基于XGBoost算法的复杂碳酸盐岩岩性测井识别

doi:10.12108/yxyqc.20200410

岩性油气藏 ›› 2020, Vol. 32 ›› Issue (4): 98–106.doi: 10.12108/yxyqc.20200410

基于XGBoost算法的复杂碳酸盐岩岩性测井识别

孙予舒^1,2, 黄芸³, 梁婷^1,2, 季汉成^1,2, 向鹏飞^1,2, 徐新蓉^1,2

1. 中国石油大学(北京)地球科学学院, 北京 102249;
2. 中国石油大学(北京)油气资源与探测国家重点实验室, 北京 102249;
3. 中国石油华北油田分公司勘探开发研究院, 河北任丘 062550

收稿日期:2019-09-12 修回日期:2019-11-29 出版日期:2020-08-01 发布日期:2020-06-16
第一作者:孙予舒(1994-),男,中国石油大学(北京)在读硕士研究生,研究方向为机器学习、沉积学及储层地质学。地址:(102249)北京市昌平区府学路18号。Email:sunyushu1022@163.com
通信作者: 季汉成(1966-),男,博士,教授,主要从事沉积学、储层地质学、石油地质学方面的教学和研究工作。Email:jihancheng@vip.sina.com。
基金资助:
中国石油天然气股份有限公司重大科技专项“华北油田持续有效稳产勘探开发关键技术研究与应用”（编号：2017E-15）和“冀中凹陷下古生界潜山及内幕优势储层成因、演化及分布特征研究”（编号：HBYT-YJY-2018-JS-274）联合资助

Identification of complex carbonate lithology by logging based on XGBoost algorithm

SUN Yushu^1,2, HUANG Yun³, LIANG Ting^1,2, JI Hancheng^1,2, XIANG Pengfei^1,2, XU Xinrong^1,2

1. College of Geoscience, China University of Petroleum(Beijing), Beijing 102249, China;
2. State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum(Beijing), Beijing 102249, China;
3. Research Institute of Exploration & Development, PetroChina Huabei Oilfield Company, Renqiu 062550, Hebei, China

Received:2019-09-12 Revised:2019-11-29 Online:2020-08-01 Published:2020-06-16

摘要/Abstract

摘要： 碳酸盐岩储层在形成过程中受到多种因素的影响，储层岩性复杂多样，基于测井资料对碳酸盐岩岩性识别具有重要意义。为了解决传统的测井岩性识别方法和机器学习方法对于复杂碳酸盐岩岩性识别准确率不高的问题，以廊固凹陷北部奥陶系碳酸盐岩为例，将XGBoost算法应用于复杂碳酸盐岩岩性识别，并将模型的性能与决策树C4.5算法和支持向量机算法进行对比。结果表明，采用的XGBoost算法的岩性识别模型对研究区碳酸盐岩岩性识别的准确率达到了88.18%，相较于决策树C4.5算法和支持向量机算法准确率均提高了10%左右，且由于XGBoost算法采用多线程和分布式计算的方法，使得训练时间大大缩短。基于XGBoost算法建立的岩性识别模型能够有效地识别复杂碳酸盐岩岩性，为复杂碳酸盐岩岩性的测井识别提供了新的思路。

关键词: XGBoost算法, 机器学习, 碳酸盐岩, 岩性识别, 测井解释

Abstract: Carbonate reservoirs are affected by a variety of factors during the formation process,and the reservoir lithology is complex and diverse. Logging data are of great significance for carbonate lithology identification. In order to solve the problem that the traditional logging lithology identification method and traditional machine learning have low recognition accuracy for complex carbonate lithology,taking the Ordovician carbonate rocks in the northern Langgu Depression as an example, based on log data, XGBoost algorithm was appied to lithology identification of complex carbonate rocks,and the performance of the model was comparied with the decision tree C4.5 algorithm and the support vector machine algorithm. The results show that the lithology identification model based on XGBoost algorithm has an accuracy rate of 88.18% for the identification of carbonate lithology in the study area. Compared with decision tree C4.5 and support vector machine,the accuracy rate is increased by about 10%. And the XGBoost algorithm uses multi-threaded and distributed computing methods,the training time is greatly shortened. It shows that the lithology identification model established by XGBoost algorithm can effectively identify complex carbonate lithology and provide a new idea for logging identification of complex carbonate lithology.

Key words: XGBoost algorithm, machine learning, carbonate, lithology identification, log interpretation

中图分类号:

P618.13

孙予舒, 黄芸, 梁婷, 季汉成, 向鹏飞, 徐新蓉. 基于XGBoost算法的复杂碳酸盐岩岩性测井识别[J]. 岩性油气藏, 2020, 32(4): 98–106.

