【摘要】 伽马曲线是反映泥质含量的重要参数,相比速度资料来说更加稳定,不受所含流体的影响。针对含气层来说,利用地震信息进行伽马反演相对于常规的波阻抗反演所描述的砂体展布更为直观。在对目前常用的伽马反演方法进行总结、讨论的基础上,指出地震体属性分析法具有明确的物理意义,并有效结合了多元逐步回归和交互验证法进行属性的优选和组合,采用褶积因子消除地震信息和测井信息的频率差异,以神经网络为拟合和预测手段,加之具有严谨的理论基础,相比其它几种方法更为准确合理。最后,通过川东T气田伽马反演的实例说明该方法的应用,预测了T气田须家河组砂体展布,取得了较好的效果。更多还原

【Abstract】 The gamma curve is an important parameter indicating the shale content. Compared with the velocity data, it is very stable and won′t be affected by reservoir fluid. For the sandbody distributing description of gas-bearing layers, the gamma inversion using seismic data is more intuitional than the conventional wave impedance inversion. On the basis of summarizing and discussing the gamma inversion, we point out that an analytic method of volume attribute is very important. The method integrates multiple stepwise regression and cross verification to optimize and combine the attribute, adopts the convolution factor to eliminate frequency difference between seismic and logging data, uses the neural network as a matching and predicting tool. Because of its religious theoretic base, the method is more accurate than other ones. At last, through a case study of the gamma inversion in T gasfield of Eastern Sichuan Basin, we predict the sandbody distribution of Xujiahe Formation.更多还原