基于Stacking集成学习的分频地震属性融合储层预测方法

doi:10.13810/j.cnki.issn.1000-7210.2024.01.002

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (7727 KB) HTML []
输出: BibTeX | EndNote (RIS)

摘要地震属性蕴含大量储层信息，融合多种地震属性可提高储层预测精度。由于地下地质结构复杂、非均质性强，依据单一的地震属性融合方法难以精细刻画储层特征。为此，提出了一种基于Stacking集成学习的分频地震属性融合储层预测方法。该方法主要包括3个部分：①根据不同厚度储层的振幅与频率关系，利用多个频率的地震信息，降低地震属性的多解性； ②联合相关性分析和无监督聚类技术优选地震属性，剔除冗余属性特征； ③利用能够综合多个差异化模型优势的Stacking集成学习模型，融合不同频段的地震属性，提高地震属性的解释精度。将该方法用于渤海湾盆地埕岛油田，并使用线性公式定量分析法进一步评估Stacking模型的泛化效果。结果显示：与单类预测模型相比，Stacking模型的综合预测性能和可靠性均有显著提升；对应的地震属性融合结果高值区形态更加清晰，融合属性与砂体厚度的相关系数可达到0.92，这表明该方法具有良好的应用前景。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	刘磊
	李伟
	杜玉山
	岳大力
	张雪婷
	侯加根

关键词 ：地震属性, 储层预测, Stacking集成学习, 分频, 智能融合

Abstract：Seismic attributes contain a large amount of reservoir information， and the fusion of multiple seismic attributes can improve the precision of reservoir prediction. However， due to the complex underground geological structure and strong heterogeneity， a single fusion method of seismic attribute is difficult to describe the reservoir characteristics. Therefore， a reservoir prediction method of fusing frequency-decomposed seismic attributes using Stacking ensemble learning is proposed. The method consists of three main parts： ①Based on the amplitude and frequency relationship of the reservoir with different sand thicknesses， seismic data with varying frequencies and bandwidths are considered to reduce the uncertainties of seismic attribute interpretation. ②Jointly preferring seismic attributes based on correlation analysis and unsupervised clustering is performed to eliminate redundant information of seismic attributes. ③A Stacking ensemble learning model， which can combine the advantages of various models， is designed to fuse seismic attributes with different frequencies and bandwidths for improving seismic attribute interpretation resolution. The proposed method is applied by taking the Chengdao Oilfield in the Bohai Bay Basin as an example. The quantitative linear formula analysis method is proposed to further evaluate the generalization effect of the Stacking model. The results prove that comprehensive prediction accuracy and the reliability of the Stacking model are significantly improved compared with those of the single-class models. The corresponding high-value areas of the fusion attribute are more evident， and the correlation coefficient between the fusion attribute and sand thickness can reach 0.92， indicating that the method has great potential in reservoir prediction.

Key words： seismic attribute reservoir prediction Stacking ensemble learning frequency-decomposed intelligent fusion

收稿日期: 2023-05-30

PACS:

P631

基金资助:本项研究受国家自然科学基金项目“坳陷湖盆洪水型湖相重力流沉积演化机理及差异构型模式”（42272186）、“坡度与水深主控的河流辫—曲转换机理及其沉积响应”（42202109）、中国石油战略合作专项“鄂尔多斯盆地致密油—页岩油储层非均质成因机理与表征技术”（ZLZX2020-02）、北京市科协青年人才托举工程（BYESS2023460）和博士后创新人才支持计划项目“少井条件下沉积过程与地震正演驱动的地震智能反演方法”（BX20220351）联合资助。

通讯作者: 岳大力,北京市昌平区府学路18号中国石油大学(北京)人工智能学院,102249。Email:yuedali@cup.edu.cn E-mail: yuedali@cup.edu.cn

作者简介: 刘磊博士研究生，1996年生； 2021年获中国石油大学（北京）智能科学与技术专业学士学位；现在该校攻读地质资源与地质工程专业博士学位，主要从事油气田开发地质方面的学习和研究。

引用本文:

刘磊, 李伟, 杜玉山, 岳大力, 张雪婷, 侯加根. 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, YUE Dali, ZHANG Xueting, HOU Jiagen. Reservoir prediction method of fusing frequency-decomposed seismic attributes using Stacking ensemble learning. Oil Geophysical Prospecting, 2024, 59(1): 12-22.

