Lithology identification is an important part of reservoir evaluation and reservoir description when processing and interpreting geophysical record data. Clustering analysis refers to the analysis process of grouping a collection of physical or abstract objects into several classes composed of similar objects. K-means clustering algorithm is an iterative clustering analysis algorithm. In this paper, seven mechanical property parameters of 49 rock samples are selected as experimental data in an engineering survey, and the geophysical logging method of K-means dynamic clustering analysis is adopted. The rock samples are divided into three categories, and the classification results are matched by mechanical property parameter method. By changing the order of data grouping, the misjudgment rates were 0.021, 0.021 and 0.102, respectively. Therefore, it is feasible and effective to use k-means dynamic clustering analysis to classify lithology. The number of samples decreased to 15, and the misjudgment rate increased to 0.267 The results of K-means dynamic clustering analysis may be different from the actual situation of rock sample data selection.

在处理和解释地球物理记录数据时，岩性识别是储层评估和储层描述的重要组成部分。聚类分析是指将物理或抽象对象的集合分组为由相似对象组成的若干类的分析过程。K-均值聚类算法是一种迭代聚类分析算法。本文选择了49个岩石样品的7个力学性能参数作为工程勘察的实验数据，并采用K均值动态聚类分析的地球物理测井方法。将岩石样品分为三类，并通过力学性能参数法对分类结果进行匹配。通过更改数据分组的顺序，误判率分别为0.021、0.021和0.102。所以，利用k均值动态聚类分析对岩性进行分类是可行和有效的。样本数量减少到15，误判率增加到0.267。K-均值动态聚类分析的结果可能与岩石样本数据选择的实际情况有所不同。