摘要
判别分析是由已知不同类型的总体建立判别函数,再用判别函数判断未知个体所属类型的多元统计方法。该方法在地质学、生物学、农学、医学等领域有着广泛的应用。以岩性系数、孔隙度、侵入系数、含油饱和度四个观测变量,用判别分析法对油层、油水同层和水层构建判别函数,并对已知样本进行了重新归类。结果证明,所建立的数学模型效果良好,可以用于试油层的判别。该数学模型与传统的方法相比,将数学函数用于油气勘探领域是一种新的有益探索。
Discriminant analysis is a kind of multivariate statistical method which use the different type whole given to establish discriminant function,then to distinguish which kind of type the unknown individual belong to based on the discriminant function.It is used widely in geology,biology,agriculture,physic fields and so on.By using lithological coefficient,porosity,invasion coefficient and oil saturation variables,the authors constructed the discriminant functions for oil layer,oil-water layer and water layer,then a new classification to known stylebook was done,and the results prove this math model work well,and the method can be used in oil layer discrimination.Using this new math model in the oil-gas exploration field could be a new and useful solution to oil and gas exploration compared to the traditional method.
出处
《物探化探计算技术》
CAS
CSCD
2009年第4期400-404,292,共5页
Computing Techniques For Geophysical and Geochemical Exploration
关键词
判别分析
费希尔准则
判别函数
油
气
水层识别
discriminant analysis
fisher principle
discriminant function
oil-gas-water recognition
