混积型碳酸盐岩储层特征及关键参数计算——以四川盆地蓬莱气区寒武系沧浪铺组为例  

Characteristics and key parameter calculation of diamictic carbonate reservoirs:An example from the Cambrian Canglangpu Formation,Penglai gas area,Sichuan Basin

作　　者：汪泽宇 赖强 吴煜宇 李淑荣 殷榕 王海青 谌丽 陈邦定 黄劲松 WANG Zeyu;LAI Qiang;WU Yuyu;LI Shurong;YIN Rong;WANG Haiqing;SHEN Li;CHEN Bangding;HUANG Jingsong(Exploration and Development Research Institute,PetroChina Southwest Oil&Gasfield Company,Chengdu,Sichuan 610041,China;Shandong Leading Petro-Tech Co.,Ltd.,Dongying,Shandong 257100,China;Southwest Branch,China National Logging Corporation,Chongqing 400021,China)

机构地区：[1]中国石油西南油气田公司勘探开发研究院,四川成都610041 [2]山东立鼎石油科技有限公司,山东东营257100 [3]中国石油集团测井有限公司西南分公司,重庆400021

出　　处：《天然气勘探与开发》2024年第2期35-44,共10页Natural Gas Exploration and Development

摘　　要：四川盆地川中地区乐山-龙女寺古隆起北斜坡蓬莱气区下寒武统沧浪铺组一段为清水浅水陆棚和混积浅水陆棚交替沉积,硅质碎屑岩和碳酸盐岩沉积物在纵向上互层或平面上交替沉积,发育泥质、硅质、方解石、白云石和黄铁矿等多种矿物,组成岩石类型多样。储层岩性纵横向变化快,且低孔低渗,不同岩性的孔渗关系迥异,传统的双矿物计算模型难以满足混积岩复杂矿物成分计算要求,相应的孔隙度计算精度也无法保证。为解决混积型碳酸盐岩储层复杂矿物含量和孔隙度两个关键参数计算问题开展研究,通过应用大量的全岩X衍射、物性、压汞实验分析数据和岩化分析、岩性扫描测井标定的常规测井信息,形成降维法、神经网络分析法、元素反演矿物组分法和最优化处理法4种计算沧浪铺组复杂岩石矿物组分方法。研究结果表明:4种方法的计算结果与全岩分析和岩性扫描均具有一致性,其中最优化处理结果最佳,降维法可操作性最强;在获得准确矿物组分含量基础上,建立基于三孔隙度曲线的变骨架参数孔隙度计算模型,用于混积型碳酸盐岩储层测井评价,具有重要意义。The first member of the Lower Cambrian Canglangpu Formation,Penglai gas area,north slope of Leshan-Longlvsi paleouplift,central Sichuan Basin,is developed with some deposits of fresh shallow-water shelf succeeded by diamictic shallow-water shelf.Among which,siliciclastic and carbonate present the depostional pattern of vertical alternation or horizontal succession.In this member,there extended a variety of minerals,such as pelite,silicalite,calcite,dolomite and pyrite,which form various rock types.Moreover,exhibiting great vertical and horizontal variation in lithology,the reservoirs are characterized by low porosity and permeability.Distinct lithology makes the relation of porosity to permeability vastly different.Most traditional dual-mineral calculation models are not available for the complex mineral composition in diamictic rocks.As a result,the porosity calculation accuracy cannot be guaranteed.So,calculating the key parameters in diamictic carbonate reservoirs was discussed,namely mineral content and porosity.Based on a host of experimental data on XRD,physical property,and mercury injection together with conventional logging data calibrated by petrochemical analysis and LithoScanner logging,four methods,containing dimensionality reduction,neural network analysis,element content to mineral content inversion,and optimal processing,were established for the Canglangpu Formation to calculate its mineral content.Results show that(i)findings from these four methods are consistent with those from whole rock analysis and lithological scanning.Specifically,the optimal processing achieves the best results,while the dimensionality reduction with the best operability;(ii)taken tri-porosity curves into account and dependent on accurate mineral content,another model to calculate porosity of variable matrix parameters was built for logging evaluation on the diamictic carbonate reservoirs.

关 键 词：混积型碳酸盐岩 降维法 神经网络方法 元素反演矿物 最优化法 变骨架参数 

分 类 号：P618.13[天文地球—矿床学]