链接本文:

http://www.ogp-cn.com/CN/10.13810/j.cnki.issn.1000-7210.2024.01.002 或 http://www.ogp-cn.com/CN/Y2024/V59/I1/12

[1]	陈芊澍, 文晓涛, 何健, 等. 基于极限学习机的裂缝带预测[J]. 石油物探, 2021, 60(1): 149-156, 174.CHEN Qianshu, WEN Xiaotao, HE Jian, et al. Prediction of a fracture zone using an extreme learning machine[J]. Geophysical Prospecting for Petroleum, 2021, 60(1): 149-156, 174.
[2]	董少群, 曾联波, 杜相仪, 等. 致密碳酸盐岩储集层裂缝智能预测方法[J].石油勘探与开发,2022,49(6): 1179-1189.DONG Shaoqun, ZENG Lianbo, DU Xiangyi, et al. An intelligent prediction method of fractures in tight carbonate reservoirs[J]. Petroleum Exploration and Development, 2022, 49(6): 1179-1189.
[3]	李婷婷, 侯思宇, 马世忠, 等. 断层识别方法综述及研究进展[J]. 地球物理学进展, 2018, 33(4): 1507-1514.LI Tingting, HOU Siyu, MA Shizhong, et al. Overview and research progress of fault identification method[J]. Progress in Geophysics, 2018, 33(4): 1507-1514.
[4]	岳大力, 李伟, 王军, 等. 基于分频融合地震属性的曲流带预测与点坝识别——以渤海湾盆地埕岛油田馆陶组为例[J]. 古地理学报, 2018, 20(6): 941-950.YUE Dali, LI Wei, WANG Jun, et al. Prediction of meandering belt and point-bar recognition based on spectral-decomposed and fused seismic attributes: A case study of the Guantao Formation, Chengdao Oilfield, Bohai Bay Basin[J]. Journal of Palaeogeography, 2018, 20(6): 941-950.
[5]	曾亚丽, 龙盛芳, 吴朦朦, 等. 基于GeoEast地震属性分析的储层预测技术及其在环县地区页岩油勘探开发中的应用[J]. 石油地球物理勘探, 2022, 57(增刊1): 196-201.ZENG Yali, LONG Shengfang, WU Mengmeng, et al. Reservoir prediction technique based on GeoEast seismic attribute analysis and its application in shale oil exploration and development in Huanxian[J]. Oil Geophysical Prospecting, 2022, 57(S1): 196-201.
[6]	钟晗. 地震属性在储层预测中的应用研究[D]. 北京: 中国石油大学(北京), 2018.ZHONG Han. The Application of Seismic Attributes in the Reservoir Prediction[D]. China University of Petroleum (Beijing), Beijing, 2018.
[7]	张树林. 海上多波多分量地震烃类直接检测方法研究及应用[J]. 物探化探计算技术, 2000, 22(3): 200-206.ZHANG Shulin. A direct hydrocarbon detecting method and application of offshore multiwave and multicomponent seismic data[J]. Computing Techniques for Geophysical and Geochemical Exploration, 2000, 22(3): 200-206.
[8]	郑江峰, 彭刚, 孙佳林, 等. 基于90°相移的振幅和频率属性融合法油气检测[J]. 石油物探, 2019, 58(1): 130-138.ZHENG Jiangfeng, PENG Gang, SUN Jialin, et al. Fusing amplitude and frequency attributes for hydrocarbon detection using 90° phase shift data[J]. Geophysical Prospecting for Petroleum, 2019, 58(1): 130-138.
[9]	岳大力, 李伟, 杜玉山, 等. 河流相储层地震属性优选与融合方法综述[J]. 地球科学, 2022, 47(11): 3929-3943.YUE Dali, LI Wei, DU Yushan, et al. Review on optimization and fusion of seismic attributes for fluvial reservoir characterization[J].Earth Science,2022,47(11): 3929-3943.
[10]	CHEN Q, SIDNEY S. Seismic attribute technology for reservoir forecasting and monitoring[J]. The Leading Edge, 1997, 16(5): 445-448.
[11]	HART B S. Channel detection in 3-D seismic data using sweetness[J]. AAPG Bulletin, 2008, 92(6): 733-742.
[12]	张家乐, 程冰洁, 徐天吉, 等. 应用地震属性主成分分析识别古河道[J].石油地球物理勘探,2023,58(1): 190-195.ZHANG Jiale, CHENG Bingjie, XU Tianji, et al. Identification of paleochannels by seismic attribute principal component analysis[J]. Oil Geophysical Prospecting, 2023, 58(1): 190-195.
[13]	MCARDLE N J, ACKERS M A. Understanding seismic thin-bed responses using frequency decomposition and RGB blending[J]. First Break, 2012,30(12): 57-65.
