论文部分内容阅读
薄层、薄互层厚度的预测是储层横向预测的一个重要环节。常规计算薄层厚度的方法是在时间域或频率域上通过提取单参数来实现的。本文则利用小坡变换，在时一频域进行最大滴分析，并提取多种特征参数；然后利用对薄层厚度敏感的地震特征参数之间的非线性关系，使用神经网络算法，建立了一套计算薄层及薄互层厚度的方法。理论模型的正反演结果表明：该算法对薄层厚度及薄互层累积厚度的预测均有较好的效果，且具有一定的抗噪声能力。我们还利用本文所述方法对TZ地区DL－92－04测线部分剖面的石炭系I油组薄互层砂岩的累积厚度进行了预测，结果令人满意。
Prediction of thin and thin interbed thickness is an important step in reservoir lateral prediction. The conventional method of calculating the thickness of a thin layer is achieved by extracting a single parameter in the time or frequency domain. In this paper, the maximum droplet analysis is carried out in the time-frequency domain by using a small slope transform, and a variety of characteristic parameters are extracted. Then, by using the nonlinear relationship between the seismic characteristic parameters sensitive to the thickness of the thin layer and the neural network algorithm, Set of methods for calculating the thickness of thin and thin interbed. The forward and inverse results of the theoretical model show that the proposed algorithm has good effect on the prediction of the thickness of the thin layer and the cumulative thickness of the thin interbed, and has certain anti-noise ability. We also use the method described in this article to predict the cumulative thickness of thin interbedded sandstones in the Carboniferous I Formation in the section DL-92-04 measured in the TZ area, and the results are satisfactory.