

西南石油大学学报(自然科学版) ›› 2023, Vol. 45 ›› Issue (6): 69-79.DOI: 10.11885/j.issn.1674-5086.2021.06.11.03

迁移深度神经网络的页岩总孔隙度预测

汪敏¹, 杨桃¹, 唐洪明^2,3, 闫建平^2,3, 廖纪佳³

1. 西南石油大学电气信息学院, 四川成都 610500;
2. 油气藏地质及开发工程全国重点实验室·西南石油大学四川成都 610500;
3. 西南石油大学地球科学与技术学院, 四川成都 610500

收稿日期:2021-06-11 发布日期:2024-01-06
通讯作者: 汪敏,E-mail:wangmin80616@163.com
作者简介:汪敏,1980年生,女,汉族,湖南邵阳人,教授,主要从事机器学习、深度学习算法理论研究及其油气应用研究。E-mail:wangmin80616@163.com;杨桃,1994年生,男,汉族,四川通江人,硕士研究生,主要从事人工智能油气地质应用研究。E-mail:1028602501@qq.com;唐洪明,1966年生,男,汉族,四川武胜人,教授,博士研究生导师,主要从事油气田开发地质学基础理论及应用技术研究和非常规油气储层评价研究。E-mail:swpithm@vip.163.com;闫建平,1980年生,男,汉族,内蒙古凉城人,教授,博士研究生导师,主要从事测井地质学、岩石物理、非常规储层测井评价技术及测井软件开发与应用的研究。E-mail:yanjp_tj@163.com;廖纪佳,1983年生,男,汉族,四川绵竹人,副教授,博士,主要从事沉积学、储层地质学的教学和研究。E-mail:liaojijia198433@163.com
基金资助:
国家自然科学基金（62006200）；中国石油-西南石油大学创新联合体科技合作项目（2020CX020000）

Prediction for Total Porosity of Shale Based on Transfer Deep Neural Network

WANG Min¹, YANG Tao¹, TANG Hongming^2,3, YAN Jianping^2,3, LIAO Jijia³

1. School of Electrical Engineering and Information, Southwest Petroleum University, Chengdu, Sichuan 610500, China;
2. National Key Laboratory of Oil & Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, Sichuan 610500, China;
3. School of Geoscience and Technology, Southwest Petroleum University, Chengdu, Sichuan 610500, China

Received:2021-06-11 Published:2024-01-06

摘要/Abstract

摘要： 页岩孔隙度是表征页岩储层孔隙结构的关键指数之一，对孔隙度进行准确预测是开展储层评价的重要一步。页岩孔隙度的准确确定需要以取芯井数据为依据，如何基于极少量取芯井数据，准确预测整口井的孔隙度，是一个有意义的问题。提出了全新迁移深度神经网络模型，基于少量测井和岩芯数据，实现孔隙度的准确预测。首先，根据皮尔逊相关系数法，优选源井深度神经网络的输入测井参数；然后，提出一种新的方法，计算源井与目标井测井数据分布相似性，定量衡量两井间的地质差异；最后，用少量的与源井测井数据分布相似的目标井测井数据再训练源井预测网络，构建预测目标井孔隙度迁移深度神经网络。在A2、B2两口井的测试结果表明： 1）该方法只需要10%的数据量，就到达了绝对均值误差为0.032 9和决定系数为0.841 6的预测性能。2）提出的计算两井相似性的方法可以有效衡量井间差异。源井测井数据与目标井测井数据分布越相似，迁移学习网络的孔隙度预测精度越高。所提出的模型能有效减小对测井和岩芯数据的依赖，极大降低页岩气勘探开发成本。

