低阻油层综合分类及支持向量机评价——以Muglad盆地Fula凹陷AG1层为例

发布时间：2018-06-15 06:30

本文选题：低阻油层 + 综合分类　；参考：《国外测井技术》2016年06期

【摘要】：Muglad盆地Fula凹陷内大部分油田进入开发中后期,低阻油层成为重要的剩余资源。低阻油层预计的采油强度和累产油量等是在油井措施实施中的关键参数,直接决定经济效益。本文对凹陷内4个主力油田AG1层已生产的51层低阻油层的储层特征、渗流特征和产能特征进行分析,并以采油强度和累产油量为核心,将低阻油层分为三类。然后利用支持向量机对低阻油层所属类型进行判别,研究其对低阻油层类型及采油强度和累产油量等指标的预测能力。结果表明,以密度、声波时差、中子孔隙度、地层深电阻率、自然伽马、自然电位幅度比等作为模型变量的支持向量机分类器检验样本判别准确率为75%,可对低阻油层采油强度和累产油等进行一定程度的预测。
[Abstract]:In the Fula sag of Muglad basin, most of the oil fields enter the middle and late stage of development, and the low resistivity reservoir becomes an important remaining resource. The predicted oil recovery intensity and cumulative oil production in low resistivity reservoirs are the key parameters in the implementation of oil well measures, which directly determine the economic benefits. In this paper, the reservoir characteristics, percolation characteristics and productivity characteristics of 51 low resistivity reservoirs produced in AG1 layer of 4 main oilfields in the sag are analyzed, and the low resistivity reservoirs are divided into three types, taking the oil recovery intensity and cumulative oil production as the core. Then support vector machine (SVM) is used to distinguish the type of low resistivity reservoir and to study its ability to predict the type of low resistivity reservoir and the indexes such as oil recovery intensity and cumulative oil production. The results show that the density, acoustic moveout, neutron porosity, resistivity of formation depth, natural gamma, The accuracy of discriminating samples by using support vector machine classifier with natural potential amplitude ratio as model variable is 75, which can be used to predict the oil recovery intensity and cumulative oil production of low resistivity reservoirs to a certain extent.
【作者单位】：中国石油勘探开发研究院;
【基金】：国家科技重大专项“复杂油气藏精细表征与剩余油分布预测”(2011ZX05009-003)。国家科技重大专项“高含水油田提高采收率新技术”(2011ZX05010)
【分类号】：TE34

【相似文献】