动态聚类分析在储层分级中的应用

在油气田，特别是海上油气田勘探开发工作中，为了提高经济效益，通常需要对储层进行分级，以便合理地制定勘探开发方案，然而，目前普遍使用的测井解释方法都无法提供储层分级的级别，通用的聚类分析也只能进行样本分类，为了从测井资料中获取储层分级信息，作者在动态聚类分析的基础上，将聚类中心的特征值之和定义为分级指数，从而成功地解决了储层分级问题，通过南海某油田ST36井的测井资料处理，并经多种资料验证，证明了用该方法进行储层分级是有效的。

APPLICATIONS OF THE DYNAMIC CLUSTERING ANALYSIS IN THE RESERVOIR GRADING

In order to increase the economic benefit, the reservoir grading is usually made in the exploration and development of oil and gas fields, especially in the sea, so that the plan for exploration and development can be drawn up reasonably. However, the current logging interpreting methods cannot present the data of reservoir grading as the common clustering analysis can only classify samples. In the paper, a grading factor is defined from the sum of eigenvalues of the clustering kernel on the basis of dynamic clustering analysis in order to abstract the information of reservoir grading and thus, the problem of reservoir grading is solved successfully. The method is verified by the examples in the logging data processing of well ST36 from the oilfield of the South China Sea.