关键词: 页岩气, 测井, 页岩孔隙度预测, 井间差异, 深度迁移学习

Abstract: Porosity is one of the key indicators to characterize the por structure of shale reservoirs. Quantitative prediction research on porosity is an important step in reservoir evaluation. The accurate value of shale porosity must be obtained through core analysis. How to obtain an accurate prediction of the porosity of the entire well based on a very small amount of coring well data is a significant problem. This paper proposes a new transfer deep neural network model, based on a small amount of core and logging data, to achieve accurate prediction of porosity. Firstly, according to Pearson correlation coefficient method, the logging parameters suitable for the source well deep neural network are selected as the input for the model. Secondly, a new method is proposed to calculate the similarity of the well logging data distribution between the source well and the target well, quantitatively measure the geological difference between two wells. Thridly, retrain the source well prediction network with a small amount of target well logging data similar to the source well logging data distribution, and build a transfer deep neural network for predicting the porosity migration of the target well. The test results of A2 and B2 show that: 1) This method requires only 10% of the data volume, and reaches the performance of an absolute mean error of 0.032 9 and a coefficient of determination of 0.841 6; 2) The proposed method for calculating the similarity of two wells can effectively measure the difference between wells. The more similar the distribution of the source well logging data and the target well logging data, the higher the accuracy of porosity prediction of the transfer learning network. The proposed model can effectively reduce the dependence on logging and core data, and greatly reduce shale gas exploration and development costs.

Key words: shale gas, well logging, shale porosity prediction, difference between wells, deep transfer learning

中图分类号:

TE122

汪敏, 杨桃, 唐洪明, 闫建平, 廖纪佳. 迁移深度神经网络的页岩总孔隙度预测[J]. 西南石油大学学报(自然科学版), 2023, 45(6): 69-79.

WANG Min, YANG Tao, TANG Hongming, YAN Jianping, LIAO Jijia. Prediction for Total Porosity of Shale Based on Transfer Deep Neural Network[J]. Journal of Southwest Petroleum University(Science & Technology Edition), 2023, 45(6): 69-79.

0
/ / 推荐

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://journal15.magtechjournal.com/Jwk_xnzk/CN/10.11885/j.issn.1674-5086.2021.06.11.03

