随着开发程度加剧,中海油增储上产不断突破下限,越来越走向深层、低渗、深水和稠油.渤海海域探明原油地质储量超60％为稠油,开发难度大.影响稠油油藏热采开发的关键地质因素有油藏类型、含油饱和度、隔夹层、小断层等,其中隔夹层影响最大.D油田为辫状河沉积、储集层非均质性强的厚层块状特超稠油油藏.针对D油田砾岩夹层薄、地震预测精度低的特点,采用高分辨率地质统计学反演技术,进行了测井曲线标准化、概率密度参数、变差函数研究,获得了岩性和厚度定量表征结果,识别了厚度＜5 m的砾岩夹层,平面展布符合研究区地质特征,为心滩内部夹层三维地质模型构建提供了依据.

复合点坝储集层内部非均质性分析是曲流河研究的难点,仅靠地震和测井资料难以解释清楚.关于复合点坝储层构型表征也较缺乏定量化指导标准.本研究选取了13条曲流河河段的260个复合点坝作为数据样本,进行参数分类统计,形成曲流河复合点坝地质知识库.将复合点坝分为4大类、25个亚类;将侧积体分为8大类、22个亚类.统计不同类型复合点坝和侧积体构型样式的分布概率关系.以此为基础,充分利用定量分布概率关系,达到在资料较少情况下分析复合点坝储集层平面非均质性的目的.

在现有的碎屑沉积地质体构型分级方案基础上,充分考虑自然界中河流沉积演化规律以及海上油田的资料基础与经济开发尺度等因素,遵循地质体分级原则与依据,建立了海上油田河流相复合砂体构型分级方案.从地质成因、主控因素、时空规模等方面系统阐述了河流相复合砂体13级构型单元的基本特征,并解析其与相关沉积地质体级次的关联性.与现有的储层构型分级的差异主要在于新增了"复合点坝"级次,复合点坝是多期残存点坝以复合体形式叠置而成的沉积单元,该级次是储层构型理论、海上油田资料分辨能力与经济开发尺度三者的契合点,是海上油田"地震导向、井震联合"构型研究思路的良好实践.河流相复合砂体构型分级对于指导海上油气开发具有一定的优势.

复合点坝侧积体定量成因分析是河流沉积学研究的难点之一.通过选取密西西比河下游段激光雷达数据,剖析复合点坝侧积体的内部特征,统计侧积体高程数据,发现其符合周期变化的特点.对高程数据进行傅里叶变换计算,得到复合点坝侧积周期,将侧积周期与河道规模进行拟合,两者符合线性关系,拟合度高.深入分析后,认为侧积周期受河流的季节性洪泛周期控制.研究结论为侧积体定量成因分析提供了理论依据.

海上油田高含水期剩余油分布复杂,优势渗流通道砂体与隔夹层分布特征决定了剩余油分布,是剩余油主控地质因素.如何准确刻画与表征储层内部的优势渗流通道砂体及隔夹层,明确剩余油分布特征,认清油田调整挖潜的剩余潜力,是油藏描述面临的主要问题.针对南海珠江口盆地西江油田(XJ油田)高含水期面临的优势渗流通道砂体认识不清以及隔夹层分布难以刻画的地质问题,探讨形成海相砂岩油田高含水期精细油藏描述的关键技术.基于"差异放大"理念的精细地层细分与对比技术,解决了精细地层格架建立困难的问题;"波形与属性结合"的优势砂体刻画技术明确了储层砂体的分布范围与空间叠置关系,为识别优势渗流通道砂体与认识渗流特征提供依据;"非均质性分级表征"的三维地质建模技术将储层内部控制剩余油分布的优势渗流通道砂体、隔夹层、储层渗流差异表征在地质模型中.在精细地质模型及数值模拟的基础上定量表征剩余油空间分布特征,为剩余油挖潜和调整方案的实施提供地质依据.

石臼坨凸起F油田沙一二段发育巨厚、低渗混积岩储层,岩性多样、孔隙结构复杂、纵向非均质性强,影响酸化层段选择.综合岩心、薄片、压汞曲线等资料,结合测井信息,对储层开展分类评价,分析储层可酸化改造的潜力并落实有利的酸化层段.结果 显示,沙一二段储层包括鲕粒白云岩、生物白云岩、白云质砂岩、凝灰质砂岩、砂岩和灰质砂岩等6类岩性;根据岩石类型、物性及孔隙结构参数将储层整体划分为I、II、III类,结合实际酸化作业效果,I类储层可改造性最强,II类次之,III类最差.

通过浅水三角洲现代沉积考察分析,结合探地雷达(GPR)资料,对浅水三角洲分类方案和对应的沉积模式进行探讨.结果 表明:根据分流河道弯曲指数和分叉参数,将浅水三角洲分为分叉平直河三角洲、分叉曲流河三角洲、分汇曲流河三角洲和分汇辫状河三角洲4种类型,并建立了各类型的沉积模式;浅水三角洲沉积模式的控制因素是分流河道下蚀作用和侧蚀作用的强度.将该沉积模式应用于渤海南部海域BZ25-1油田新近系明化镇组下段砂体内部构型的研究认为:明化镇组下段4油层组2小层(NmⅣ2)时期是低位体系域早期,发育分汇曲流河三角洲,河道间为侧积隔夹层;明化镇组下段4油层组1小层(NmⅣ1)时期湖平面上升,发育分汇辫状河三角洲,河道间为垂积隔夹层.这为下一步的开发方案设计提供了指导.

根据海上油气资源勘探开发特点,在SPE油气资源分类理念下建立了具有中国海洋石油特色的《海上油气资源储量和潜在资源量分类》体系.分类体系将已发现油气资源划分为储量和潜在资源量2个类别,作为油气资源的分类基础.采用2个维度体现分类的主控要素,垂向上为油气田商业性状态和勘探开发阶段,横向上为油气资源的动用情况和开发状态.油气田划分为含油气构造、正在评价含油气构造、认定商业性油气田、在建设油气田和在生产油气田5种状态.油气资源的动用情况划分为已动用和未动用2种情况:已动用包括在生产中动用、方案中动用和计划中动用;未动用分为方案未动用和计划未动用.储量划分为已开发和未开发2种状态,潜在资源量划分为待开发和暂不开发2种状态,难动用储量划分为原油难动用储量和天然气难动用储量.

井间地震具有主频高、频带宽的特点,可清晰反映井间地层、构造、储层特征.波阻抗反演将地震反射数据转化为波阻抗信息,可直接用于储层特征分析,但井间地震低频信息缺失,将影响反演精度.首先,通过地质模型正演数据反演的方法,分析了低频信息对波阻抗反演的影响;其次,以工区井间地震为研究对象,对比分析了不同方法构建的低频模型对反演精度的影响,并将波阻抗反演数据用于井间砂体对比分析.结果 表明,(1)正演分析中,含更多低频成分的子波旁瓣的幅度比含低频成分较少的子波旁瓣要小,用其约束反演,可提高波阻抗反演精度;(2)实例分析中,采用低频信息更加丰富的低频模型约束井间地震波阻抗反演时,有助于提高波阻抗反演精度,更加准确地反映地下储层特征;(3)井间地震波阻抗反演数据反映了储层空间展布特征,有助于提高薄储层表征精度,为开发后期的井位部署和井位调整提供依据.

The architecture theory and characterization method of fluvial reservoir, which were based on dense wells data of onshore oilfields, have become relatively mature after decades of research. Nevertheless, it is challenging to apply this theory and method to offshore oilfields, where only sparse wells drilled. Through years of exploration, our research team proposes a new architecture theory and character-ization method for fluvial reservoir of offshore oilfield by combining well data with seismic data and taking compound sand-body as major research target. Compared with previous study, this new approach employs technologies such as 3 D prototype model reconstruction, seismic architecture facies prediction and seismic driving deterministic geological reservoir modeling to detailedly study the 5 th, 6 thand 7 tharchitecture units of fluvial reservoir (compound channel belt, single channel belt and compound point bar) , which shows obvious advantages on substrata division and correlation and discontinuous permeation barriers prediction in development phase. At present, there are still challenges applying this theory and method to fine geological modeling and numerical simulation. It's important to make full use of seismic data and develop artificial intelligence ways to improve seismic interpretation.

西江W油田珠江组储层为一套海相辫状河三角洲前缘沉积砂体,内部结构复杂,微相类型多样.运用高分辨率层序地层学原理及测井频谱旋回分析技术,综合岩心、测井等资料,在西江W油田珠江组内部识别出5个长期和18个中期基准面旋回,搭建了精细层序地层格架.通过几何参数的统计和砂体发育规律的总结,在精细层序格架约束下,把辫状河三角洲砂体构型样式划分为孤立型、侧叠型、双向迁移型和堆叠型4种.研究表明基准面旋回变化控制了储层构型叠置样式和演化规律,不同的中期基准面旋回位置发育不同的构型样式,在中期基准面旋回上升早期或下降晚期,砂体以堆叠和侧叠为主,而中期基准面旋回上升晚期或下降早期则以双向迁移型和孤立型为主.构型分析成果有效指导了H11层剩余油的挖潜,并为海上油田三角洲储层构型研究提供了一定思路.

Considering large well spacing in offshore oilfields development,traditional fine reservoir research thoughts from onshore oilfield,which focuses on well-seismic combination and well-based study,is not applicable on offshore oilfields.Therefore,the concept of fluvial compound sandbody is proposed.In this concept,compound sandbody is defined as sandbodies composed of several spatially genetically associated subunits within certain geologic period.Compound sandbody is hierarchical inside,where each level is composed of lower level sandbody and its interbeds,and compound sandbody unit boundaries of different levels are constrained by interfaces.Based on this concept,this paper explored the fundamental methods and approaches of characterizing compound sandbody,put forward the characterization principle of "axis-variation delimitation,pattern-fixing and well-point fixing",and determined three-step method for characterizing complex superimposed channels based on seismic sensitive attributes.Taking Bohai Q oilfield as an example,the theory is applied on residual oil prediction and injection-production connectivity analysis,which achieved better effects on tapping the potential oil.It demonstrates that this compound sandbody theory can somehow solve development challenges in oilfield production and provide support for the efficient development of offshore oilfield.The results in this paper can provide references for reservoir subdivision and comparison,development layer division,development well location deployment and well types selection.

Considering the great heterogeneity in fluvial reservoir of offshore oilfield,reservoir architecture characterization is the core for reservoir fine description.Based on previous study and under the research concept of historical comparison,this paper conducted GPR detection on modern sediments of meandering rivers,namely Hailar River and Chaobai River.This paper,through prototype model reconstruction,reviewed and explored fluvial reservoir architectural units features,and concluded the following understandings:Firstly,due to the transformation of deposition dynamic mechanism and dominating factors,fluvial reservoir formation takes the 6th order interface as breaking point.As a whole,from lower order toward higher order,the sedimentation process of fluvial reservoir architecture tends to transfer from lateral to vertical,and the 6th and 7th orders are key interfaces.Secondly,under the influence of ancient valley landform,fluvial deposits present the characteristics of multi-stacked complex longitudinally,which is an indication of multi-stage valley terraces.Relatively continuous and stable sequence interfaces may be developed among different stages.Thirdly,individual point bars were mostly reserved in the form of remnant bodies.Generally,point bar complex,which is composed of several genetically related individual point bar remnants and takes abandoned channels as lateral boundary,may have favorable conditions to form isolated sedimentary units.Therefore,point bar complex is the most important unit for meandering river sand body architecture characterization.

构型研究是目前储层研究的重点和难点领域之一.分阶段论述了这一学科的形成背景和实际需求,总结了储层构型的研究历程和发展趋势.储层构型研究手段已由常规的野外露头、现代沉积分析向3D地震、探地雷达、水槽实验、数值模拟等新技术新方法转变,层次约束、模式拟合、多维互动和地震沉积学等方法进一步加强,研究对象也由传统的曲流河沉积向三角洲、深水沉积和碳酸盐岩储层转变.储层构型研究向着精细化和定量化方向发展.同时,针对当前储层构型所涉及的热点领域,如地下构型分析、构型地震响应及解释方法、复合砂体构型与井网井距关系等方面进行了概述与讨论.

渤海A油田目的层为曲流河沉积,河道砂体厚度薄,砂体厚度精确预测是油田开发方案设计的关键.首先介绍了各种谱分解方法的特点,然后介绍了薄层识别原理,最后应用最大熵谱分解方法对渤海新近系A油田明化镇组下段储层厚度进行了定量预测.有效预测了厚度小于10 m的储层,与已钻井有很好的吻合度.最大熵谱分解方法具有分辨率高、效率高的特点,适合油田勘探及开发初期少井储层预测.