SUN Yushu, HUANG Yun, LIANG Ting, JI Hancheng, XIANG Pengfei, XU Xinrong. Identification of complex carbonate lithology by logging based on XGBoost algorithm[J]. Lithologic Reservoirs, 2020, 32(4): 98–106.

参考文献

[1] 江凯, 王守东, 胡永静, 等. 基于Boosting Tree算法的测井岩性识别模型.测井技术, 2018, 42(4):396. JIANG K, WANG S D, HU Y J, et al. Lithology identification model by well logging based on boosting tree algorithm. Well Logging Technology, 2018, 42(4):396.
[2] 王瑞, 朱筱敏, 王礼常.用数据挖掘方法识别碳酸盐岩岩性. 测井技术, 2012, 36(2):197. WANG R, ZHU X M, WANG L C. Using data mining to identify carbonate lithology. Well Logging Technology, 2012, 36(2):197.
[3] 吴施楷, 曹俊兴.基于连续限制玻尔兹曼机的支持向量机岩性识别方法.地球物理学进展, 2016, 31(2):821-828. WU S K, CAO J X. Lithology identification method based on continuous restricted Boltzmann machine and support vector machine. Progress in Geophysics, 2016, 31(2):821-828.
[4] 杨冬.BP神经网络技术在碳酸盐岩岩性识别中的应用.石化技术, 2016, 23(1):58. YANG D. Application of BP neural network technology in carbonate lithology identification. Petrochemical Industry Technology, 2016, 23(1):58.
[5] 张翔, 肖小玲, 严良俊, 等.基于模糊支持向量机方法的岩性识别. 石油天然气学报(江汉石油学院学报), 2009, 31(6):115-118. ZHANG X, XIAO X L, YAN L J, et al. Lithologic identification based on fuzzy support vector machines. Journal of Oil and Gas Technology(Journal of Jianghan Petroleum Institute), 2009, 31(6):115-118.
[6] 钟仪华, 李榕.基于主成分分析的最小二乘支持向量机岩性识别方法.测井技术, 2009, 33(5):425-429. ZHONG Y H, LI R. Application of principal component analysis and least square support vector machine to lithology identification. Well Logging Technology, 2009, 33(5):425-429.
[7] 赵忠军, 黄强东, 石林辉, 等.基于BP神经网络算法识别苏里格气田致密砂岩储层岩性.测井技术, 2015, 39(3):363-367. ZHAO Z J, HUANG Q D, SHI L H, et al. Identification of lithology in tight sandstone reservoir in Sulige Gas Field based on BP neural net algorithm. Well Logging Technology, 2015, 39(3):363-367.
[8] 范存辉, 梁则亮, 秦启荣, 等.基于测井参数的遗传BP神经网络识别火山岩岩性:以准噶尔盆地西北缘中拐凸起石炭系火山岩为例.石油天然气学报, 2012, 34(1):68-71. FAN C H, LIANG Z L, QIN Q R, et al. Identification of volcanic-rock lithology by using genetic BP neural network based on logging parameters:By taking carboniferous volcanic rocks in Zhongguai uplift of northwestern margin of Junggar Basin for instance. Journal of Oil and Gas Technology, 2012, 34(1):68-71.
[9] 王振洲, 张春雷, 高世臣.利用决策树方法识别复杂碳酸盐岩岩性:以苏里格气田苏东41-33区块为例.油气地质与采收率, 2017, 24(6):25-33. WANG Z Z, ZHANG C L, GAO S C. Lithology identification of complex carbonate rocks based on decision tree method:an example from block Sudong 41-33 in Sulige Gas Field. Petroleum Geology and Recovery Efficiency, 2017, 24(6):25-33.
[10] 仲鸿儒, 成育红, 林孟雄, 等.基于SOM和模糊识别的复杂碳酸盐岩岩性识别.岩性油气藏, 2019, 31(5):84-91. ZHONG H R, CHENG Y H, LIN M X, et al. Lithology identification of complex carbonate based on SOM and fuzzy recognition. Lithologic Reservoirs, 2019, 31(5):84-91.
[11] 马峥, 张春雷, 高世臣.主成分分析与模糊识别在岩性识别中的应用.岩性油气藏, 2017, 29(5):127-133. MA Z, ZHANG C L, GAO S C. Lithology identification based on principal component analysis and fuzzy recognition. Lithologic Reservoirs, 2017, 29(5):127-133.
[12] 宋延杰, 王团, 付健, 等.雷64区块砂砾岩储层岩性识别方法研究.哈尔滨商业大学学报(自然科学版), 2015, 31(1):73-78. SONG Y J, WANG T, FU J, et al. Research on technology of lithology identification of sand-conglomerate rock in Lei 64. Journal of Harbin University of Commerce(Natural Sciences Edition), 2015, 31(1):73-78.