http://journal15.magtechjournal.com/Jwk_xnzk/CN/Y2023/V45/I6/69

参考文献

[1] 李成博，宁传奇，钟长林，等. 中国油页岩矿勘查控制程度探讨[J]. 吉林大学学报(地球科学版)， 2021， 51(1):13-21. doi:10.13278/j.cnki.jjuese.20200072 LI Chengbo, NING Chuanqi, ZHONG Changlin, et al. Discussion on extent of exploration control of oil shale deposits in China[J]. Journal of Jilin University (Earth Science Edition), 2021, 51(1): 13-21. doi: 10.13278/j.cnki.jjuese.20200072 [2] 刘树根，王世玉，孙玮，等. 四川盆地及其周缘五峰组-龙马溪组黑色页岩特征[J]. 成都理工大学学报(自然科学版)， 2013， 40(6):621-639. doi:10.3969/j.issn.1671-9727.2013.06.02 LIU Shugen, WANG Shiyu, SUN Wei, et al. Characteristics of black shale in Wufeng Formation and Longmaxi Formation in Sichuan Basin and its peripheral areas[J]. Jonurnal of Chengdu University of Technology (Science & Tenchnology Edition), 2013, 40(6): 621-639. doi: 10.3969/j.issn.1671-9727.2013.06.02 [3] 徐传正，冯烁，田继军，等. 龙马溪组岩相类型及其对孔隙特征的影响因素[J]. 西南石油大学学报(自然科学版)，2021，43(1):51-60. doi:10.11885/j.issn.1674-5086.2019.05.04.01 XU Chuanzheng, FENG Shuo, TIAN Jijun, et al. Lithofacies types of Longmaxi Formation and its influencing factors on pore characteristics[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2021, 43(1): 51-60. doi: 10.11885/j.issn.1674-5086.2019.05.04.01 [4] JARVIE D M, HILL R J, RUBLE T E. Unconventional shale-gas systems: The missisippian barnett shale of northcentral texas as one mode for thermongenic shale-gas assessmet[J]. AAPG Bulletin, 2007, 91(4): 475-499. [5] ROSS D J K, BUSTIN R M. The importance of shale composition and pore structure upon gas storage potential of shale gas reservoirs[J]. Marine and Petroleum Geology, 2009, 26(6): 916-927. doi: 10.1016/j.marpetgeo.2008.06.004 [6] 赵越，李磊，司运航，等. 浅层页岩气储层孔隙分形特征及控制因素:以云南昭通地区龙马溪组为例[J]. 吉林大学学报(地球科学版)， 2022， 52(6):1813-1829. doi:10.13278/j.cnki.jjuese.20220148 ZHAO Yue, LI Lei, SI Yunhang, et al. Fractal characteristics and controlling factors of pores in shallow shale gas reservoirs: A case study of Longmaxi Formation in Zhaotong Area, Yunnan Province[J]. Journal of Jilin University (Earth Science Edition), 2022, 52(6): 1813-1829. doi: 10.13278/j.cnki.jjuese.20220148 [7] LUFFEL D L, GUIDRY F K. New core analysis methods for measuring reservoir rock properties of devonian shale[J]. Journal of Petroleum Technology, 1989, 44(11): 1184-1190. doi: 10.2118/20571-PA [8] 吴逸豪，卢双舫，陈方文，等. 泥页岩储层有机孔隙定量评价研究[J]. 特种油气藏， 2015， 22(5):65-68. doi:10.3969/j.issn.1006-6535.2015.05.013 WU Yihao, LU Shuangfang, CHEN Fangwen, et al. Quantitative characterization of organic pores in muddy shale reservoirs[J]. Special Oil and Gas Reservoirs, 2015, 22(5): 65-68. doi: 10.3969/j.issn.1006-6535.2015.05.013 [9] 钟光海，陈丽清，廖茂杰，等. 页岩气储层品质测井综合评价[J]. 天然气工业， 2020， 40(2):54-60. doi:10.3787/j.issn.1000-0976.2020.02.006 ZHONG Guanghai, CHEN Liqing, LIAO Maojie, et al. A comprehensive logging evaluation method of shale gas reservoir quality[J]. Natural Gas Industry, 2020, 40(2): 54-60. doi: 10.3787/j.issn.1000-0976.2020.02.006 [10] 王洪亮，穆龙新，时付更，等. 基于循环神经网络的油田特高含水期产量预测方法[J]. 石油勘探与开发， 2020， 47(5):1009-1015. doi:10.11698/PED.2020.05.15 WANG Hongliang, MU Longxin, SHI Fugeng, et al. Production prediction at ultra-high water cut stage via recurrent neural network[J]. Petroleum Exploration and Development, 2020, 47(5): 1009-1015. doi: 10.11698/PED.2020.05.15 [11] GU Ming, GOKARAJU D, CHEN Dingding, et al. Shale fracturing characterization and optimization by using anisotropic acoustic interpretation, 3D fracture modeling, and supervised machine learning[J]. Petrophysics: The SPWLA Journal of Formation Evaluation and Reservoir Description, 2016, 57(6): 573-587. [12] 惠钢，陈胜男，王海，等. 基于改进残差的神经网络方法预测页岩气甜点[J]. 