摘要
有效、精确的岩性识别是火山岩油气藏勘探的基础之一。针对准噶尔盆地东部西泉地区石炭系火山岩,基于大量岩心和镜下薄片资料,优选出对岩性响应敏感的6条测井曲线,利用主成分分析法对各测井响应值进行了标准化处理,并建立了一个随机样本,构建出累计方差贡献率为86.798%的2个综合主成分变量Y1和Y2,提高了不同岩性之间测井响应特征差异敏感性;再利用典型判别分析法建立了研究区石炭系火山岩9种不同岩性的判别模型,并对其进行了识别。通过对原始样本进行回判性检验及交叉验证表明:该方法提高了岩性识别的准确性,识别火山岩岩性的综合准确率高于91%,不仅有效地消除了测井曲线之间携带岩性信息的相关性和重叠性,而且在火山岩薄互层识别上比常规岩性识别方法的精度更高。这一方法为研究区火山岩岩性的精确识别提供了参考依据。
Effective and accurate lithology identification is one of the most important foundations for volcanic reservoir exploration.Aiming at the Carboniferous volcanic rocks in Xiquan area of the eastern Junggar Basin,based on a large amount of coring data and micro-slice data,six well logging curves which were sensitive to lithology were selected,and each of them were standardized to establish a random sample by principal component analysis,and two comprehensive variables Y1 and Y2 with a cumulative variance contribution rate of 86.798%were constructed to improve the sensitivity of logging response characteristics between different lithologies.The typical discriminant analysis method was used to establish discriminant models for nine different lithologies of Carboniferous volcanic rocks in the study area to identify volcanic rocks.The back-judgmental inspection and cross-validation of the original samples show that the accuracy of lithology identification was improved by using this method,its comprehensive accuracy of lithology identification was over 91%.The correlation and overlap of lithologic information carried by logging curves were effectively eliminated,and the accuracy of this method is higher than that of conventional lithology identification methods in thin interbedded volcanic rocks identification.The study provides a reference for accurate lithology identification of volcanic rocks in the study area.
作者
李树博
郭旭光
郑孟林
王泽胜
刘新龙
LI Shubo;GUO Xuguang;ZHENG Menglin;WANG Zesheng;LIU Xinlong(Research Institute of Exploration and Development,PetroChina Xinjiang Oilfield Company,Karamay 834000,Xinjiang,China)
出处
《岩性油气藏》
CSCD
北大核心
2021年第1期258-266,共9页
Lithologic Reservoirs
基金
中国石油股份公司科技专项“准噶尔盆地石油勘探重点领域成藏规律及勘探技术研究”(编号:2017E-0401)资助。
关键词
火山岩
岩性识别
数据挖掘方法
西泉地区
准噶尔盆地
volcanic rocks
lithology identification
data mining method
Xiquan area
Junggar Basin