[13] 李洪奇, 谭锋奇, 许长福, 等.基于决策树方法的砾岩油藏岩性识别:以克拉玛依油田六中区克下组油藏为例.石油天然气学报(江汉石油学院学报), 2010, 32(3):73-79. LI H Q, TAN F Q, XU C F, et al. Lithological identification of conglomerate reservoirs base on decision tree method. Journal of Oil and Gas Technology(Journal of Jianghan Petroleum Institute), 2010, 32(3):73-79.
[14] 李百强, 张小莉, 王起琮, 等.低渗-特低渗白云岩储层成岩相分析及测井识别:以伊陕斜坡马五段为例. 岩性油气藏, 2019, 31(5):70-83. LI B Q, ZHANG X L, WANG Q C, et al. Analysis and logging identification of diagenetic facies of dolomite reservoir with low and ultra-low permeability:a case study from Ma 5 memberin Yishan slope, Ordos Basin. Lithologic Reservoirs, 2019, 31(5):70-83.
[15] CHEN T Q, GUESTRIN C. XGBoost:a scalable tree boosting system. Proceedings of the 22 nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, California, USA, 2016:785-794.
[16] 刘宇, 乔木.基于聚类和XGboost算法的心脏病预测.计算机系统应用, 2019, 28(1):229. LIU Y, QIAO M. Heart disease prediction based on clustering and XGboost. Computer Systems & Applications, 2019, 28(1):229.
[17] FRIEDMAN J H. Greedy function approximation:a gradient boosting machine. The Annals of Statistics, 2001, 29(5):1189-1232.
[18] 李超, 张文辉, 林基明.基于XGBoost算法的恒星/星系分类研究.天文学报, 2019, 60(2):75. LI C, ZHANG W H, LIN J M. Research on star/galaxy classification based on XGBoost algorithm. Acta Astronomica Sinica, 2019, 60(2):75.
[19] 沈倩倩, 邵峰晶, 孙仁诚.基于XGBoost的乳腺癌预测模型. 青岛大学学报(自然科学版), 2019, 32(1):97. SHEN Q Q, SHAO F J, SUN R C. Prediction model of breast cancer based on XGBoost. Journal of Qingdao University (Natural Science Edition), 2019, 32(1):97.
[20] 罗菊兰, 陈彦竹, 高波, 等.基于矿物组合分类的碳酸盐岩储层岩性识别模型的建立.国外测井技术, 2018, 39(2):21-26. LUO J L, CHEN Y Z, GAO B, et al. Establishment of lithology recognition model for carbonate reservoir based on mineral assemblage classification. World Well Logging Technology, 2018, 39(2):21-26.
[21] 高雅琴, 谢润成, 吕志洲, 等.基于多元概率因子识别复杂碳酸盐岩岩性方法的应用.石化技术, 2018, 25(2):122-123. GAO Y Q, XIE R C, LYU Z Z, et al. The application of methods of identifying lithology of complex carbonate rocks based on multiple probability factor. Petrochemical Industry Technology, 2018, 25(2):122-123.
[22] 孙哲, 韦阿娟, 江尚昆, 等.元素录井技术在渤海潜山岩性识别中的应用.特种油气藏, 2017, 24(5):78-84. SUN Z, WEI A J, JIANG S K, et al. Application of element logging technology in identifying buried hill lithologies in Bohai Sea. Special Oil and Gas Reservoirs, 2017, 24(5):78-84.
[23] 关新, 陈世加, 苏旺, 等.四川盆地西北部栖霞组碳酸盐岩储层特征及主控因素.岩性油气藏, 2018, 30(2):67-76. GUAN X, CHEN S J, SU W, et al. Carbonate reservoir characteristics and main controlling factors of Middle Permian Qixia Formation in NW Sichuan Basin. Lithologic Reservoirs, 2018, 30(2):67-76.
[24] 刘冬冬, 杨东旭, 张子亚, 等.基于常规测井和成像测井的致密储层裂缝识别方法:以准噶尔盆地吉木萨尔凹陷芦草沟组为例.岩性油气藏, 2019, 31(3):76-85. LIU D D, YANG D X, ZHANG Z Y, et al. Fracture identification for tight reservoirs by conventional and imaging logging:a case study of Permian Lucaogou Formation in Jimsar Sag, Junggar Basin. Lithologic Reservoirs, 2019, 31(3):76-85.
[25] 杨柳, 王钰.泛化误差的各种交叉验证估计方法综述.计算机应用研究, 2015, 32(5):1288-1289. YANG L, WANG Y. Survey for various cross-validation estimators of generalization error. Application Research of Computers, 2015, 32(5):1288-1289.
[26] KHALID S, KHALIL T, NASREEN S. A survey of feature selection and feature extraction techniques in machine leaing. 2014 Science and Information Conference. London, UK, 2014:372-378.
[27] RAMASUBRAMANIAN K, SINGH A. Machine learning using R. New York:Apress, 2017:181-184.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于XGBoost算法的复杂碳酸盐岩岩性测井识别

