电成像测井新参数在碳酸盐岩岩相识别中的应用  

Application of New Parameters of Electrical Imaging Logging in Carbonate Facies Identification

作　　者：李昌 沈安江[1,2] 孟贺 LI Chang;SHEN An-jiang;MENG He(Petrochina Hangzhou Research Institute of Geology, Hangzhou 310023, China;Key Laboratory of Carbonate Reservoirs, CNPC, Hangzhou 310023, China)

机构地区：[1]中国石油杭州地质研究院,杭州310023 [2]中国石油天然气集团公司碳酸盐岩储层重点实验室,杭州310023

出　　处：《科学技术与工程》2021年第26期11130-11135,共6页Science Technology and Engineering

基　　金：国家科技重大专项(2016ZX05004002)。

摘　　要：电成像测井识别碳酸盐岩岩相主要有定性图版法、定量参数法和机器学习法。图版法简单实用,但其效率低且受人工经验影响大,图像纹理参数种类过多且使用复杂,机器学习法需要大量样本标签,应用受到局限。为此,提出了一种简单又高效的新定量参数法,即通过新定量参数判别岩石构造特征来识别不同岩相。首先采用地质统计法对电成像测井动态图像进行全井眼插值,然后针对全井壁覆盖图像,利用数字图像处理技术获取二值图像,最后分别统计二值图像在纵向和横向黑色斑点(块)最大个数,并将二者的比值定义为视岩石构造数ARSN(aparent rock structure number)。ARSN可以有效区分块状和薄层状构造。据野外露头及岩心观察,一般颗粒云岩相为块状或厚层状和泥晶云岩相为薄层状或薄互层状构造特征。因此,利用ARSN可以区分颗粒云岩和泥晶云岩两大类岩相。以四川盆地M地区龙王庙组为例,经取心井验证,岩相识别符合率80%以上。该方法效率高且不受人的因素影响,实现高精度岩相识别,为该区碳酸盐岩沉积微相精细研究提供了有力技术支撑。The identification methods for carbonate lithofacies by electrical imaging logging mainly include qualitative plate method,quantitative parameter method and machine learning method.For the simple and practical plate method,it has low efficiency and is significantly affected by artificial experience.The quantitative parameter method contains many kinds of texture parameters,making it complex and difficult to use.Machine learning method requires a large number of sample labels,which limits its application.Therefore,a simple and efficient quantitative parameter method was proposed to identify carbonate lithofacies by distinguishing the rock structural characteristics.Firstly,the geostatistics method was used for the whole borehole interpolation of dynamic image of electrical imaging logging.Then,the binary image was obtained by using the digital image processing technology.Finally,the maximum number of black spots or blocks was counted respectively in the vertical and horizontal directions of the binary image.The ratio of the two numbers was defined as the apparent rock structure number(ARSN),which could effectively distinguish the massive and thin-layer structures.Generally,according to the field outcrop and core observation,the dolograinstone facies is massive or thick bed and the dolomudstone facies is thin bed or thin interbed,thus ARSN can identify the dolograinstone and the dolomudstone.Taking Longwangmiao formation in M area,Sichuan Basin as an example,the coincidence rate of lithofacies identification was more than 80%through cores verification.Because it is not affected by human factors,this method realizes high-precision lithofacies identification with high efficiency,which provides strong technical support for the fine study of carbonate sedimentary microfacies in this area.

关 键 词：电成像测井 碳酸盐岩 岩相测井识别 定量参数 四川盆地 龙王庙组 

分 类 号：TE122[石油与天然气工程—油气勘探]