基于聚类分析的地震属性优化及储层预测——以敖包塔油田敖9工区为例  

Optimization of Seismic Attributes and Reservoir Prediction Based on Cluster Analysis——By Taking Ao 9 Working Area in Aobaota Oilfield for Example

作　　者：王新桐[1] 卢双舫[1] 肖佃师[1] 

机构地区：[1]东北石油大学地球科学学院油气藏形成机理与资源评价黑龙江省重点实验室,黑龙江大庆163318

出　　处：《石油天然气学报》2013年第3期61-66,165,共6页Journal of Oil and Gas Technology

基　　金：国家科技重大专项(2011ZX05004-001;2011ZX05007-001)

摘　　要：应用常规地震属性分析技术进行储层预测已受到普遍重视并得到广泛应用,然而由于地震属性与所预测对象之间关系复杂,应用单一地震属性预测储层精度不高,且地震属性种类繁多不能同时参与预测,而地震属性优化技术恰能较好地解决这个问题。为此,采用基于聚类分析的地震属性优化方法,通过计算属性间的相关系数,确定相关程度,优选属性组合进行储层预测。在松辽盆地大庆长垣南部敖包塔油田葡萄花油层储层预测中,单一地震属性预测储层砂体砂岩厚度和有效厚度的相关系数分别为0.7317和0.6734,而采用聚类分析的地震属性优化方法优选属性组合后预测储层砂体相关系数可达到0.8515和0.7704,预测精度明显提高。The application of conventional seismic attribute analysis technology for reservoir prediction drew great attention and it was widely applied,however because the relationship between seismic attributes and the predicted objects was complex,the accuracy of single seismic attribute to predict reservoirs was not high enough,and a wide variety of seismic attributes could not be involved in reservoir prediction,while seismic attribute optimization technology could be better solve the problem.Therefore the attribite optimization method based on cluster analysis was used to determine the degree of relevance by computing the correlation coefficient between the attribute,the combination of attributes was optimized for reservoir prediction.In the reservoir prediction of Putaohua reservoir of Aobaota Oilfield in the south Changyuan Area of Daqing Oilfield of Songliao Basin,using a single seismic attribute to predict reservoir sand bodies of sandstone thickness and effective thickness,the related coefficient was 0.732 and 0.693 respectively,for cluster analysis,the predicted coefficient of the reservoir sand bodies was 0.8515 and 0.7704 respectively,the prediction accuracy is obviously improved.

关 键 词：地震属性 储层预测 聚类分析 属性优化 

分 类 号：P631.44[天文地球—地质矿产勘探]