基于改进Stacking算法的碳酸盐岩储层测井岩性识别方法与应用

doi:10.3969/j.issn.1006-6535.2025.04.007

摘要/Abstract

摘要： 针对川中地区碳酸盐岩储层传统岩性识别方法精度低、模型泛化能力弱的问题,提出一种基于改进Stacking算法的测井岩性识别方法。该方法融合多种机器学习模型的优势,优化特征加权策略,可提高对测井曲线关键信息的提取能力,同时增强对复杂岩性的识别准确性和稳定性。相比传统方法,该模型能够更有效地捕捉测井数据的非线性关系,并降低不同岩性类别间的预测混淆度。研究结果表明:该方法在四川盆地川中地区碳酸盐岩储层的岩性识别精度达到96%,较传统模型提升6个百分点,且平均相对误差更低,预测效果更优。改进的Stacking算法结合高效计算框架,可显著提升训练和预测效率,使岩性识别更加高效、可靠。该方法可有效地识别复杂岩性,为碳酸盐岩储层岩性识别提供参考。

关键词: Stacking, 集成学习, 特征加权, 碳酸盐岩, 岩性识别

Abstract: Conventional lithology identification methods for carbonate reservoirs in the Central Sichuan Area have low accuracy and weak model generalization. To address this, a lithology identification method in well logging based on an improved Stacking algorithm is proposed. This method integrates the advantages of multiple machine learning models, optimizes feature weighting strategies, and enhances the extraction of key information from logging curves, improving the accuracy and stability of complex lithology identification. Compared to traditional methods, it better captures the nonlinear relationships in logging data and reduces prediction confusion between lithology categories. The study results show that the accuracy of this method in identifying lithologies in central Sichuan carbonate reservoirs reaches 96%, an improvement of over 6 percentage point compared to traditional models. It also has lower average relative errors and better prediction performance. Combined with an efficient computing framework, the improved Stacking algorithm significantly enhances training and prediction efficiency, making lithology identification both efficient and reliable. This method effectively identifies complex lithologies and provides a valuable reference for carbonate reservoir lithology identification.

Key words: Stacking, ensemble learning, feature weighting, carbonate, lithology identification

中图分类号:

TE122

罗水亮, 漆影强, 唐松, 阮基富, 高达, 刘乾乾, 李生. 基于改进Stacking算法的碳酸盐岩储层测井岩性识别方法与应用[J]. 特种油气藏, 2025, 32(4): 58-67.

LUO Shuiliang, QI Yingqiang, TANG Song, RUAN Jifu, GAO Da, LIU Qianqian, LI Sheng. Lithology logging identification method and application to carbonate reservoirs based on improved Stacking algorithm[J]. Special Oil & Gas Reservoirs, 2025, 32(4): 58-67.

