论文部分内容阅读
薄层、薄互层厚度预测是储层横向预测的一重要环节。常规计算薄层厚度的方法是在时间域或频率域实现，主要使用单参数计算。本文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方法。通过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，具有一定的抗噪声能力。对塔中DL92─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 part of horizontal reservoir prediction. The conventional method of calculating the thickness of a thin layer is achieved in the time or frequency domain, mainly using single parameter calculations. In this paper, we use the neural network algorithm to establish a set of methods for calculating the thickness of thin and thin interbed using the nonlinear relationship between the seismic characteristic parameters that are sensitive to the thickness of the thin layer. The results of forward and inversion of the model show that this algorithm has a good effect on the prediction of the thickness of the thin layer and the cumulative thickness of the thin interlaminar layer, and has a certain anti-noise ability. The cumulative thickness of the thin interbedded sandstone of Carboniferous Ⅰ oil group in section DL9204 measured in the tower was predicted. The result was satisfactory.