应用Bayes逐步判别分析自动确定岩性

AUTOMATIC DETERMINATION OF LITHOLOGY BY USING OF BAYES STEP-BY-STEP DISCRIM INATIVE ANALYSIS

石油大学(华东)

为了应用计算机自动确定整个钻井剖面地层的岩性,本文根据Bayes逐步判别分析的原理,利用有系统取芯的关键井的测井资料,建立各类测井相的判别模型和测井相-岩性数据库。然后,用所建立的判别模型和数据库确定关键井及其它未取芯井地层的岩性,得出这些井连续而完整的岩性剖面图。我们应用此法及相应的测井相分析软件处理了几口井的实际测井资料,结果与岩芯、岩屑录井等地质资料吻合得相当好。证明这种方法是成功的,所开发的测井相分析软件是很有效的。

In order to automatically determine the lithology of the geological profile of the well with computer, models for discriminating the electrofacies and the databases of the electro-facies-lithology were established on the basis of the Bayes step-by-step discriminative principle by using the log data of the key - well with systematic core-sampling.With such discriminative models and databases, lithologies of the key-well and other wells without core-sampling were determined and the integrated well-to-well lithology section will thus be obtained. Real log data from several wells were processed with the technique recommended. Results show that they are in coincidence with the cores and the cuttings very well.