参考文献

[1] 赵文智,沈安江,乔占峰,等.中国碳酸盐岩沉积储层理论进展与海相大油气田发现[J].中国石油勘探,2022,27(4):1-15.
ZHAO Wenzhi,SHEN Anjiang,QIAO Zhanfeng,et al.Theoretical progress in carbonate reservoir and discovery of large marine oil and gas fields in China[J].China Petroleum Exploration,2022,27(4):1-15.
[2] 成大伟,袁选俊,周川闽,等.测井岩性识别方法及应用——以鄂尔多斯盆地中西部长7油层组为例[J].中国石油勘探,2016,21(5):117-126.
CHENG Dawei,YUAN Xuanjun,ZHOU Chuanmin,et al.Logging-lithology identification methods and their application:a case study on Chang 7 Member in central-western Ordos Basin,NW China[J].China Petroleum Exploration,2016,21(5):117-126.
[3] 刘凯,邹正银,王志章,等.基于机器学习的火山岩岩性智能识别及预测[J].特种油气藏,2022,29(1):38-45.
LIU Kai,ZOU Zhengyin,WANG Zhizhang,et al.Intelligent identification and prediction of lithology of volcanic reservoirs based on machine learning[J].Special Oil & Gas Reservoirs,2022,29(1):38-45.
[4] 武中原,张欣,张春雷,等.基于LSTM循环神经网络的岩性识别方法[J].岩性油气藏,2021,33(3):120-128.
WU Zhongyuan,ZHANG Xin,ZHANG Chunlei,et al.Lithology identification based on LSTM recurrent neural network[J].Lithologic Reservoirs,2021,33(3):120-128.
[5] 徐晗,姚孔轩,程丹仪,等.基于非开挖随钻检测系统与随机森林的地层岩性识别[J].地质科技通报,2021,40(5):272-280.
XU Han,YAO Kongxuan,CHENG Danyi,et al.Stratigraphic lithology identification based on no-dig logging while drilling system and random forest[J].Bulletin of Geological Science and Technology,2021,40(5):272-280.
[6] 孙岿.基于改进KNN算法的潜山复杂岩性测井识别方法[J].特种油气藏,2022,29(3):18-27.
SUN Kui.Logging identification method of complex lithology in buried hill based on the improved KNN algorithm[J].Special Oil & Gas Reservoirs,2022,29(3):18-27.
[7] 宋梓豪,巩红雨,冉爱华,等.基于ADASYN-GS-XGBOOST混合模型的火山岩测井岩性识别[J].海相油气地质,2024,29(2):188-196.
SONG Zihao,GONG Hongyu,RAN Aihua,et al.Lithology logging identification of volcanic rock based on ADASYN-GS-XGBOOST hybrid model[J].Marine Origin Petroleum Geology,2024,29(2):188-196.
[8] 史鹏宇,徐思慧,冯加明,等.基于改进Stacking算法的致密砂岩储层测井流体识别[J].地球物理学进展,2024,39(1):280-290.
SHI Pengyu,XU Sihui,FENG Jiaming,et al.Log identification of fluid types in tight sandstone reservoirs using an improved stacking algorithm[J].Progress in Geophysics,2024,39(1):280-290.
[9] 汪海燕,黎建辉,杨风雷.支持向量机理论及算法研究综述[J].计算机应用研究,2014,31(5):1281-1286.
WANG Haiyan,LI Jianhui,YANG Fenglei.Overview of support vector machine analysis and algorithm[J].Application Research of Computers,2014,31(5):1281-1286.
[10] 张驰,潘懋,胡水清,等.融合储层纵向信息的机器学习岩性识别方法[J].地质科技通报,2023,42(3):289-299.
ZHANG Chi,PAN Mao,HU Shuiqing,et al.A machine learning lithologic identification method combined with vertical reservoir information[J].Bulletin of Geological Science and Technology,2023,42(3):289-299.
[11] WANG X D,YANG S C,ZHAO Y F,et al.Lithology identification using an optimized KNN clustering method based on entropy-weighed cosine distance in Mesozoic strata of Gaoqing Field,Jiyang Depression[J].Journal of Petroleum Science and Engineering,2018,166:157-174.
[12] 武中原.序列数据机器学习方法在岩石物理参数预测中的应用研究[D].北京:中国地质大学(北京),2021.
WU Zhongyuan.Application of machine learning methods for sequence data in the prediction of rock physical parameters[D].Beijing:China University of Geosciences(Beijing),2021.
[13] BREIMAR L.Random Forests[J].Machine Learning,2001,45(1):5-32.
[14] 李玉伟,李子健,邵力飞,等.基于物理信息约束的页岩油储层可压性评价新方法[J].煤田地质与勘探,2023,51(10):37-51.
LI Yuwei,LI Zijian,SHAO Lifei,et al.A new physics-informed method for the fracability evaluation of shale oil reservoirs[J].Coal Geology & Exploration,2023,51(10):37-51.
[15] 邢强,张晋言,王镇方,等.基于XGBoost的测井解释规则库自动获取方法[J].石油物探,2022,61(2):356-363.
XING Qiang,ZHANG Jinyan,WANG Zhenfang,et al.Automatic acquisition of a rule base for logging interpretaion using the XGBoost algorithm[J].Geophysical Prospecting for Petroleum,2022,61(2):356-363.
[16] 王俊,曹俊兴,尤加春.基于GRU神经网络的测井曲线重构[J].石油地球物理勘探,2020,55(3):510-520.
