R语言在特低渗透孔隙结构与可动流体关系研究中的应用  

Application of R language in studying the relationship between the super-low permeability pore structure and movable fluid

作　　者：蒋建方 黄勇[1,2] 熊婷 仇晓丽 李诗豪 张甜 王馨玉 JIANG Jianfang;HUANG Yong;XIONG Ting;QIU Xiaoli;LI Shihao;ZHANG Tian;WANG Xinyu(Research Institute of Enhanced Oil Recovery,China University of Petroleum,Beijing 102249,China;MOE Key Laboratory of Petroleum Engineering,China University of Petroleum,Beijing 102249,China)

机构地区：[1]中国石油大学提高采收率研究院,北京102249 [2]中国石油大学石油工程教育部重点实验室,北京102249

出　　处：《大庆石油地质与开发》2019年第3期46-53,共8页Petroleum Geology & Oilfield Development in Daqing

基　　金：“十三五”国家科技重大专项课题“低渗—致密油藏高效提高采收率新技术”(2017ZX05009-004)

摘　　要：R语言是一种用于统计分析、绘图的编程语言和平台,主要用于实现复杂数据的可视化。多孔介质的微观孔隙结构对可动流体影响复杂,因此把R语言数据分析方法引入到多因素影响关系的研究中。对长庆油田长8储层岩心进行了恒速压汞、核磁共振实验,分别测定了特低渗砂岩储层孔隙结构及可动流体等参数。恒速压汞实验表明,储层孔隙半径主要为60~120μm,喉道半径集中在1~3μm,孔喉比均值为344.78;核磁共振测定结果表明,可动流体比例为39.69%~55.15%,平均值为46.16%。将R语言用于孔隙半径、喉道半径等与可动流体比例关系研究中的结果显示:单因素分析未考虑多变量间的相互影响,孔隙结构与可动流体关系不理想;通过R语言建立了多元线性模型,拟合优度较高(R^2=0.98),其中,渗透率对可动流体比例的影响相对权重最高(32.92%),余下依次为主流喉道半径(16.36%)、孔隙度(13.99%)、孔喉比(13.30%)、平均喉道半径(12.29%)和孔喉分选系数(11.14%)。研究表明,用R语言能够从新的层面对孔隙结构与可动流体关系进行分析,对可动流体赋存特征研究具有重要的指导意义。R language is one of the programming languages and platforms for statistical analysis and drawing, and mainly used for the visualization of the complex data. The microscopic pore structure of the porous media has the complex influence on the movable fluid, so the R language data analysis method was introduced into the study of the multi-factor impacting relationship. The pore structure and movable fluid parameters were determined by constant-speed mercury intrusion (CSMI) and nuclear magnetic resonance (NMR) experiments of the cores from Chang 8 super-low permeability sandstone reservoir in Changqing Oilfield. CSMI results demonstrate that the reservoir pore radii are mainly between 60 μm and 120 μm, the throat radii are concentrated in the range from 1 μm to 3 μm, and the average pore-throat ratio is 344.78;NMR results display that the percentages of the movable fluid are between 39.69% and 55.15% with an average of 46.16%. The results of studying the relationship between the pore radius, throat radius etc. and the movable fluid percentage by means of the R-language show that the single-factor analysis doesn't consider the interaction among the multiple variables, and the relationship between the pore structure and mobile fluid is not ideal;A multiple linear model was established through the R language, with mach higher degree of the goodness of fit ( R^2=0.98), among them, the permeability has the highest relative weight on the percentage of the movable fluid (32.92%), the next are the mainstream throat radius (16.36%), porosity (13.99%), pore throat ratio (13.30%), mean throat radius (12.29%), and sorting coefficient of the pore throat (11.14%). The study results show that the R language can analyze the relationship between the pore structure and the movable fluid from a new aspect, and moreover it has important guiding significance for studying the existence characteristics of the movable fluid.

关 键 词：R语言 特低渗储层 孔隙结构 可动流体 拟合优度 

分 类 号：TE312[石油与天然气工程—油气田开发工程]