基于因子分析的岩石可钻性指标分析

岩石可钻性级别的准确划分是提高钻进效率的有效措施之一,但因其指标极多,故这项工作至今没有圆满解决.传统的方法是通过直接测定岩石的各项物理力学性质来综合判定岩石的可钻性,但这种方法有两大缺陷.首先,各项数据之间数量级和量纲的差距极大；其次,各项数据之间存在很高的相关性,直接使用原始数据判定可钻性会带来误差.用标准差化的方法对原始数据预处理可以消除数量级和量纲的差距.利用多元统计中因子分析模型则可以将原始数据分类构成少数综合因子,再利用数学变换将综合因子转化为综合指标,最终不仅达到降低原始数据维数,简化分析计算的目的,而且综合指标之间的低相关性也避免了原始数据间高相关性带来的误差.

Analysis of Rock Drill-Ability Factor Based on Factor Analysis in Statistics

Accurate classification of the rock drill-ability is an effective way to improve the drilling efficiency.However,there are so many indexes that it is difficult to be accomplished.Conventionally,the rock drill-ability is determined by measuring directly the physical and mechanical character of rocks.But this approach still has two major flaws: a huge difference of the magnitude and the dimension in the data,and a high correlation between these data,which are easy to leave errors in the measurement.This paper...