WANG Jun,CAO Junxing,YOU Jiachun.Reconstruction of logging traces based on GRU neural network[J].Oil Geophysical Prospecting,2020,55(3):510-520.
[17] 王俊,曹俊兴,尤加春,等.基于门控循环单元神经网络的储层孔渗饱参数预测[J].石油物探,2020,59(4):616-627.
WANG Jun,CAO Junxing,YOU Jiachun,et al.Prediction of reservoir porosity,permeability,and saturation based on a gated recurrent unit neural network[J].Geophysical Prospecting for Petroleum,2020,59(4):616-627.
[18] 罗辉,吕慧栋,周嘉安琪,等.密度泛函理论与分子力学方法研究沥青质分子间的π-π堆积作用[J].中国石油大学学报(自然科学版),2023,47(1):189-196.
LUO Hui,LYU Huidong,ZHOU Jiaanqi,et al.Study on π-π stacking between asphaltene molecules by DFT and MM methods[J].Journal of China University of Petroleum(Edition of Natural Science),2023,47(1):189-196.
[19] 刘磊,李伟,杜玉山,等.基于Stacking集成学习的分频地震属性融合储层预测方法[J].石油地球物理勘探,2024,59(1):12-22.
LIU Lei,LI Wei,DU Yushan,et al.Reservoir prediction method of fusing frequency-decomposed seismic attri-butes using Stacking ensemble learning[J].Oil Geophysical Prospecting,2024,59(1):12-22.
[20] 李泉伦,陈争光,焦峰.基于Stacking集成学习的近红外光谱油页岩含油率预测[J].光谱学与光谱分析,2023,43(4):1030-1036.
LI Quanlun,CHEN Zhengguang,JIAO Feng.Prediction of oil content in oil shale by near-infrared spectroscopy based on stacking ensemble learning[J].Spectroscopy and Spectral Analysis,2023,43(4):1030-1036.
[21] 陈怡.基于集成学习的储层物性参数预测方法研究[D].大庆:东北石油大学,2023.
CHEN Yi.Research on reservoir property parameter prediction method based on ensemble learning[D].Daqing:Northeast Petroleum University,2023.
[22] 谭文侃,胡南燕,叶义成,等.基于四大集成学习的岩爆烈度分级预测[J].岩石力学与工程学报,2022,41(增刊2):3250-3259.
TAN Wenkan,HU Nanyan,YE Yicheng,et al.Rockburst intensity classification prediction based on four ensemble learning[J].Chinese Journal of Rock Mechanics and Engineering,2022,41(S2):3250-3259.
[23] 刘宏,王高峰,刘南,等.磨溪气田嘉陵江组二段陆表海台地高频层序发育特征及对储层的控制[J].石油与天然气地质,2016,37(5):713-720.
LIU Hong,WANG Gaofeng,LIU Nan,et al.Characterization of high-frequency sequences in the epicontinental carbonate platform of Jia 2 Member in Moxi Gas Field and its controls over reservoirs[J].Oil & Gas Geology,2016,37(5):713-720.
[24] 张艳.苏里格气田东区碳酸盐岩储层评价及含气性预测[D].北京:中国地质大学(北京),2018.
ZHANG Yan.Evaluation of carbonate reservoirs and prediction of gas content in the eastern area of Sulige Gasfield[D].Beijing:China University of Geosciences,(Beijing),2018.
[25] 孙予舒,黄芸,梁婷,等.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,et al.Identification of complex carbonate lithology by logging based on XGBoost algorithm[J].Lithologic Reservoirs,2020,32(4):98-106.
[26] 江宝得,李秀春,罗海燕,等.异质集成学习在滑坡易发性评价中的对比研究[J].土木工程学报,2023,56(10):170-179.
JIANG Baode,LI Xiuchun,LUO Haiyan,et al.A comparative analysis of heterogeneous ensemble learning methods for landslide susceptibility assessment[J].China Civil Engineering Journal,2023,56(10):170-179.
[27] 查文舒,乔奇,刘子雄,等.基于相关性分析的Bi-LSTM测井曲线预测方法[J].合肥工业大学学报(自然科学版),2022,45(5):700-706.
ZHA Wenshu,QIAO Qi,LIU Zixiong,et al.A method for well log prediction using bidirectional long short-term memory based on correlation analysis[J].Journal of Hefei University of Techenology(Natural Science),2022,45(5):700-706.
[28] 王俊,曹俊兴,尤加春,等.基于门控循环单元神经网络的储层孔渗饱参数预测[J].石油物探,2020,59(4):616-627.
WANG Jun,CAO Junxing,YOU Jiachun,et al.Prediction of reservoir porosity,permeability,and saturation based on a gated recurrent unit neural network[J].Geophysical Prospecting for Petroleum,2020,59(4):616-627.
[29] 巩维嘉.基于集成学习的岩性识别研究与应用[D].大庆:东北石油大学,2022.
GONG Weijia.Research and application of lithology identification based on ensemble learning[D].Daqing:Northeast Petroleum University,2022.
[30] 王新领,祝新益,张宏兵,等.基于随机树嵌入的随钻测井岩性识别方法[J].吉林大学学报(地球科学版),2024,54(2):701-708.
WANG Xinling,ZHU Xinyi,ZHANG Hongbing,et al.Lithology identification method for logging while drilling based on random tree embedding[J].Journal of Jilin University(Earth Science Edition),2024,54(2):701-708.