基于支持向量机模型的烃源岩有机碳含量预测——以鄂尔多斯盆地为例

doi:10.11764/j.issn.1672-1926.2019.01.014

天然气地球科学

基于支持向量机模型的烃源岩有机碳含量预测——以鄂尔多斯盆地为例

张成龙¹,陶士振¹,白斌¹，王倩茹²

1.中国石油勘探开发研究院石油地质研究所，北京 100083；
2.北京大学地球与空间科学学院，北京 100089

收稿日期:2018-10-29 修回日期:2019-01-22 出版日期:2019-05-10
作者简介:张成龙（1994-），男，湖北宜昌人，硕士研究生，主要从事鄂尔多斯盆地致密油富集相关研究.E-mail：Zhang-chll@163.com.
基金资助:
国家科技重大专项课题“致密油形成条件、富集规律与资源潜力”（编号：2016ZX05046-001）资助.

Source rock TOC content prediction based on the support vector machine model:An application in Ordos Basin

Zhang Cheng-long¹,Tao Shi-zhen¹,Bai Bin¹,Wang Qian-ru²

1.Institute of Petroleum Geology,PetroChina Research Institute of Petroleum Exploration and Development,Beijing 100083,China;
2.School of earth and Space Sciences,Peking University,Beijing 100089,China

Received:2018-10-29 Revised:2019-01-22 Online:2019-05-10

摘要/Abstract

摘要： 根据测井数据预测烃源岩有机质丰度为烃源岩评价提供了相对容易和廉价的替代方案。目前被广泛运用于预测有机质丰度的方法是ΔLogR法。为了适应复杂地质条件下的TOC预测，选取区域沉积背景下的一口密集采样模型井，使用线性滤波预处理数据，统一测井数据和TOC数据的精度；计算测井响应的皮尔森矩阵筛选测井训练特征；利用测井训练特征和TOC实测值作为输入建立支持向量机模型；使用交叉验证法选取地区最优模型以增强泛化能力。该新方法在鄂尔多斯盆地盐池地区的应用显示，相比ΔLogR预测方法，对TOC低值和高值的预测误差更小，能够有效反映延长组长7油层组源岩有机质丰度的纵向变化。

关键词: 支持向量机, TOC计算, ΔLogR法, 鄂尔多斯盆地

Abstract: Calculating TOC content using Logging data is a relatively easy and inexpensive way to evaluate source rocks.The traditional ΔLogR method for calculation is widely used now.In order to adapt to the TOC calculation under complex geological conditions,a representative model well is selected under the sedimentary background.For the purpose of uniting logging data and TOC data,linear filter is used to preprocess the data.The Pearson matrix can help screening logging characteristics for training.Final feature and measured TOC values are used as inputs to build the support vector machine models,and cross-validation is used to select regional optimal models for the final prediction.The application of this support vector machine method in the Ordos Basin shows that it has high accuracy and generalization ability,and is superior to the ΔLogR method when TOC is relatively high or low.The result can effectively reflect the heterogeneity of organic matter abundance in the Chang 7 high quality source rock in Ordos Basin.

Key words: Support vector machine, TOC , calculate, ΔLogR method, Ordos Basin

中图分类号:

TE122.1

张成龙, 陶士振, 白斌, 王倩茹. 基于支持向量机模型的烃源岩有机碳含量预测——以鄂尔多斯盆地为例[J]. 天然气地球科学, 2019, 30(5): 761–768.

Zhang Cheng-long, Tao Shi-zhen, Bai Bin, Wang Qian-ru. Source rock TOC content prediction based on the support vector machine model:An application in Ordos Basin[J]. Natural Gas Geoscience, 2019, 30(5): 761–768.