[14]	李伟, 岳大力, 胡光义, 等. 分频段地震属性优选及砂体预测方法——秦皇岛32-6油田北区实例[J]. 石油地球物理勘探, 2017, 52(1): 121-130.LI Wei, YUE Dali, HU Guangyi, et al. Frequency-segmented seismic attribute optimization and sandbody distribution prediction: an example in North Block, Qinghuangdao 32-6 Oilfield[J]. Oil Geophysical Prospecting, 2017, 52(1): 121-130.
[15]	ZENG H. Thickness imaging for high-resolution stratigraphic interpretation by linear combination and color blending of multiple-frequency panels[J]. Interpretation, 2017, 5(3): T411-T422.
[16]	LI W, YUE D, WANG W, et al. Fusing multiple frequency-decomposed seismic attributes with machine learning for thickness prediction and sedimentary facies interpretation in fluvial reservoirs[J]. Journal of Petroleum Science and Engineering, 2019, 177: 1087-1102.
[17]	YUE D, LI W, WANG W, et al. Fused spectral-decomposition seismic attributes and forward seismic modelling to predict sand bodies in meandering fluvial reservoirs[J]. Marine and Petroleum Geology, 2019, 99: 27-44.
[18]	蔡义峰, 熊婷, 姚卫江, 等. 地震多属性分析技术在薄层砂体预测中的应用[J]. 石油地球物理勘探, 2017, 52(增刊2): 140-145.CAI Yifeng, XIONG Ting, YAO Weijiang, et al. Thin sandstone prediction with seismic multi-attribute analysis[J]. Oil Geophysical Prospecting,2017,52(S2): 140-145.
[19]	蔡国刚, 杨光达, 周艳, 等. 机器学习驱动的地震多属性分析表征扇三角洲沉积[J]. 科学技术与工程, 2021, 21(29): 12454-12460.CAI Guogang, YANG Guangda, ZHOU Yan, et al. Sedimentary facies characterization of fan delta based on machine learning driven multi-seismic attributes analysis[J]. Science Technology and Engineering, 2021, 21(29): 12454-12460.
[20]	张宪国, 吴啸啸, 黄德榕, 等. 极限学习机驱动的地震多属性融合识别曲流带单一点坝[J]. 石油地球物理勘探, 2021, 56(6): 1340-1350.ZHANG Xianguo, WU Xiaoxiao, HUANG Derong, et al. Single point bar interpretation in meandering belt with extreme learning machine driven multiple seismic attributes fusion[J]. Oil Geophysical Prospecting, 2021, 56(6): 1340-1350.
[21]	曲志鹏, 王芳芳, 张云银, 等. 基于关联规则与随机森林的地震多属性砂体厚度预测[J]. 地质科技通报, 2021, 40(3): 211-218.QU Zhipeng, WANG Fangfang, ZHANG Yunyin, et al. Thickness prediction of seismic multi-attributes sand based on association rules and random forests[J]. Bulletin of Geological Science and Technology, 2021, 40(3): 211-218.
[22]	陈欢庆, 王珏, 杜宜静. 储层非均质性研究方法进展[J]. 高校地质学报, 2017, 23(1): 104-116.CHEN Huanqing, WANG Jue, DU Yijing. Advances of research methods on reservoir heterogeneity[J]. Geological Journal of China Universities, 2017, 23(1): 104-116.
[23]	石鹏. 委内瑞拉JN-4区块储层非均质性研究[D]. 山东青岛:中国石油大学(华东), 2015.SHI Peng. The Reservoir Heterogeneity Research in JN-4 Block, Venezuela[D]. China University of Petroleum (East China), Qingdao,Shandong, 2015.
[24]	徐继伟, 杨云. 集成学习方法:研究综述[J]. 云南大学学报(自然科学版), 2018, 40(6): 1082-1092.XU Jiwei, YANG Yu. A survey of ensemble learning approaches[J]. Journal of Yunnan University (Natural Sciences Edition), 2018, 40(6): 1082-1092.
[25]	史佳琪, 张建华. 基于多模型融合Stacking集成学习方式的负荷预测方法[J]. 中国电机工程学报, 2019, 39(14): 4032-4041.SHI Jiaqi, ZHANG Jianhua. Load forecasting based on multi-model by Stacking ensemble learning[J]. Proceedings of the CSEE, 2019, 39(14): 4032-4041.
[26]	鲍海波, 吴阳晨, 张国应, 等. 基于特征加权Stacking集成学习的净负荷预测方法[J]. 电力建设, 2022, 43(9): 104-116.BAO Haibo, WU Yangchen, ZHANG Guoying, et al. Net load forecasting method based on feature-weighted Stacking ensemble learning[J]. Electric Power Construction, 2022, 43(9): 104-116.
[27]	BERGSTRA J, BENGIO Y. Random search for hyper parameter optimization[J]. Journal of Machine Learning Research, 2012, 13(1): 281-305.
[28]	ZENG H. Geologic significance of anomalous instantaneous frequency[J]. Geophysics, 2010, 75(3): P23-P30.