基于BP神经网络技术的H区长3储层分类评价研究  

Research on classification and evaluation of Chang 3 reservoir in H area based on BP neural network technology

作　　者：张晓博 程国栋 赵军龙 代文鑫[1] 陶英 ZHANG XiaoBo;CHENG GuoDong;ZHAO JunLong;DAI WenXin;TAO Ying(No.2 Production Plant,Changqing Oilfield Company,Qingyang 745100,China;School of Earth Sciences and Engineering,Xi'an Shiyou University,Xi'an 710065,China;Shaanxi Key Laboratory of Petroleum Accumulation Geology,Xi'an Shiyou University,Xi'an 710065,China)

机构地区：[1]长庆油田分公司第二采油厂,庆阳745100 [2]西安石油大学地球科学与工程学院,西安710065 [3]西安石油大学陕西省油气成藏地质学重点实验室,西安710065

出　　处：《地球物理学进展》2023年第3期1272-1281,共10页Progress in Geophysics

基　　金：陕西省自然科学基础研究计划一般项目“基于生烃期古构造的致密油有利区识别方法—以鄂尔多斯盆地为例”(2019JM-359)资助。

摘　　要：为服务于陇东油田H区长3储层开发方案调整设计,本文在相关文献调研基础上,梳理储层评价分类主要方法及应用,结合收集到的地质资料、岩心资料、测井资料和生产动态资料等,开展长3储层特征分析,参考鄂尔多斯盆地低孔低渗储层分类评价认识,提出研究区长3储层分类评价标准,借助BP神经网络技术,开展长3储层分类评价.研究表明,长3储层从下至上可细分为长33、长32、长31等三个亚层、9个小层,长3为典型的低孔特低渗储层,平均孔隙度为10%,渗透率为0.3×10^(-3)μm^(2);长3储层主要为岩屑质长石砂岩、长石砂岩及长石质岩屑砂岩,长石溶孔是本区长3最主要的储集空间类型;根据毛管压力曲线及动态生产资料,结合前人认识将储层类型划分为三种类型;优选声波时差、深感应电阻率等储层类型敏感测井响应,结合储层砂地比、孔隙度和渗透率等参数,利用BP神经网络技术开展储层分类评价,精细刻画了研究区长3储层类型平面展布特征.研究建立的储层分类评价技术具有一定的参考意义,揭示的长3储层类型分布特征对开发方案调整设计具有重要支撑作用.In order to serve the adjustment and design of the development plan for Chang 3 reservoir in Zone H of Longdong Oilfield,based on relevant literature research,this article summarizes the main methods and applications of reservoir evaluation and classification,and combines collected geological data,core data,logging data,and production performance data,and carries out the characteristics analysis of Chang 3 reservoir.Referring to the understanding of low porosity and low permeability reservoir classification and evaluation in the Ordos Basin,a classification and evaluation standard for Chang 3 reservoir in the research area is proposed,Using BP neural network technology,carry out classification and evaluation of Chang 3 reservoir.Research shows that the Chang 3 reservoir can be subdivided into three sub layers and nine sub layers from bottom to top,including Chang 33,Chang 32,and Chang 31.Chang 3 is a typical low porosity and ultra-low permeability reservoir with an average porosity of 10%and permeability of 0.3×10^(-3)μm^(2);Chang 3 reservoir is mainly lithic arkose,arkose and feldspathic lithic sandstone,and design of development plans.Feldspathic solution pores are the main reservoir space type of Chang 3 in this area;According to capillary pressure curve and dynamic means of production,reservoir types can be divided into three types based on previous understanding;Selecting sensitive logging responses for reservoir types such as acoustic time difference and deep induction resistivity,combined with parameters such as sand to soil ratio,porosity,and permeability of the reservoir,BP neural network technology was used to carry out reservoir classification and evaluation,and the planar distribution characteristics of the Chang 3 reservoir types in the study area were finely characterized.The research on the established reservoir classification and evaluation technology has certain reference significance,and the revealed distribution characteristics of the Chang 3 reservoir type have an important supporting role i

关 键 词：储层分类 测井评价 BP神经网络技术 长3储层 陇东油田H区 

分 类 号：P631[天文地球—地质矿产勘探]