论文部分内容阅读
主分量分析法（ＰＣＡ）的数学原理是坐标转换，它将原变量变换成互不相关的变量。在实际应用中，它为综合解释法提供了一种变量组合的途径。从物理意义上来说，它可以求出各物理参数的权系数，从而判断出，在各个主分量中各个参数的贡献大小。这样，我们便可以有目的地选择各参数，利用主分量ＰＣＡ曲线进行地质解释。本文概述了主分量分析法ＰＣＡ的基本原理和自编的分析程序流程，并以新疆塔北地区为例，说明了分层、计算泥质含量及孔隙度的成果。应用实例表明，在测并解释工作中应用主分量ＰＣＡ曲线可以取得更准确、更细致的效果。具体地说，在新疆塔北地区砂泥岩薄互层组上，可以用主分量ＰＣＡ曲线进行薄互层组的地质解释。
The principle of principal component analysis (PCA) is the coordinate transformation, which transforms the original variables into irrelevant variables. In practical application, it provides a way for the combination of variables to be used in comprehensive interpretation. From a physical point of view, it can find the weight coefficient of each physical parameter, so as to judge the contribution of each parameter in each main component. In this way, we can purposefully select each parameter and use the principal component PCA curve for geologic interpretation. In this paper, the basic principles of PCA and the procedure of self-compiled analysis program are summarized. Taking Tarim region of Xinjiang as an example, the results of stratification, calculation of shale content and porosity are described. The application examples show that PCA curve can be used to obtain more accurate and detailed results in the measurement and interpretation work. Specifically, PCA curves of principal components can be used to interpret the thin interbedded layers on sand-shale thin interbedded layers in Tabei, Xinjiang.