利用分频反演技术预测烃源岩

通过寻找地震属性与总有机碳含量(TOC)之间的关系对优质烃源岩进行预测的方法逐渐得到应用并取得了较好效果,但其结果往往存在多解性,而导致预测结果不稳定.为此,笔者利用分频反演方法对涠西南凹陷流沙港组烃源岩进行精细预测,克服了多属性预测存在的多解性.通过引入AVF信息,使反演结果更加可信,并且解决了薄层识别及横向不连续性问题.采用分频反演方法得到了TOC反演体,通过对TOC反演体进行沿层计算,得到流沙港组二段顶部、底部油页岩以及内部泥岩的定量分析结果:流沙港组二段顶部和底部的油页岩TOC值大于 3%,流沙港组二段内部泥岩中的TOC值普遍大于2%,且最厚的部位基本与A洼陷、B洼陷吻合,属于优质烃源岩;井点处的TOC值、厚度的计算值与实测值基本吻合.优质烃源岩定量评价结果说明分频反演方法行之有效,可以定量评价优质烃源岩.

Hydrocarbon source rock prediction with frequency-divided inversion

The relationship between seismic attributes and total organic carbon (TOC) can be used to predict high-quality hydrocarbon source rock, and usually its application can achieve good results. However multiple solutions of this relationship lead also wrong prediction. For this reason, frequency-divided inversion is used to predict hydrocarbon source rocks in Liushagang Formation, Weixinan Sag. This method overcomes multiple solutions caused by multi-attribute prediction. Because of AVF introduction, frequency-divided inversion can solve problems such as thin bed identification and lateral discontinuity, so it is more intuitive and credible than other methods. TOC inversion cube is obtained with frequency-divided inversion, then through calculation along horizon, oil shale at the top and the bottom of Liushagang Formation and mudstone inside Liushagang Formation is quantitatively obtained: TOC of the oil shale at the top and bottom of Liushagang Formation is more than 3%, TOC of the mudstone inside Liushagang Formation is generally more than 2%, and the most thick parts coincide with the Depressions A and B. The predicted TOC and thickness value are coincident with measured value at well location. The results show that frequency-divided inversion is effective in high-quality source rock quantitative evaluation.