摘要
油藏分类方法研究是对油藏进行科学管理、提高油田开发效果的一项重要工作。针对油藏特征的属性参数,利用聚类分析方法,根据油藏特征相似关系的远近进行分类,克服了主观因素的影响,并对各类别的属性值建立判别函数,引用多组判别法来计算未知油藏归属于标准类型的概率大小,判断其属于标准类型油藏的相似几率程度,由此便可以准确地判断出未知油藏类型。结果表明,将聚类分析和贝叶斯准则下的判别分析相结合可以很好地对油藏进行科学分类和管理。
The reservoir classification study is important work for managing scientifically the reservoir and raising the effects of oilfield development.In viewof the reservoir characteristic parameters,the reservoirs are classified by cluster analysis method on the basis of the similar relationship of the reservoir features,which overcomes the influence of human factors.According to the attribute value of each category,the discriminant functions are established.The multibank discriminant methods are applied to calculate the probability of the standard type for unknown reservoir,and then the similar degree with the standard one is decided.Thus the type of the unknown reservoir is determined accurately.The result shows that the use of these two kinds of mathematical methods can classify and manage the reservoir scientifically.
出处
《岩性油气藏》
CSCD
2010年第2期123-127,共5页
Lithologic Reservoirs
基金
国家973项目(编号:2009CB219604)"高丰度煤层气富集机制及提高开采效率基础研究"资助
关键词
油藏分类
渗透率
聚类聚合法
油藏评价
多组判别法
reservoir classification
permeability
clusteringpolymerization method
reservoir evaluation
multi-bank discriminant method