Identification of complex carbonate lithology by logging based on XGBoost algorithm

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10

[1]	赵军, 李勇, 文晓峰, 徐文远, 焦世祥. 基于斑马算法优化支持向量回归机模型预测页岩地层压力[J]. 岩性油气藏, 2024, 36(6): 12-22.
[2]	张天择, 王红军, 张良杰, 张文起, 谢明贤, 雷明, 郭强, 张雪锐. 射线域弹性阻抗反演在阿姆河右岸碳酸盐岩气藏储层预测中的应用[J]. 岩性油气藏, 2024, 36(6): 56-65.
[3]	桂金咏, 李胜军, 高建虎, 刘炳杨, 郭欣. 基于特征变量扩展的含气饱和度随机森林预测方法[J]. 岩性油气藏, 2024, 36(2): 65-75.
[4]	李长海, 赵伦, 刘波, 赵文琪, 王淑琴, 李建新, 郑天宇, 李伟强. 滨里海盆地东缘北特鲁瓦油田石炭系碳酸盐岩储层裂缝网络连通性评价[J]. 岩性油气藏, 2024, 36(2): 113-123.
[5]	陈叔阳, 何云峰, 王立鑫, 尚浩杰, 杨昕睿, 尹艳树. 塔里木盆地顺北1号断裂带奥陶系碳酸盐岩储层结构表征及三维地质建模[J]. 岩性油气藏, 2024, 36(2): 124-135.
[6]	杨午阳, 魏新建, 李海山. 智能物探技术的过去、现在与未来[J]. 岩性油气藏, 2024, 36(2): 170-188.
[7]	王雪柯, 王震, 计智锋, 尹微, 姜仁, 侯珏, 张艺琼. 滨里海盆地东缘石炭系盐下碳酸盐岩油气藏成藏规律与勘探技术[J]. 岩性油气藏, 2023, 35(6): 54-62.
[8]	罗贝维, 尹继全, 胡广成, 陈华, 康敬程, 肖萌, 朱秋影, 段海岗. 阿联酋西部地区白垩系森诺曼阶高孔渗灰岩储层特征及控制因素[J]. 岩性油气藏, 2023, 35(6): 63-71.
[9]	范蕊, 刘卉, 杨沛广, 孙星, 马辉, 郝菲, 张珊珊. 阿曼盆地A区白垩系泥岩充填型碳酸盐岩溶蚀沟谷识别技术[J]. 岩性油气藏, 2023, 35(6): 72-81.
[10]	郭海峰, 肖坤叶, 程晓东, 杜业波, 杜旭东, 倪国辉, 李贤兵, 计然. 乍得Bongor盆地花岗岩潜山裂缝型储层有效渗透率计算方法[J]. 岩性油气藏, 2023, 35(6): 117-126.
[11]	刘亚明, 王丹丹, 田作基, 张志伟, 王童奎, 王朝锋, 阳孝法, 周玉冰. 巴西桑托斯盆地复杂碳酸盐岩油田火成岩发育特征及预测方法[J]. 岩性油气藏, 2023, 35(6): 127-137.
[12]	唐昱哲, 柴辉, 王红军, 张良杰, 陈鹏羽, 张文起, 蒋凌志, 潘兴明. 中亚阿姆河右岸东部地区侏罗系盐下碳酸盐岩储层特征及预测新方法[J]. 岩性油气藏, 2023, 35(6): 147-158.
[13]	许鑫, 杨午阳, 张凯, 魏新建, 张向阳, 李海山. 三维初至波旅行时层析速度反演算法优化[J]. 岩性油气藏, 2023, 35(4): 79-89.
[14]	王建功, 李江涛, 李翔, 高妍芳, 张平, 孙秀建, 白亚东, 左洺滔. 柴西地区新生界湖相微生物碳酸盐岩岩相组合差异性及控制因素[J]. 岩性油气藏, 2023, 35(3): 1-17.
[15]	宋兴国, 陈石, 杨明慧, 谢舟, 康鹏飞, 李婷, 陈九洲, 彭梓俊. 塔里木盆地富满油田F_Ⅰ16断裂发育特征及其对油气分布的影响[J]. 岩性油气藏, 2023, 35(3): 99-109.