递归拟合测井高斯曲线快速实现岩性识别

Quickly Realize Lithology Identification by Recursive Fitting Well Logging Gaussian Curve

某种岩石矿物成分、物性结构的数值在其值域统计结果往往呈高斯分布或近似高斯分布。1口井全部测井数值直方差曲线为该井穿过所有地层岩性直方差值曲线的叠加。直方差值曲线的包络是直方差在其值域数值的最优化表示。对整条测井曲线逐次分段可以得出不同深度的直方差曲线包络。1组曲线包络可以拟合出1条高斯曲线,该高斯曲线可以表示该组包络的某种特征。比较不同井深曲线包络拟合后的高斯曲线与该地区标准岩性的高斯曲线可以实现对该井深的岩性的识别。

The mineral composition and physical structure of rock are the main factors of lithology identification.The mineral composition and the physical structure are usually distributed by Gaussian or approximatly Gaussian distribution. A completely well variance curve is superimposed by all segments.The variance envelope is the best representation in range of variance.Different depth envelope curves are drewed by whole well curves.A group of envelope curves can be fitted a Gaussian curve.The Gaussian curve can represent some characteristics of the group envelope curves.It can achieve lithology identification by comparison of Gaussian curves and the standard rock lithology Gaussian curves.