西南石油大学学报(自然科学版)， 2021， 43(5):19-32. doi:10.11885/j.issn.1674-5086.2021.02.27.01 HUI Gang, CHEN Shengnan, WANG Hai, et al. Application of improved residual neural network-based machine learning method in the prediction of shale gas sweet spot[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2021, 43(5): 19-32. doi: 10.11885/j.issn.1674-5086.2021.02.27.01 [13] SILVA A A, LIMA N I A, MISSAGIA R M, et al. Artificial neural networks to support petrographic classification of carbonate-siliciclastic rocks using well logs and textural information[J]. Journal of Appliced Geophysics, 2015, 117: 118-125. [14] 杨柳青，陈伟，查蓓. 利用卷积神经网络对储层孔隙度的预测研究与应用[J]. 地球物理学进展， 2019， 34(4):1548-1555. doi:10.6038/pg2019CC0528 YANG Liuqing, CHEN Wei, ZHA Bei. Prediction and application of reservoir porosity by convolutional neural network[J]. Progress in Geophysics, 2019, 34(4): 1548-1555. doi: 10.6038/pg2019CC0528 [15] 聂舟，马诗杰，伍秋姿，等. 长宁地区海相页岩天然裂缝发育特征及其对含气性的影响[J]. 断块油气田， 2022， 29(5):591-597. NIE Zhou, MA Shijie, WU Qiuzi, et al. Development characteristics of natural fractures in marine shale in Changning area and their influence on gas-bearing properties[J]. Fault-Block Oil and Gas Field, 2022, 29(5): 591-597. [16] 胡月，陈雷，周昊，等. 海相页岩纹层特征及其对页岩储层发育的影响:以川南长宁地区龙马溪组为例[J]. 断块油气田， 2021， 28(2):145-150. doi:10.6056/dkyqt202102001 HU Yue, CHEN Lei, ZHOU Hao, et al. Lamina characteristics of marine shale and its influence on shale reservoir development: A case study of Longmaxi Formation, Changning Area, South Sichuan Basin[J]. FaultBlock Oil and Gas Field, 2021, 28(2): 145-150. doi: 10.6056/dkyqt202102001 [17] 范高锋，王伟胜，刘纯，等. 基于人工神经网络的风电功率预测[J]. 中国电机工程学报，2008，28(34):118-123. doi:10.3321/j.issn:0258-8013.2008.34.019 FAN Gaofeng, WANG Weisheng, LIU Chun, et al. Wind power prediction based on artificial neural network[J]. Proceedings of the CSEE, 2008, 28(34): 118-123. doi: 10.3321/j.issn:0258-8013.2008.34.019 [18] PAN S J, TSANG I W, KWOK J T, et al. Domain adaptation via transfer component analysis[J]. IEEE Transactions on Neural Networks, 2011, 22(2): 199-210. [19] 周飞燕，金林鹏，董军. 卷积神经网络研究综述[J]. 计算机学报， 2017， 40(6):1229-1251. doi:10.11897/SP.J.1016.2017.01229 ZHOU Feiyan, JIN Linpeng, DONG Jun. Review of convolution neural network[J]. Chinese Journal of Computers, 2017, 40(6): 1229-1251. doi: 10.11897/SP.J.1016.2017.01229 [20] 刘建伟，刘媛，罗雄麟. 深度学习研究进展[J]. 计算机应用研究， 2014， 31(7):1921-1930， 1942. doi:10.3969/j.issn.1001-3695.2014.07.001 LIU Jianwei, LIU Yuan, LUO Xionglin. Research and development on deep learning[J]. Application Research of Computers, 2014, 31(7): 1921-1930, 1942. doi: 10.3969/j.issn.1001-3695.2014.07.001 [21] 熊中敏，郭怀宇，吴月欣. 缺失数据处理方法研究综述[J]. 计算机工程应用， 2021， 57(14):27-38. doi:10.3778/j.issn.1002-8331.2101-0187 XIONG Zhongmin, GUO Huaiyu, WU Yuexin. Review of missing data processing methods[J]. Computer Engineering and Applications, 2021, 57(14): 27-38. doi: 10.3778/j.issn.1002-8331.2101-0187 [22] 陈娟，王献雨，罗玲玲，等. 缺失值填补效果:机器学习与统计学习的比较[J]. 统计与决策， 2020， 36(17):28-32. doi:10.13546/j.cnki.tjyjc.2020.17.006 CHEN Juan, WANG Xianyu, LUO Lingling, et al. Comparison of machine learning and statistical learning in the imputation of missing values[J]. Statistics and Decision, 2020, 36(17): 28-32. doi: 10.13546/j.cnki.tjyjc.2020.17.006 [23] 魏杰，杨斌，刘锋，等. 基于岩性识别的BP神经网络孔隙度预测[J]. 石油化工应用， 2020， 39(3):105-110. doi:10.3969/j.issn.1673-5285.2020.03.023 WEI Jie, YANG Bin, LIU Feng, et al. Porosity prediction of BP neural network based on lithology identification[J]. Petrochemical Industry Application, 2020, 39(3): 105-110. doi: 10.3969/j.issn.1673-5285.2020.03.023 [24] HAMADA G M, ELSHAFEI M A. Neural network prediction of porosity and permeability of heterogeneous gas sand reservoirs[C]. SPE 126042-MS, 2009. doi: 10.2118/126042-MS [25] KANFAR R, SHAIKH O, YOUSEFZADEN M, et al. Real-time well log prediction from drilling data using deep learning[C]. IPTC 19693-MS, 2020. doi: 10.2523/IPTC-19693-MS [26] 张勇，陶一凡，巩敦卫. 迁移学习引导的变源域长短时记忆网络建筑负荷预测[J]. 控制与决策，2021，36(10):2328-2338. doi:10.13195/j.kzyjc.2020.0215 ZHANG Yong, TAO Yifan, GONG Dunwei. Load forecasting of buildings using LSTM based on transfer learning with variable source domain[J]. Control and Decision, 2021, 36(10): 2328-2338. doi: 10.13195/j.kzyjc.2020.0215

