渤海低渗透油藏“甜点”地震响应特征及预测

Using seismic response characteristics to predict“sweet spot”locations in a low-permeability reservoir in the Bohai Sea

渤海N油田深层沙河街组为辫状河三角洲沉积,部分油藏单元具有低孔低渗特征,确定相对高渗透“甜点”储层发育区是保障该油田注气开发的关键因素。目的层埋深超过3200 m,地震资料主频不到15 Hz,有效频带宽度在6~30 Hz,且地震资料是对“甜点”岩性和物性双重响应的结果,导致基于地震资料的“甜点”预测精度较低。为此,采用相对高渗百分含量参数划分低渗储层中相对高渗条带类型,实现对“甜点”储层的定量表征。建立低渗透油藏地震机理模型,分析相对高渗储层的地震响应规律,优选与相对高渗百分含量参数相关性好的地震属性,建立相对高渗储层的敏感地震属性集。在沉积微相约束下进行多种敏感地震属性的聚类分析,优化沉积微相在预测中的权重以提高预测精度,并利用预测结果指导沉积微相的修改,再用于新一轮预测,多次循环迭代直至预测结果变化率小于5%时终止,从而实现相对高渗储层平面分布的精确预测。相对高渗储层的平面分布预测结果与沉积规律一致,与检验井基本吻合,为油田4口开发井的井位优化提供了重要依据。

The Shahejie Formation in the Bohai oilfield is a braided river delta deposit,and some of its reservoir units exhibit low porosity and permeability.Determining the location of relatively high-permeability reservoirs(“sweet spots”)is key to ensuring efficient gas extraction in this oilfield.The burial depth of the target layer exceeds 3200 m,the dominant frequency of seismic data is less than 15 Hz,and the effective frequency range is 6~30 Hz.Seismic responses depend on the lithology and physical properties of“sweet spots.”Therefore,the accuracy of“sweet spot”prediction based on seismic data is low.Based on permeability parameters,relatively high-permeability strip types were identified in the low-permeability reservoir.The response of a seismic model was subsequently analyzed to determine the locations of seismic attributes characteristic of“sweet spots”.Under the constraint of sedimentary microfacies,a cluster analysis of various seismic attributes was conducted,and the prediction accuracy was improved by optimizing the weight of sedimentary microfacies.The sedimentary microfacies were modified using the prediction results and then applied to subsequent predictions.Multiple cycles were performed until the change rate of prediction results was less than 5%,allowing for the final prediction of the locations of relatively high-permeability reservoirs.These results were consistent with the law of superposition.Additionally,a blind well test showed that the prediction results were mostly consistent with the test wells and therefore accurate,which provided an important basis for optimizing the locations of four wells in the oilfield.