应用自组织神经网络进行自动地层划分的设计与实现

THE APLICATION OF SELF ORGANIZING NERUAL NETWORK IN AUTOMATIC STRATUM SEGMENTATION

本文针对多井对比中的地层划分问题,提出了利用自组织神经网络进行地层划分的解决办法。该方法利用未分层井段的测井信号,结合自组织神经网络的自适应算法对测井信号进行反复学习,最终得到样本空间的分类结果。该方法不仅可以针对每一口井的测井信号进行处理,而且可以将某一口井的学习结果进行保留,用于其它井测井信号进行分类的分类器。

In this paper, we put forward a method to solve the problem of stratum segmentation in multiwell correlation by self-organizing neural network. By using adaptive self-organizing neural network, it gets the labeled result from the original wellog signal. This method not only can process one well every time, but also can save the learning result to seve as classifier of other welllog signals.