迁移深度神经网络的页岩总孔隙度预测

Prediction for Total Porosity of Shale Based on Transfer Deep Neural Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	杨兆中, 杜慧龙, 易良平, 李小刚, 苟良杰. 页岩气井液岩相互作用机理与焖井制度研究进展[J]. 西南石油大学学报(自然科学版), 2023, 45(6): 80-94.
[2]	何磊, 牟必鑫, 杨平, 陈杨, 刘治成. 康滇隆起东缘上三叠统页岩气有利区优选[J]. 西南石油大学学报(自然科学版), 2023, 45(5): 43-56.
[3]	祝效华, 李瑞, 刘伟吉, 李枝林, 陆灯云. 深层页岩气水平井高效破岩提速技术发展现状[J]. 西南石油大学学报(自然科学版), 2023, 45(4): 1-18.
[4]	冯林杰, 蒋裕强, 曹脊翔, 杨长城, 宋林珂. 川中北部须家河组烃源岩测井解释及评价[J]. 西南石油大学学报(自然科学版), 2023, 45(4): 31-42.
[5]	王伟俊, 刘红岐, 朱光亚, 辛炯垄, 王友净. 碳酸盐岩油藏储层测井挖掘效应的综合分析[J]. 西南石油大学学报(自然科学版), 2023, 45(1): 43-54.
[6]	钟显康, 李浩男, 扈俊颖. 页岩气开采与集输过程中腐蚀问题与现状分析[J]. 西南石油大学学报(自然科学版), 2022, 44(6): 162-174.
[7]	刘成川, 陈俊, 程洪亮. 川西致密砂岩气藏测井评价技术及其发展展望[J]. 西南石油大学学报(自然科学版), 2022, 44(3): 59-69.
[8]	覃建华, 杨琨, 丁艺, 张博宁, 唐慧莹. 基于KL-E的地质力学模型参数反演及应用[J]. 西南石油大学学报(自然科学版), 2022, 44(2): 65-78.
[9]	白杨, 李道雄, 李文哲, 李宏波, 罗平亚. 长宁区块龙马溪组水平段井壁稳定钻井液技术[J]. 西南石油大学学报(自然科学版), 2022, 44(2): 79-88.
[10]	郭敬民, 孙恩慧, 汪巍, 李博, 杨东东. 深度挖掘水平井资料预测地层倾角新方法[J]. 西南石油大学学报(自然科学版), 2022, 44(1): 66-78.
[11]	梁兴, 单长安, 蒋佩, 张朝, 朱斗星. 浅层页岩气井全生命周期地质工程一体化应用[J]. 西南石油大学学报(自然科学版), 2021, 43(5): 1-18.
[12]	惠钢, 陈胜男, 王海, 顾斐. 基于改进残差的神经网络方法预测页岩气甜点[J]. 西南石油大学学报(自然科学版), 2021, 43(5): 19-32.
[13]	李俊乾, 卢双舫, 李文镖, 蔡建超. 基于过程分析法的页岩原位含气性评价[J]. 西南石油大学学报(自然科学版), 2021, 43(5): 42-55.
[14]	盛广龙, 黄罗义, 赵辉, 饶翔, 马嘉令. 页岩气藏压裂缝网扩展流动一体化模拟技术[J]. 西南石油大学学报(自然科学版), 2021, 43(5): 84-96.
[15]	陈学忠, 鲁友常, 吴易一, 马丽丽. 中美页岩气开采压裂返排液处置政策分析及启示[J]. 西南石油大学学报(自然科学版), 2021, 43(5): 212-219.