参考文献

［1］Schmoker J W.Determination of organic-matter content of appalachian devonian shales from gamma-ray logs[J].AAPG Bulletin,1981,65(7):1285-1298.
[2]Passey Q R,Moretti F J,Kulla J B,et al.Practical model for organic richness from porosity and resistivity logs[J].AAPG Bulletin,1990,74(12):1777-1794.
[3]Liu Chao,Lu Shuangfang,Xue Haitao.Variable-coefficient ΔLogR method and its application in shale organic evaluation[J].Progress in Geophysics,2014,29(1):312-317.
刘超,卢双舫,薛海涛.变系数ΔLogR方法及其在泥页岩有机质评价中的应用[J].地球物理学进展,2014,29(1):312-317.
[4]Huo Qiuli,Zeng Huasen,Fu Li,et al.The advance of ΔLogR method and it’s application in Songliao Basin[J].Journal of Jilin University：Earth Science Edition,2011,41(2):586-591.
霍秋立,曾花森,付丽,等.ΔLogR测井源岩评价方法的改进及其在松辽盆地的应用[J].吉林大学学报：地球科学版,2011,41(2):586-591.
[5]He Cong,Su Ao,Zhang Mingzhen,et al.Optimal selection and application of prediction means for organic carbon content of source rocks based on logging data in Yanchang Formation,Ordos Basin[J].Natural Gas Geoscience,2016,27(4):754-764.
贺聪,苏奥,张明震,等.鄂尔多斯盆地延长组烃源岩有机碳含量测井预测方法优选及应用[J].天然气地球科学,2016,27(4):754-764.
[6]Wang Qinghui,Feng Jin.The TOC logging evaluation methods and application of source rock:A case study of Wenchang Formation in Pearl River Mouth Basin[J].Natural Gas Geoscience,2018,29(2):251-258.
王清辉,冯进.烃源岩TOC测井评价方法及应用——以珠江口盆地文昌组为例[J].天然气地球科学,2018,29(2):251-258.
[7]Zhu Zhenyu,Wang Guiwen,Zhu Guangyu.The application of artificial neural network to the source rock’s evaluation[J].Progress in Geophysics,2002,17(1):137-140.
朱振宇,王贵文,朱广宇.人工神经网络法在烃源岩测井评价中的应用[J].地球物理学进展,2002,17(1):137-140.
[8]Xiong Lei,Zhang Chaomo,Zhang Chong，et al.Research on logging evaluation method of TOC content of shale gas reservoir in A area[J].Lithologic Reservoirs,2014,26(3):74-78,83.
熊镭,张超谟,张冲,等.A地区页岩气储层总有机碳含量测井评价方法研究[J].岩性油气藏,2014,26(3):74-78,83.
[9]Zhang Han,Lu Shuangfang,Li Wenhao，et al.Application of ΔLogR technology and BP neural network in organic evaluation in the complex lithology tight stratum[J].Progress in Geophysics,2017,32(3):1308-1313.
张晗,卢双舫,李文浩,等.ΔLogR技术与BP神经网络在复杂岩性致密层有机质评价中的应用[J].地球物理学进展,2017,32(3):1308-1313.
[10]Deng Xiuqin,Fu Jinhua,Yao Jingli，et al.Sedimentary facies of the Middle-Upper Triassic Yanchang Formation in Ordos Basin and breakthrough in petroleum exploration[J].Journal of Palaeogeography,2011,13(4):443-455.
邓秀芹,付金华,姚泾利,等.鄂尔多斯盆地中及上三叠统延长组沉积相与油气勘探的突破[J].古地理学报,2011,13(4):443-455.
[11]Yuan Xuanjun,Lin Senhu,Liu Qun，et al.Lacustrine fine-grained sedimentary features and organic-rich shale distribution pattern:A case study of Chang 7 member of Triassic Yanchang Formation in Ordos Basin,NW China[J].Petroleum Exploration and Development,2015,42(1):34-43.
袁选俊,林森虎,刘群,等.湖盆细粒沉积特征与富有机质页岩分布模式——以鄂尔多斯盆地延长组长7油层组为例[J].石油勘探与开发,2015,42(1):34-43.
[12]Zhang Wenzheng,Yang Hua,Li Jianfeng，et al.Leading effect of high-class source rock of Chang 7 in Ordos Basin on enrichment of low permeability oil-gas accumulation：Hydrocarbon generation and expulsion mechanism[J].Petroleum Exploration and Development,2006,33(3):289-293.
张文正,杨华,李剑锋,等.论鄂尔多斯盆地长7段优质油源岩在低渗透油气成藏富集中的主导作用——强生排烃特征及机理分析[J].石油勘探与开发,2006,33(3):289-293.
[13]Yang Hua,Zhang Wenzheng.Leading effect of the seventh member of high-quality source rock of Yanchang Formation Ordos Basin during the enrichment of low-penetrating oil-gas accumalation:Geology and Geochemistry[J].Geochimica,2005,34(2):147-154.
杨华,张文正.论鄂尔多斯盆地长7段优质油源岩在低渗透油气成藏富集中的主导作用:地质地球化学特征[J].地球化学,2005,34(2):147-154.
[14]Yang Zhi,Hou Lianhua,Tao Shizhen，et al.Formation conditions and “sweet spot
” evaluation of tight oil and shale oil[J].Petroleum Exploration and Development,2015,42(5):555-565.
杨智,侯连华,陶士振,等.致密油与页岩油形成条件与“甜点区”评价[J].石油勘探与开发,2015,42(5):555-565.
[15]Yang Bin,Kuang Lichun,Sun Zhongchun，et al.On support vector machines method to identify oil & gas zone with logging and mudlog information[J].Well Logging Technology,2005,29(6):511-514,571.
杨斌,匡立春,孙中春,等.一种用于测井油气层综合识别的支持向量机方法[J].测井技术,2005,29(6):511-514,571.
[16]Li Xinhu.Lithology identification methods contrast based on support vector machines at different well logging parameter[J].Coal Geology & Exploration,2007,35(3):72-76,80.
李新虎.基于不同测井曲线参数集的支持向量机岩性识别对比[J].煤田地质与勘探,2007,35(3):72-76,80.
[17]Mou Dan,Wang Zhuwen,Huang Yulong，et al.Lithological identification of volcanic rocks from SVM well logging data:A case study in the eastern depression of Liaohe Basin[J].Chinese Journal of Geophysics,2015,58(5):1785-1793.
牟丹,王祝文,黄玉龙,等.基于SVM测井数据的火山岩岩性识别——以辽河盆地东部坳陷为例[J].地球物理学报,2015,58(5):1785-1793.
[18]Burges Christopher J C.A tutorial on support vector machines for pattern recognition[J].Data Mining and Knowledge Discovery,1998,22(2):121-167.
[19]Cui Shasha.Evaluation of Hydrocarbon Source Rock by Logging Method:A Case Study from the Nanniwan Oilfiled[D].Xi’an:Xi’an Shiyou University,2016:21-23.
崔莎莎.烃源岩测井评价方法——以南泥湾油田为例[D].西安：西安石油大学,2016：21-23.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed