论文详情
顺北地区超深层断控储集体沿下古生界北东、北西走向单剪走滑断裂展布,主要储集空间为与走滑断裂相关的洞穴、构造高角度缝和沿缝溶蚀孔洞,埋藏深(大于7000m)、非均质性强。断控储集体空间位置的准确刻画、储集体内部的物性和流体性质的精确预测,是断控储集体油气藏勘探开发的重点和难点。首先利用测井解释成果,将测井曲线分类为储层相和致密相,然后根据叠后地震敏感属性分析结果提取井旁储层敏感地震属性,在井位处对敏感属性和岩相分类结果进行标定,制作标签。利用贝叶斯分类算法,建立断控储集体储层敏感属性的岩相概率分类量版,将多个地震敏感属性体转化为储层相概率体,并以概率的方式表征其空间展布特征。最后在储层相的控制下,根据岩石物理分析得到的认识,结合叠前同步反演结果预测出断控储集体储层物性和流体性质敏感数据体。研究结果表明,地震几何属性要比地震动力学和运动学属性更准确可靠地刻画断控储集体,多属性融合概率表征有助于储层的量化研究,形成了一套适用于顺北地区断控储集体精细描述的技术流程。
The ultra-deep fault-controlled reservoirs in the Shunbei area are distributed along the NE- and NW-trending single-shear strike-slip faults of the Lower Paleozoic.The main storage spaces are caves related to strike-slip faults,structural high-angle fractures,and dissolution pores along fractures,with deep burial (more than 7000 m) and strong heterogeneity.Accurately describing the spatial location and predicting the physical and fluid properties during the exploration and development of fault-controlled reservoirs is a key challenge.We first used the logging interpretation results to classify the logging curves into reservoir facies and tight facies and then extracted the sensitive seismic attributes along the wells according to the seismic sensitive attribute analysis.At the well location,we calibrated the sensitive attributes and lithofacies classification results as labels.We used the Bayesian classification algorithm to prepare the lithofacies classification probability distribution cross-plot of the sensitive attribute in the fault-controlled reservoirs,fused the multiple seismic sensitive attributes to generate the reservoir facies probability volume,and characterized its spatial distribution characteristics in a probabilistic manner.Finally,under the control of reservoir facies,based on petrophysical analysis and the results of prestack simultaneous inversion,we predicted the volumes of physical and fluid properties of fault-controlled reservoirs.In this study,we demonstrated that seismic geometric attributes can more accurately and reliably describe fault-controlled reservoirs compared with the attributes of seismic dynamics and kinematics,and multi-attribute fusion probabilistic characterization is helpful for quantitative research on reservoirs.A set of technical processes suitable for the fine description of fault-controlled reservoirs in the Shunbei area was developed.The field data results show that the predicted spatial distribution characteristics and internal heterogeneity of fault-controlled reservoirs are consistent with the geological knowledge as well as drilling and logging information in the study area.
中国石化科技部项目(P21033-2)资助。