利用多元统计方法研究干酪根类型参数的相关性及分类问题

A STUDY OF RELATIONSHIP BETWEEN KEROGEN PARAMETERS AND THE VALIDITY IN THE CLASSIFICATION OF KEROGEN BY MULTIVARIATE STATISTICAL ANALYSIS

本文利用R—Q型因子分析、聚类分析等多元统计方法,对干酪根参数进行了研究。结果表明:因子分析对多参数的干酪根分类确为一有效方法;干酪根的H/C、O/C和1460cm^(-1)/1600cm^(-1)以及岩样的I_H等参数,较岩样的S_2/S_3、干酪根的δ^(13)C和I_o在分类中更为有效,特别是对区分Ⅰ、Ⅱ型干酪根效果很好。干酪根δ^(13)C虽在分类中有效性较差,但对区分Ⅲ型干酪根有一定意义。文中还对参数间的相关性进行了分析讨论,据此,筛选出了有效、经济的干酪根分类参数。

This paper deals with the study of kerogen parameters by means of R-Q mode factor analysis and cluster analysis, and the result shows that this is an effective method to classify kerogen by factor analysis. Kerogen parameters: H/C, O/C, rock I_H and kerogen IR 1460cm^(-1)/1600cm^(-1) are more effective than parameters: kerogen δ^(13)C, rock S_2/S_3, I_o. These indexes are useful especially in distinguishing Ⅰ, Ⅱ kerogen. Though the validity of the classification of kerogen δ^(13)C is not very good, yet it is of some significance in distinguishing Ⅲ kerogen. The authors have also analysed and discussed relationship between these parameters and have selected some economic and valid parameters for the classification of kerogen.