根据物源供给与可容纳空间比值,将研究区新近系曲流河相沉积复合砂体叠合模式分为堆叠型、侧叠型以及孤立型,其中堆叠型不甚发育.根据该划分模式,通过单井井震标定及联井小层对比各叠合模式的地震响应特征.通过提取均方根振幅、振幅变化率、平均振幅、有效带宽、弧长、峰值频率、能量半时间等7种属性,将模型试验确定的4类砂岩储层叠置模式,作为地震属性聚类分析的4类预测目标结果,得到NmⅡ-3小层均方根振幅与聚类分析平面图,结合测井资料验证,较准确地预测了砂岩储层叠合分布特征.

对于一个已经建立好的三维地质模型,单纯通过地震、井数据对模型进行修正精度不高.地震驱动建模作为一种充分整合地震数据的建模方法,可以充分挖掘地震中潜在的信息,通过地震驱动把地震残差转化为模型误差的修正量,从而实现模型的更新修正.基于此,本文提出地震正演循环迭代模型修正技术,并以实际工区为例阐述了该方法的技术流程.通过地震对比和井对比技术对模型修正结果进行了定性和定量评价,结果表明该方法能够降低模型修正后的累积误差,使模型与真实值更接近,更好的表征了储层,更准确的表征了储层非均质.

In recent years,the theory of composite sandbody architecture proposed by CNOOC has contributed to several major oil and gas discoveries.This paper systematically introduces the context for proposing this theory,its fundamental concept,main content and characterization method as well as practical application,and discusses the significances of this theory.Ground penetrating radar reveals that the common architecture element of modern and ancient meandering river deposits is point bar composite instead of a single point bar.When their scales being larger than seismic resolution,they commonly correspond to a single channel belt or composite channel belts and represent a sub -layer or sand group.A sand body of this kind is defined as a composite sand body and is treated as a separate development unit.The composite sand body is originated from abrupt change in accommodation and source supply caused by tectonic uplift or subsidence of ba -sin valley.The relationship between depositional process and composite sand body architecture pattern is revealed by river terrace geomorphology.This paper illustrates architecture unit,architecture bounding surfaces,architecture model,scale characteristics and their controlling factors,and defines modern sedimentary geomorphic markers of architecture bounding surfaces.Fine characterization of composite sand body architecture is the key to precisely describing remaining oil distri -bution and quantifying development index,and further optimizing injection-production pattern and increasing success rate of infilling drilling.

According to the fine reservoir subdividing and correlating method by "the markers,cycles comparison,hierarchy control,seismic constraint",the oil-bearing reservoirs can be gradually subdivided into following four levels:oil-bearing series,reservoir group,sandstone group and single oil layer.Because of the lack of the reliable subdivision and correlating marker for the fluvial facies sedimentary reservoirs,it leads to the uncertainty of the subdivision and correlation.According to the classifying system of the composite sandbody architecture,the fluvial sandbody architecture can be subdivided into four general categories and seven subclasses,and moreover this kind of the method is proven to be more accurate and effective.

Clastic reservoir architecture,which was applied to fluvial facies reservoir characterization by Miall in 1985,refers to geometric shape,scale,orientation and superposed relationship of reservoir units of different orders.Clastic reservoir architecture is an intuitive quantitative measure of reservoir heterogeneity characterization,influencing the effect of reservoir development directly.Thus,how to undertake fine architecture characterization is the key to whether reasonable quantitative development indexes or proper description of remaining oil distribution can be described so as to effectively reduce the ratio of low-efficiency wells.With an aim of overcoming the natural restriction that the thickness of fluvial facies deposition reservoirs of the Minghuazhen Formaion in Bohai bay is less than minimum requirement of seismic resolution,this study carried out the classification for the various reservoir architectures constrained by seismic data in a distinguishable scale for seismic data.A classification of three architectures and seven types has been established.Based on seismic response characteristics of each architecture and description of relevant seismic attributes,"seismic architecture facies",which combines architecture types with their seismic responses,constitutes a characterization method of "seismic architecture facies" based high-resolution conceptual model.Our study shows that although various reservoir architectures cannot be well gridding on the basis of present geological and numerical modeling technologies,the equivalent conceptual model may be an applicable technological approach.